IDENTIFYING AND REDUCING ERRORS IN THE ...Identifying and Reducing Errors in the Operating Theatre 26 September 2005 Version 1.0 8 15.3 Characterising the Paediatric Heart Surgery

Identifying and Reducing Errors in the Operating Theatre

26 September 2005 Version 1.0 1

IDENTIFYING AND REDUCING ERRORS IN THE OPERATING THEATRE

Patient Safety Research Programme

Research Contract PS 012

FINAL REPORT Version 1.0

26 September 2005

K.R. Catchpole1

P.J. Godden2 A.E.B. Giddings3

G. Hirst4 T. Dale4

M. Utley2 S. Gallivan2 M. de Leval1

1Cardiothoracic Unit, Great Ormond Street Hospital, London, WC1N 3JH, UK and Institute of Child Health, London, WC1N 1EH, UK

2Clinical Operational Research Unit, University College London.

3King’s College Hospital NHS Trust, Denmark Hill, London, UK.

4Atrainability, Cranleigh, Surrey, UK.

Competing Interests: None

© Queen’s Printer and Controller of HMSO 2005.

This report may be freely reproduced for the purposes of private research and study and extracts may be included in professional journals provided that suitable

acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to PSRP.

CLINICAL OPERATIONAL RESEARCH UNIT



ABBREVIATIONS ACT Activated Clotting Time

ASD Atrial Septal Defect

ASO Arterial Switch Operation

CAIR Checklist for Assessing Institutional Resilience

CPB Cardiopulmonary Bypass

CVP Central Venous Pressure

DHCA Deep Hypothermic Circulatory Arrest

ETCO2 End Tidal Carbon Dioxide

FMEA Failure Modes and Effects Analysis

HCT Haematocrit

ICU Intensive Care Unit

IRAS Incident Reporting Attitude Survey

LAP Left Atrial Pressure

MRSA Methycilin Resistant Staphylococcus Aureus

MUF Modified Ultrafiltration

OTTMAQ Operating Theatre Team Management Attitude Questionnaire

RAP Right Atrial Pressure

SA Situation Awareness

SOP Standard Operating Procedure

SVC Superior Vena Cava

TGA Transposition of the Great Arteries

THR Total Hip Replacement Operation

TKR Total Knee Replacement Operation

VSD Ventricular Septal Defect A Glossary of Terms follows the Reference section The word count for the main text of this document is 47,568.



EXECUTIVE SUMMARY Introduction

A systems approach to error in healthcare is vital if adverse events are to be reduced

and surgical quality improved in the future. This project utilised three methods to

examine the aetiology of error in orthopaedic and paediatric cardiac surgery:

• Observation of errors in the operating theatre;

• Assessment of institutional resilience and attitudes to safety;

• Mathematical modelling of the impact of latent system conditions on patient

safety.

Identifying errors in the operating theatre

The prospective identification of unstable or error-inducing configurations through direct

observation can be used to examine the aetiology of error and chains of failure, identify

system deficiencies before they have an impact on outcome, and provide solutions that

are resistant to hindsight bias.

For the assessment of errors in surgery three methods were adopted; direct observation

by one or more expert observers; the use of video recordings to play back events; and

the evaluation of the non-technical skills. A model for the direct observation of surgical

error was adapted from previous work in aviation and healthcare. Structure and

consistency was achieved in the analysis by distinguishing minor failures (small events

with a negative effect) from major failures (events that come close to an accident or

incident), and applying a failure source model that made assumptions about the threats

and errors that underlie the causes of failure. Non-technical skills, described as cognitive

and social abilities, were evaluated using a scale adapted from aviation.

In 24 paediatric cardiac operations, 366 minor failures and seven major failures were

observed. The minor failures were of 29 different types, with communication and co-

ordination, absences from theatre, and equipment problems being noted as most

frequent. Task threats were the most salient source of failure, and non-technical errors

were observed more frequently than technical errors. In 20 orthopaedic cases, 421 minor

failures and one major failure were observed. The most frequent types of minor failure



were distractions, equipment management failures, safety consciousness failures, and

co-ordination and communication failures. The most frequently observed threats were

cultural and organisational problems. Non-technical errors were more frequent than

technical errors. In both cases, all but one major failure was associated with an

accumulation of minor failures in higher risk operations.

Adverse events in surgery are likely to be associated with a co-incidental accumulation

of a number of these minor recurring failures.

Assessment of culture and institutional resilience

Three tools were piloted for the assessment of safety culture and institutional resilience

in health care organisations. The Checklist for Assessing Institutional Resilience (CAIR)

was used in the paediatric cardiac centre, and an Incident Reporting and Attitude Survey

(IRAS) and the Operating Theatre Team Management Questionnaire (OTTMAQ) were

used in the paediatric cardiac centre and the orthopaedic centre. Data collection

required for the CAIR was intensive, and the IRAS and OTTMAQ required up to 25

minutes each to complete. Low response rates were observed despite considerable

efforts to encourage staff to participate. This limited the conclusions that could be drawn

from the data.

IRAS and OTTMAQ data from both centres were remarkably similar, and showed that

attitudes to safety were better in comparison to other similar studies. However,

practitioners felt that information regarding incident reports would either be badly used or

not used at all, which discouraged error reporting. Since this was also reflected in the

results from the CAIR, it appeared that this was a key barrier to safety.

Mathematical modelling of the impact of latent system conditions on patient safety

Computer simulation allows the examination of particular configurations in a complex

system without the expense of monitoring changes in the real system, so may be of

considerable value in making improvements in patient safety.

The modelling framework considered the hospital process as analogous to an industrial

process. The patient journey was modelled as a series of phases of care, each

associated with a specific location, mode of care and care team. Consideration of the



movement of both patient and information along the care pathway allows the occurrence

of and recovery from errors in the surgical care system to be examined.

A systems approach to reducing errors in the operating theatre

These studies demonstrate the value of utilising knowledge from other industries for

improving patient safety, while addressing the unique demands of healthcare. The

battery of techniques employed here provided an analysis of systems safety in

orthopaedic and paediatric cardiac units in two different hospitals. Though there were

differences in safety attitudes and the range, type and severity of intraoperative failures,

by and large both surgical centres were faced with similar problems. It is our belief that

these deficiencies afflict most health service providers in the UK, and fundamental

changes will be required in healthcare if there is to be any major reduction in the

incidence of adverse events related to surgery. Based on this work, the

recommendations for the reduction of errors in operating theatre are:

• Reduce the opportunity for errors. Checklists and standard operating procedures

could increase awareness, improve the process, and promote consistency of

care. A user-centred approach to the design, procurement and maintenance of

medical equipment could reduce equipment related problems. Interruptions,

absences, and violations of safety practices should be discouraged.

• Enhance the response to errors and the ability to avoid error-inducing situations.

The value of pre-operative briefings, post-operative debriefings, and non-

technical skills training should be evaluated as a method for improving the ability

of teams to anticipate and respond to safety threats. Involving healthcare

practitioners more in the management of and response to incident data may

encourage better incident reporting, but more information about unsafe situations

is also needed.

• Sustain efforts to understand how, why and when errors occur in operating

theatres. Intraoperative error observation techniques should be used to further

enhance the body of safety-related knowledge. The ethical and cultural barriers

to effective safety research should be examined and removed where possible.

Knowledge transfer from other industries should be encouraged.

• Computer simulation methods have the potential to complement other empirical

methods in the study of systemic errors. This should be explored further.



TABLE OF CONTENTS

ABBREVIATIONS ........................................................................................................... 2

EXECUTIVE SUMMARY.................................................................................................3

TABLE OF CONTENTS ..................................................................................................5

LIST OF FIGURES........................................................................................................10

LIST OF TABLES..........................................................................................................11

GENERAL INTRODUCTION.........................................................................................13

AIMS OF THE PROJECT..............................................................................................15

PART 1: IDENTIFYING ERRORS IN THE OPERATING THEATRE .............................16

1. Introduction ...............................................................................................................16 1.1 Summary ........................................................................................................... 21

2. Task Analysis and Failure Modes and Effects Analysis .............................................22 2.1 Task Analysis .................................................................................................... 22 2.2 Failure Modes and Effects Analysis ................................................................... 27 2.3 Summary ........................................................................................................... 31

3. General Methods.......................................................................................................32 3.1 Introduction........................................................................................................ 32 3.2 Ethics and Consent Issues ................................................................................ 32 3.3 Observational Methods ...................................................................................... 35 3.4 Analysis Methods............................................................................................... 41 3.5 Summary ........................................................................................................... 47

4. Systems Failures in Paediatric Cardiac Surgery........................................................49 4.1 Introduction........................................................................................................ 49 4.2 Method .............................................................................................................. 50 4.3 Results .............................................................................................................. 50 4.4 Discussion ......................................................................................................... 57 4.5 Summary ........................................................................................................... 60

5. Systems Failures in Orthopaedic Surgery .................................................................61 5.1 Introduction........................................................................................................ 61 5.2 Data Collection .................................................................................................. 62 5.3 Results .............................................................................................................. 62 5.4 Discussion ......................................................................................................... 70 5.5 Summary ........................................................................................................... 73

6. Major Event Vignettes ...............................................................................................74 6.1 Introduction........................................................................................................ 74 6.2 Mastoid strip causes compression of right coronary artery................................. 74 6.3 Omission of surgical step................................................................................... 75 6.4 Ex-sanguination mitigated fortuitously................................................................ 76 6.5 Breakdown in team co-ordination....................................................................... 78 6.6 Aortic homograft ruptured during sternotomy. .................................................... 79 6.7 Incorrectly Labelled Donor Tissue...................................................................... 80 6.8 ACT management difficulties in high-risk case................................................... 80 6.9 Multiple uncertainty leads to teamwork and task breakdown.............................. 83 6.10 From minor failures to major failures.................................................................. 84



6.11 Summary ........................................................................................................... 87

7. Non-Technical Skills and Errors In Surgery ...............................................................88 7.1 Introduction........................................................................................................ 88 7.2 Method .............................................................................................................. 90 7.3 Results .............................................................................................................. 90 7.4 Discussion ......................................................................................................... 95 7.5 Summary ........................................................................................................... 98

8. General Findings from the Observational Studies ...................................................100 8.1 Overview of the Studies................................................................................... 100 8.2 Threat and Error Profiles in Orthopaedic and Paediatric Cardiac Surgery........ 101 8.3 Reducing Errors in Surgery.............................................................................. 103 8.4 Methodological Considerations ........................................................................ 108 8.5 Summary ......................................................................................................... 110

PART II: ASSESSMENT OF CULTURE AND INSTITUTIONAL RESILIENCE............111

9. Introduction .............................................................................................................111

10. Checklist for Assessing Institutional Resilience.....................................................112 10.1 Introduction...................................................................................................... 112 10.2 Methods........................................................................................................... 112 10.3 Results ............................................................................................................ 114 10.4 Discussion ....................................................................................................... 115 10.5 Summary ......................................................................................................... 117

11. Incident Reporting Attitude Survey ........................................................................118 11.1 Introduction...................................................................................................... 118 11.2 Method ............................................................................................................ 118 11.3 Results from Paediatric Cardiac Surgical Centre ............................................. 119 11.4 Results from the Orthopaedic Surgical Centre................................................. 124 11.5 Discussion ....................................................................................................... 129 11.6 Summary ......................................................................................................... 130

12. Operating Theatre Team Management Attitude Questionnaire..............................131 12.1 Introduction...................................................................................................... 131 12.2 Methods........................................................................................................... 132 12.3 Results from the Paediatric Cardiac Unit ......................................................... 134 12.4 Results from the Orthopaedic Centre............................................................... 140 12.5 Discussion ....................................................................................................... 148 12.6 Summary ......................................................................................................... 150

13. General Findings from the Assessment of Organisational Safety Health ...............151 13.1 Summary ......................................................................................................... 153

PART III: MATHEMATICAL MODELLING OF THE IMPACT OF LATENT SYSTEM CONDITIONS ON PATIENT SAFETY.........................................................................155

14. Introduction ...........................................................................................................155 14.1 General Introduction ........................................................................................ 155 14.2 Simulation Modelling for Analysing Complex Systems..................................... 155 14.3 Summary ......................................................................................................... 160

15. Modelling of Errors in Surgical Care Processes.....................................................161 15.1 Concepts Underpinning the Model Structure.................................................... 161 15.2 Mapping the Patient Pathway for Congenital Heart Surgery ............................ 162



15.3 Characterising the Paediatric Heart Surgery Patient........................................ 166 15.4 Modelling the occurrence of and recovery from errors ..................................... 167 15.5 Summary ......................................................................................................... 168

16. Development and Integration of Sub-Simulations ..................................................169 16.1 Introduction...................................................................................................... 169 16.2 Patient Admission and Discharge .................................................................... 169 16.3 Bed Management on the Ward ........................................................................ 172 16.4 Diagnostic Procedures..................................................................................... 172 16.5 Cross-Matching of Blood.................................................................................. 173 16.6 Integrating the Sub-Simulations ....................................................................... 175 16.7 Simulation Modelling for Reducing Errors in the Operating Theatre................. 175 16.8 Summary ......................................................................................................... 176

17. General Findings from the Mathematical Modelling...............................................177

PART IV: REDUCING ERRORS IN THE OPERATING THEATRE .............................178

18. Summary of the Research Findings ......................................................................178

19. Learning from other industries ...............................................................................179

20. Organisational Genesis of Adverse Events in Surgery. .........................................181

21. Reducing Errors in the Operating Theatre .............................................................183

CONCLUSIONS AND RECOMMENDATIONS............................................................185

ACKNOWLEDGEMENTS ...........................................................................................187

REFERENCES ...........................................................................................................188

GLOSSARY OF TERMS.............................................................................................198 General Terms......................................................................................................... 198 Paediatric Cardiac Surgery Terms ........................................................................... 198 Orthopaedic Surgery Terms..................................................................................... 200

APPENDIX 1: Tasks completed and variations from proposed plan of work................201

APPENDIX 2: Detail of task analyses..........................................................................203 Generic Task Analysis ............................................................................................. 203 Paediatric Cardiac Surgery ...................................................................................... 207 Total Hip Replacement Surgery ............................................................................... 210 Total Knee Replacement Surgery ............................................................................ 212

APPENDIX 3: Failure Modes and Effects Analyses.....................................................215 Paediatric Cardiac Surgery ...................................................................................... 215 Total Hip Replacement Surgery ............................................................................... 222 Total Knee Replacement Surgery ............................................................................ 225

APPENDIX 4: Sample data collection packs ...............................................................229 Data Collection Pack for the Arterial Switch Operation............................................. 229 Data Collection Pack for the Total Knee Replacement Operation ............................ 241

APPENDIX 5: Checklist for Assessing Institutional Resilience ....................................254

APPENDIX 6: Evidence gathered for CAIR process....................................................256 Review of Policies.................................................................................................... 256 Interviews with Staff Members ................................................................................. 256 The Trust Risk Management Team.......................................................................... 259



Reviews of Meetings and Handovers in the Cardiac Surgery Centre ....................... 260

APPENDIX 7: Commentary on CAIR results ...............................................................264

APPENDIX 8: The IRAS Questionnaire.......................................................................268

APPENDIX 9: Rewording of the IRAS questionnaire for analysis purposes ................271

APPENDIX 10: Full breakdown of responses to IRAS.................................................273 Responses to IRAS questionnaire: Peadiatric Cardiac Unit ..................................... 273 Responses to the IRAS questionnaire: Orthopeadic Surgery ................................... 276

APPENDIX 11: The Operating Team Management Attitude Questionnaire.................278

APPENDIX 12: Rewording of the OTTMAQ for analysis purposes ..............................284

APPENDIX 13: Responses to the OTTMAQ ...............................................................286 Responses to section 1 of the OTTMAQ questionnaire: Peadiatric Cardiac Surgery .................................................................................................................... 286 Responses to section 1 of the OTTMAQ questionnaire: Orthopaedic Surgery ......... 289

APPENDIX 14: Processes at each stage of the patient journey ..................................293

APPENDIX 15: Information flow at each stage of the patient journey..........................295

APPENDIX 16: Details of sub-simulations...................................................................297

APPENDIX 17: Calibration of model parameters.........................................................298

APPENDIX 18: Visual Logic code for the simulation model.........................................299



LIST OF FIGURES Figure 1.1: Model of threat, error, and failures adopted for the observational studies....19 Figure 2.1: Generic task analysis for surgical procedures .............................................23 Figure 2.2: Level 1 task analysis for paediatric cardiac surgery.....................................24 Figure 2.3: Level 1 task analysis for total knee replacement surgery.............................25 Figure 2.4: Before and after TKR: anatomy of knee joint and prosthetic knee joint........25 Figure 2.5: Level 1 task analysis for total hip replacement surgery................................26 Figure 2.6: Illustration of total hip replacement operation ..............................................26 Figure 3.1: Video equipment configuration for paediatric cardiac surgery......................38 Figure 3.2: Video equipment configuration for orthopaedic surgery...............................39 Figure 3.3: Output from the video recording. .................................................................39 Figure 3.4: Images of video equipment installed in the paediatric cardiac

operating theatre. ................................................................................... 40 Figure 4.1: Minor and major failures by operation and risk level. ...................................51 Figure 4.2: Mean rate per operation of minor failure types. ...........................................53 Figure 4.3: Sources of threat and error..........................................................................55 Figure 4.4: Threat rate, threat type, and operative risk. .................................................56 Figure 4.5: Error rate, error type, and operative risk. .....................................................56 Figure 5.1: Minor and major failures by operation and risk level. ...................................64 Figure 5.2: Minor failures in each dual-observed operation by risk level and observers.65 Figure 5.3: Bland-Altman plot for assessing measurement agreement between minor

failure observations.................................................................................................66 Figure 5.4: Mean rate per operation of minor failure types. ...........................................67 Figure 5.5: Sources of threat and error..........................................................................69 Figure 5.6: Threat rate, threat type, and operative risk. .................................................69 Figure 5.7: Error rate, error type, and operative risk. .....................................................70 Figure 7.1: Distribution of non-technical errors along NOTECHS dimensions. ..............91 Figure 7.2: Non-technical errors attributed to NOTECHS elements...............................92 Figure 7.3: Bland-Altman plot for observed non-technical errors between observers ....94 Figure 8.1: Threat and error profiles for orthopaedic and paediatric cardiac surgery ...103 Figure 14.1: Patient flow schema. ...............................................................................158 Figure 14.2: Patient contacts with successive clinical care teams during each

phase of the health care process.......................................................... 158 Figure 15.2: Diagrammatical representation of communication interfaces in the

cardiac patient pathway. ....................................................................... 164 Figure 15.3: Descriptive flowchart for the progress of patients through the cardiac

pathway................................................................................................ 165 Figure 15.4: A chart of how information flows through the different team members

and departments in the cardiac patient pathway................................... 166 Figure 16.2: Staff groups associated with each work centre within the admission

and discharge sub-simulation. .............................................................. 171 Figure 16.3: Screenshot of the Bed Management sub-simulation................................172 Figure 16.4: Screenshot of the diagnostic procedures sub-simulation.........................173 Figure 16.5: Screenshot of the cross-matching of blood sub-simulation. .....................174 Figure A1: Treatment sub-task for the arterial switch operation...................................205



LIST OF TABLES Table 1.1: Relevant extracts from the Bristol Royal Infirmary enquiry. ...........................18 Table 1.2: Summary of error-related parameters used in the observational studies ......20 Table 2.1: FMEA Scales................................................................................................27 Table 2.2: Critical failure modes for cardiac surgery......................................................28 Table 2.4: Critical failure modes for total knee replacement operations.........................29 Table 2.3: Critical failure modes for total hip replacement operations............................29 Table 3.1: Descriptions and examples of minor failure types.........................................43 Table 3.2: Sources of failure. ........................................................................................45 Table 3.3: Surgical NOTECHS dimensions and elements. ............................................46 Table 3.4: Scoring categories for surgical NOTECHS behavioural marker scale...........47 Table 4.1: Surgery to correct congenital heart defect ....................................................50 Table 5.1: Total knee replacement surgery. ..................................................................62 Table 6.1: Minor failures, in approximate chronology, associated with major failures. ...86 Table 7.1: Correlations between non-technical errors and other intraoperative

performance parameters. ....................................................................... 91 Table 7.2: Correlations between NOTECHS ranks and other intraoperative

performance parameters. ....................................................................... 93 Table 7.3: Agreement between non-technical errors and ranked NOTECH scores .......94 Table 7.4: Crosstabulation of NOTECHS scores from both observers...........................95 Table 10.1: The completed Checklist for Assessing Institutional Resilience ................115 Table 11.1: Attitudes to the value of reporting incidents (cardiac surgery) ...................121 Table 11.2: Adverse personal consequences for reporting incidents (cardiac

surgery) ................................................................................................ 121 Table 11.3: The perceived value of reporting minor incidents and near misses

(cardiac surgery). ................................................................................. 122 Table 11.4: Attitudes to the risk management team and incident investigations

(cardiac surgery) .................................................................................. 122 Table 11.5: Attitudes to the performance of the directorate or management concerning

incidents (cardiac surgery)....................................................................123 Table 11.6: Barriers to incident reporting (cardiac surgery) .........................................123 Table 11.7: Free text comments (cardiac surgery). .....................................................124 Table 11.8: Attitudes to the value of reporting incidents (orthopaedic surgery)............126 Table 11.9: Adverse personal consequences for reporting incidents (orthopaedic

surgery) ................................................................................................126 Table 11.10: The perceived value of reporting minor incidents and near misses

(orthopaedic surgery). ..........................................................................127 Table 11.11: Attitudes to the risk management team and incident investigations

(orthopaedic surgery) ........................................................................... 127 Table 11.12: Attitudes to the performance of the directorate or management

concerning incidents (orthopaedic surgery). ......................................... 128 Table 11.13: Barriers to incident reporting (orthopaedic surgery) ...............................128 Table 11.14: Free text comments (orthopaedic surgery). ............................................129 Table 12.1: The respondents own practice (cardiac surgery) ......................................136 Table 12.2: Hierarchy and leadership in teams (cardiac surgery) ................................136 Table 12.3: Expressing concern at the actions of other team members (cardiac

surgery) ................................................................................................ 137 Table 12.4: Job and workplace satisfaction (cardiac surgery) .....................................137 Table 12.5: Working as a team (cardiac surgery). .......................................................138 Table 12.6: Conflict and stress at work (cardiac surgery) ............................................138



Table 12.7: Preferred and experienced leadership styles (cardiac surgery) ................138 Table 12.8: Work goals (cardiac surgery) ....................................................................139 Table 12.9: Free text responses to the question “How can the effectiveness of

operating theatre teams be increased?” ............................................... 140 Table 12.10: Free text responses to the question “How can the job satisfaction of

operating theatre teams be increased?” ............................................... 140 Table 12.11: The respondents own practice (orthopaedic surgery) .............................143 Table 12.12: Hierarchy and leadership in teams (orthopaedic surgery) ....................... 143 Table 12.13: Expressing concern at the actions of other team members

(orthopaedic surgery) ........................................................................... 144 Table 12.14: Job and workplace satisfaction (orthopaedic surgery) ............................ 144 Table 12.15: Working as a team (orthopaedic surgery). .............................................. 145 Table 12.16: Conflict and stress at work (orthopaedic surgery) ................................... 145 Table 12.17: Preferred and experienced leadership styles (orthopaedic surgery) ....... 145 Table 12.18: Work goals (orthopaedic surgery) ........................................................... 146 Table 12.19: Free text responses to the question “How can the effectiveness of

operating theatre teams be increased?” ............................................... 147 Table 12.20: Free text responses to the question “How can the job satisfaction of

operating theatre teams be increased?” ............................................... 148 Table A10.1: Numbers of responses given to each question for the first section of

the IRAS survey ................................................................................... 274 Table A10.2: Number of responses and percentage response for each section of

the second section of the IRAS survey. ................................................ 275 Table 10.3: Numbers of responses given to each question for the first section of

the IRAS survey. .................................................................................. 277 Table A10.4: Number of responses and percentage response for each section of

the second section of the IRAS survey. ................................................ 277 Table A13.1: Numbers of responses given to each question for the first section of

the OTTMAQ survey............................................................................. 289 Table A13.2: Numbers of responses given to each question for the first section of

the OTTMAQ survey............................................................................. 292



GENERAL INTRODUCTION

“Human error is the inevitable by-product of the pursuit of success in an imperfect,

unstable, resource constrained world.” (Dekker, 2002)1

It is estimated that incidents in which unintended harm results from the receipt of

healthcare cost the UK approximately £2 billion a year in additional hospital stays alone2.

Retrospective studies suggest that medical error is among the leading causes of death in

both the USA and the UK3;4. Though some approaches5 have sought to associate, by

implication, individuals with medical failures, errors of individual skill or expertise are

unlikely to be the only cause of catastrophic events6-9, since the study of these events

usually reveals a series of minor failures that were known and tolerated, and only

created the catastrophe when occurring in close temporal or spatial proximity10-14. This

means that human errors must be seen as the consequence of inadequate components

of the healthcare system, a philosophy often described as the systems approach to

safety11;15. Recent high profile failures16 and reports2 have highlighted the need to adopt

the systems approach, since it has two central advantages. Firstly, it acknowledges that

humans are fallible at every stage of the design, maintenance, configuration, support,

and operation of any given system, and consequently that all system components – not

just the humans performing the most safety-critical tasks – can be faulty. This allows the

redesign or replacement of faulty system components or processes. Secondly, by

acknowledging that individuals can be forced into making errors, fear of blame is

reduced, which will improve error reporting, and result in more and better information on

the deficient system components. This allows the prospective identification of error-

inducing system components before they can have a catastrophic effect. The application

of this approach to the study of surgical error will undoubtedly be successful in improving

safety in medicine17-22, and it forms the fundamental basis for this research.

Surgery is the apex of the healthcare system, where every component of healthcare –

diagnostics, treatments, technologies, skills, teamwork, infrastructure, management –

has some influence on the events that occur in the short and critical time a patient is in

the operating theatre. Human factors is the study of the relationship between humans,

the tasks that they perform, the equipment or tools they use to perform those tasks, and

the environment in which they perform them. It is central to a systems approach to the



identification and reduction in error, and thus to improvements in safety, quality and

efficiency in surgery. Operational research uses mathematical models and other

analytical approaches to understand and assess the processes in an organisation, and

may therefore be of help in identifying deficiencies that arise due to configuration of

different elements within the system. In this research, human factors principles were

used to examine the aetiology of human error within the operating theatre, and to

observe and analyse behaviours that were indicative of systemic deficiencies. Cultural

attitudes to safety within the operating theatre teams were assessed using a number of

questionnaires and assessment tools, while operational research techniques were used

to examine the path of patients and information through the healthcare system. This

provided a total systems approach to the identification and reduction of errors in

operating theatres.



AIMS OF THE PROJECT

The original aims of the project, as stated in the proposal (page 2), were to:

• prospectively identify latent conditions which may influence operating theatre team

performance;

• develop an integrated video and trained observer method for identifying active errors

in the operating theatre;

• apply Operational Research techniques to examine linkages between latent

conditions and safety;

• develop error management strategies to reduce the frequency of latent conditions and

active errors.

Details regarding the tasks completed and variations from the original proposed plan of

work can be found in Appendix 1. Here, it is worth noting a number of difficulties

encountered during the course of the studies. The original first applicant was unable to

lead the project, which also suffered from a lack of support from staff, some of whom

were unwilling to participate in the study. There were also a number of difficulties with

ethics approval associated particularly with the utilisation of video in the operating

theatre, which are discussed in more detail within the document.



PART 1: IDENTIFYING ERRORS IN THE OPERATING THEATRE

1. Introduction

The system of surgery requires the successful co-ordination of complex and independent

system components. It requires a skilled team of surgeons, anaesthetists, nursing staff

and other specialists, who must all work together to provide the best treatment for the

patient. It requires a patient who has undergone an appropriate diagnostic process, a set

of well rehearsed procedures that match closely to the requirements of the patient, and a

range of equipment, drugs and blood products that must be appropriately organised in a

workspace that supports the individuals23. Furthermore, surgery requires an organisation

and culture which supports the progress of the patient through their treatment, and the

activities of the team in the operating theatre24. In such a complex system, there is a

great deal of capacity for failure7;16;25. The first aim of this study was to develop a

methodology for identifying weaknesses in the system of surgery.

Reacting to failures in safety after a poor outcome can help reduce the risk of similar

events, but is prone to hindsight bias26-28 and may not always address the fundamental

causes of the error. This can result in the perpetuation of the error-inducing system

elements embedded in brittle and ineffective safety systems6. Consequently, a systems

approach to safety benefits from prospective studies to identify existing process failures

within the system, and the removal of the sources of those failures before they can affect

outcome29;30. The prospective evaluation of human error in surgery has begun to benefit

from direct observation of error7;8;31-33, but so far has failed comprehensively to address

the systemic causes of failure. Following previous attempts to identify the system

components of patient safety18;34;35, we hoped to offer a method to evaluate aspects

system failure and thus to arrive at generalisable observations of surgical error. Of

substantive interest were the range, frequency and types of failures that could be

identified, and whether there was evidence that minor failures could accumulate to form

more serious events. The application of human factors principles to the prospective

observation of healthcare error was therefore also a primary goal for this work.

Studies began in paediatric cardiac surgery and progressed to orthopaedic surgery. It

has been argued that cardiac surgery shares many properties with other safety-critical



industries36. This type of surgery was central to perhaps the most exhaustively

investigated healthcare catastrophe of recent years16;25, and is particularly predisposed

to errors because it features multiple specialties, close coupling of concurrent tasks,

uncertainty, changing plans and high workload13;37 to which paediatric cases may be

particularly susceptible38. Finally, previous prospective direct observation analysis in this

surgical domain had offered strong evidence that small negative events can influence

outcome7. Consequently paediatric cardiac surgery was selected as a suitable surgical

domain for study. Orthopaedic surgery was selected to contrast with paediatric cardiac

surgery, because it carries low risk39, is highly proceduralised and generally invariant, so

may be suited to safety practices found in the aviation and nuclear industries which

utilise clearly defined rules. It is also dependent upon a distributed set of specialists who

perform their tasks outside the operating theatre, sometimes weeks or months before or

after the operation, but will influence, or be influenced by, the critical events in theatre.

Since humans are a fundamental component to every system, we expected many

similarities across both surgical domains that would also reflect known behaviours in

other complex systems such as aviation. However, since the teams, tasks, equipment,

specialist skills, patients and treatment centres were different, we expected a different

spectrum of problems in each surgical domain. The extent to which they differed would

offer a new perspective on the requirements for safety between and within surgical

specialties and treatment centres.

It was hoped to identify the analysis method prior to data collection. To that end, task

analyses were produced for both types of surgery that defined surgical tasks in some

detail, and a structured error capture checklist was produced based upon this work.

During the data collection period Failure Modes and Effects Analyses (FMEAs) were

performed based on each task analysis in the hope of identifying the areas of surgical

failure that were most at risk. The task analyses and FMEAs are detailed in section 2. It

quickly became clear that surgery was too complex to produce process-based error

capture tools that were sufficiently well defined to identify all individual failures. Since the

most serious or easily recognisable problems already had defences and could be

anticipated by the theatre teams, it was neither individual errors, nor isolated failures of

system components, that directly caused unsafe or undesirable events in theatre.

Rather, it was important to identify the tasks or processes that did not go smoothly, were

unexpectedly difficult, or became unsafe, due to the interaction between sub-optimal



system properties and human fallibility. The Kennedy report16 was useful in the early

stages, since it described many undesirable events that were observable in operating

theatres and were implicated in the tragic events at the Bristol Royal Infirmary. Table 1.1

summarises some of the events recorded in the Kennedy report that were also observed

during the present studies. The challenge was to develop a method to quantify the

systemic causes of human error and their relationship with adverse outcomes, and thus

arrive at a method for prospectively identifying and reducing errors in surgery.

Extracts from the Kennedy report (2001): “The avoidance of open discussion was compounded by……the uneasy relationship between anaesthetists and surgeons, which made it difficult for any anaesthetist to appear critical of a surgeon”. (p.141) “The difficulties…encountered reveal both the territorial loyalties and boundaries within the culture of medicine and of the NHS…”(p.141) “An organisation offering a service must, of course, have dedicated staff. But that is not enough. It must also have in place within it systems that allow it to learn, develop, and prosper, quite apart from any external mechanisms. A key feature of such systems is that all involved must feel able to be open about their work and the work of colleagues.” (p.156). Criticism was also made of the absence of anaesthetists during surgery. (p.193) Difficulties with equipment were also mentioned… (p.193) “Perhaps the most significant deficiency was the lack of availability of cardiological advice and assistance to the surgeons in the operating theatre.” (p.193) “The fundamental problem…..was the stark question: who was in charge? Anaesthetists and surgeons carried out separate ward rounds…….. staff felt that they received conflicting instructions. A course of action indicated by one clinician might be changed by another…..” (p.195). NB: While this was not studied on the ward, conflicting commands were apparent in theatre, particularly in relation to perfusion. “We reach one conclusion…..the problem of poor teamwork and the implications this had for performance and outcome. The crucial importance of effective teamwork in this complex area of surgery was very widely recognised….” (p.197) “[perfusion] is a shared responsibility between the perfusion technician….the surgeon doing the plumbing side….and the anaesthetist who has overall responsibility for the physiology of the rest of the body whilst the heart is being looked at and operated on by the surgeons…..the role of the anaesthetist is to help the perfusionist interpret blood gas levels….and to help him manage that to help the perfusionist control blood pressure so it is not too low and not too high, because we know that in both those situations that if there is a lot of blood coming back because perfusion is not good, the surgeon may not be able to do the operation as quickly and as efficiently as possible. So there is that aspect of making the surgeon’s job easier and also protecting the patient. It is not possible…to say really perfusion equals perfusionist; perfusion equals all three of those elements” (p.629) “….a choice on the part of the anaesthetist between those patients [in the ITU] and [in the theatre]……. Often I would have to leave my patient in theatre with a trainee anaesthetist while I went to CIC to assess patients. If I was the on call anaesthetist on a Monday I would wait until my patient in theatre was safely established on CPB before visiting ICU.” (p.631). “….On occasion the surgeons were surprised by some of the anatomy that they found once surgery had begun.” (p.632) Table 1.1: Relevant extracts from the Bristol Royal Inf irmary enquiry. All these problems identified as part of the Bristol Royal Infirmary enquiry were observed intraoperatively in the present studies.



Threats Organisation/Culture Environment Task Patient

Errors Technical Non-Technical

Minor Failure

Major Failure

Adverse Outcome

Figure 1.1: Model of threat, error, and failures adopted for the observational studies. Threats either predispose errors that cause minor failures in process, or directly cause minor failures themselves. These minor failures either lead to more threats, and so to more errors, or lead directly to major failures, which are events that may be dangerous. These major failures can lead to more threats and failures, or directly to an adverse outcome. Since human errors and their causes may not always be directly observed40, there is a

need to infer deficiencies from the undesirable events that occur. A threat and error

model was adapted from similar direct observation studies on the flight deck41;42 to

provide the assessment framework. In this model, which is illustrated in figure 1.1,

threats are elements of the system that provide the circumstances in which human errors

can occur. Threats – sub-optimal properties of the system – can directly result in minor

failures in process (e.g. equipment failure). Alternatively, failure to compensate for

threats can lead to a human error that results in a minor failure (e.g. failure to notice a

critical read-out on a complex piece of equipment). These minor failures can expose

more threats, and so lead to more errors, or can lead directly to events that may be

dangerous, described as major failures. Major failures can also lead to more threats,

errors and failures, or directly to an adverse outcome. Essentially, threats predispose



errors which may then accumulate or cascade14 to create an increasingly undesirable or

unsafe situation. By preventing the transition between threats, errors and failures, or

capturing and mitigating current failures, the chance for further failures can be reduced,

and the error chain broken.

Parameter Definition

Minor failure Small undesirable process events; events observed in theatre

that were judged to have had small negative effects on the

duration or difficulty of the operation, the risk to the patient, or

the demand for resources.

Major failure Events that come close to an incident or accident.

Threat Elements of the system that predispose error; properties of the

culture, organisation, patient, environment or task that were

associated with an intraoperative failure.

Error An undesirable deviation from intent (Reason 1990), or

alternatively, a human action or set of human actions that

exceed some limit of acceptability (Miller et al. 1987); any form

of human error associated with intraoperative minor failure.

Technical error Human errors that arise from imperfect psychomotor ability or

incomplete utilisation of experience, training or expertise.

Non-technical error Human errors that arise from imperfect general cognitive or

social skills.

Table 1.2: Summary of error-related parameters used in the observational studies

The surgical system was viewed as being composed of four fundamental types of threat:

patient threats, task threats, environmental threats (which included equipment and

resources), and culture and organisational threats35;43. In terms of human errors, it was

important to distinguish between errors of technical skill or expertise8;33;44 that are

formally taught and assessed in healthcare, and non-technical skills, such as teamwork

and general cognitive and social skills18;45;46, which are not. Non-technical skills can be

both a source of failure, and a method for the management of threats and errors, so may

offer considerable opportunity for reducing surgical error. By providing a structured

approach to the identification of system threats and human error from observations of



undesirable events in the operating theatre, it was hoped to measure the relative

incidence of different sources of failure within the system of surgery.

A summary of the definitions used in these studies is provided in table 1.2, and the

general methods adopted are described in section 3. Sections 4 and 5 describe the

results of the application of this method in paediatric cardiac surgery and orthopaedic

surgery respectively, and section 6 describes in detail the major failures observed. The

final study, described in section 7, focused on non-technical skills in relation to the

interoperative observations of failures, threats and errors.

1.1 Summary

• A systems approach to error in healthcare is vital if adverse events are to be avoided

in the future.

• The prospective identification of unstable or error-inducing configurations through

direct observation can:

o quantifiably examine the aetiology of error and chains of failure

o identify system deficiencies before they have an impact on outcome

o provide solutions that are resistant to hindsight bias

• A model for the direct observation of surgical error was adapted from previous work in

aviation and healthcare. This defines:

o Threats as elements of the system that predispose errors.

o Errors as failures of intention or human actions that exceed acceptability.

o Minor failures as small events with a negative effect on the process.

o Major failures as events that come close to an accident or incident.

o Technical skills as experience, expertise and psychomotor performance.

o Non-technical skills as cognitive or mental abilities and social or

interpersonal abilities possessed by an individual and displayed by the

team.



2. Task Analysis and Failure Modes and Effects Analysis

2.1 Task Analysis

Task analysis is a technique that can be used to reduce high level system goals to

individual actions or procedures. It can be used to help understand the similarities and

differences between different types of surgery, and to allow methodological study and a

detailed understanding of the requirements for the successful completion of a given

operation. A generic task analysis for surgical procedures was initially developed to

provide structural uniformity across the three operations documented here. Specific

analyses were then developed for each operation by studying text-book descriptions of

the procedures and surgical manuals, and through information searches, observation of

operations, and discussions with specialists. These task analyses were used for the

development of explicit procedural-based error-capture checklists for our observational

studies, described in the general methods section, and for an FMEA assessment of each

type of surgery, described later in this section.



Generic Surgical Task Analysis

Figure 2.1 shows a task analysis that is generic for most surgical procedures that require

general anaesthesia or heavy sedation. The complexity of each task and sub-task will be

dependent upon the procedure, but, in general the tasks exist in one form or another.

The sub-tasks that will vary considerably between procedures are shown in boxes with

dashed outlines. These procedure-specific variations are described in subsequent

sections. A description of each stage has been provided in Appendix 2.

Figure 2.1: Generic task analysis for surgical procedures

Transfer to Theatre

Access Treatment Closure Transfer from Theatre

Preparation Assessment

Move Patient in

Transfer Monitoring lines

Transfer Pump Lines

Transfer patient to table

Preparation Incision Exposure of Treatment Area

Anatomical Inspection

Finalise Treatment Plan

Execution of Treatment

Stabilise Patient

Assess Treatment Area

Assess Treatment Success

Stabilise Patient for Treatment Plan

Move Patient Out

Prepare for transfer

Transfer Monitoring lines

Transfer Pump Lines

Transfer patient to bed

Close Skin Close Treatment Site

Count equipment

Highly procedure-specific variations

Theatre



Paediatric Cardiac Surgery

This task analysis (figure 2.2) examined paediatric cardiac surgery to correct congenital

heart defects. Though it does not specifically describe each operation, it describes tasks

common to that surgical domain, and focuses purely on the surgical procedure. A

second level of detail in this task analysis can be found in Appendix 2.

Figure 2.2: Level 1 task analysis for paediatric cardiac surgery

Orthopaedic Surgery

Task analyses in orthopaedic surgery examined total knee replacement (TKR) surgery,

and total hip replacement surgery (THR). As these operations are relatively

straightforward and invariant, it was possible to break down each task into greater detail

than was possible for the considerably longer, more complex and unpredictable cardiac

procedures.

The TKR analysis (figures 2.3 and 2.4) is based upon the requirements for a PFC

implant, as specified in the manufacturers manual. It was adapted, through pilot

observations, to the specific safety and team procedures in place at the orthopaedic

centre. Primary implants of other types demonstrate minor procedural deviations from

this. Computer navigation-aided procedures, undergoing trials at the orthopaedic centre

Access Closure Treatment

Stabilise Patient

Assess Treatment

Close Skin Count equipment

Theatre

First Incision

Exposure of Treatment Area

Anatomical Inspection

Prepare forTreatment

Surgical Tasks

Maintenance of effective perfusion

Bloodgas checks

Temperature Control

Wean from bypass

Cross clamp protocols

Cardioplegia administration

DHCA Procedures



during the period of the data collection, and not documented here, vary in a considerable

number of respects. However, many of the steps and basic requirements remain the

same. The total hip replacement (THR) task analysis (figures 2.5 and 2.6) is based on

the Exeter total hip replacement, as defined by the surgical manual provided by the

prosthetics manufacturer, and adapted, through pilot observations, to the specific safety

and team procedures in place at the orthopaedic centre. A second level of detail in these

task analyses can be found in Appendix 2.

Figure 2.3: Level 1 task analysis for total knee replacement surgery

Figure 2.4: Before and after TKR: anatomy of knee joint and prosthetic knee joint


Stabilise Patient Close Skin Count equipment

Theatre

First Incision


Femoral Preparation

Tibial Preparation

Patella Resurfacing

Insertion of Implants



Figure 2.5: Level 1 task analysis for total hip replacement surgery Figure 2.6: Illustration of total hip replacement operation


Stabilise Patient

Assess Treatment

Close Skin Count equipment

Theatre

First Incision


Femoral preparation 1

Acetabular preparation

Acetabular cementing

Acetabular implantation

Trial reduction

Femoral preparation 2

Femoral cementing

Stem implantation

Reduction

Femoral resection

1. Remove femoral head 2. Insert acetabular component

3.Insert femoral component



2.2 Failure Modes and Effects Analysis

Failure Modes and Effects Analysis (FMEA) is a technique that attempts to predict how

critical each possible failure in the system is by taking into account frequency, severity,

and detectability, in order to identify those failures that provide the greatest risk (BS5760

part 5, 1991). It has been used in many other high risk industries, such as aviation,

nuclear and chemical industries to identify the areas of greatest risk and develop

methods for reducing those risks. An example of FMEA application in healthcare can be

found in a recent Institute for Healthcare Improvement Report47. From the task analysis,

each element was assigned one or more failure modes, from which effects were

predicted. Each failure mode was scored from 1 to 10 on three dimensions; chance of

failure, severity of failure, and chance of detection. These three scores were then

multiplied together to give a criticality index, which defines the level of risk for each

failure. The higher the score, the more serious the potential for failure, and so the more

priority should be placed on avoiding that failure mode. The scales and anchors used for

each dimension are shown in table 2.1.

Occurrence Scale Description Remote (1) No known occurrence Low (2, 3, 4) Possible, but no known data Moderate (5, 6) Infrequent High (7, 8) Frequent Very High (9, 10) Almost certain

Severity Scale Description Slight annoyance (1) May affect the system Moderate System Problem (2, 3) May affect the patient Major System Problem (4, 5) May affect the patient Minor Injury (6) Major Injury (7) Terminal Injury or Death (8, 9)

Detection Scale Description Very High (1) Error always detected High (2, 3) Error likely to be detected Moderate (4, 5, 6) Moderate likelihood of detection Low (7, 8) Low likelihood of detection Remote (9) Detection not possible at any point

Table 2.1: FMEA Scales . Criteria used for estimating chance of occurrence, severity, and chance of detection of failure mode.




The FMEA table for paediatric cardiac surgery can be found in Appendix 3. The highest

risk is failure to administer heparin, because the effect is severe, and because it can be

difficult to detect if the defences are not in place beforehand. However, the criticality

index in this case does not reflect the fact that there are several levels of defence for

avoiding this particular failure. The surgeon must first ask for the heparin, the

anaesthetist confirms that the heparin is in, and then confirms that the Activated Clotting

Time (ACT) is high enough to initiate bypass. This process has not been modelled as

part of this FMEA, though clearly a more detailed assessment might. The next most

critical event is finding that the anatomy of the patient is different from expectation.

Though this only has an impact on the process of the operation, it is both frequent and

difficult to detect until discovered by the surgeon. Rupture of arteries during the

sternotomy is also high on the criticality index, as though relatively infrequent, and

readily detectable, it carries serious implications for patient safety. Arterial and venous

cannulation also appears on the list of critical tasks. Even though they cause only a

moderate system problem, the chance of occurrence is moderate, and the ability to

detect the problem is low. Poorly placed cannulae can create problems for the surgical

team once on bypass, are difficult to correct once bypass has been achieved, and can

result in considerable down stream impact on the success of the bypass, the ease of the

operation, the ability to perfuse, and consequently the physiological well-being of the

patient. De-airing the heart following treatment is as critical, carrying a much higher

severity – as it can lead directly to cardiac failure – but a lower chance of occurrence,

and a higher chance of detection.

Task Failure Mode Effects Index 1.3.4.3 Perfusionist confirms full flow bypass heparin not given Patient death 144

1.3.2 Treatment area is examined for consistency with expectation

Found to be different More difficult operation 84

1.2.3 Sternal saw is used Rupture of arteries Extensive bleeding 72

1.3.3.2 Arterial cannulation Poor cannulation Poor perfusion 70

1.3.3.3 Venous cannulation Poor cannulation Poor perfusion 70

2.5.1 Heart is de-aired. not de-aired correctly embolism 70

Table 2.2: Crit ical fai lure modes for cardiac surgery

Orthopaedic Surgery

The FMEA tables for TKR and THR surgery can be found in Appendix 3. The most

critical TKR tasks, shown in table 2.4, all relate to effects on the success of the

prosthetic implantation. They involve failures in femoral cuts, tibial cuts, drilling or the



seating of measurement jigs. Failures in these tasks are difficult to detect and difficult to

recover from.

Task Failure

Mode Effects Index

2.1.7 Distal femoral cut: Oscillating blade saw placed through slot poor cut poor implantation 60

2.1.17 Anterior, posterior and chamfer cuts with oscillating saw poor cut poor implantation 60

2.2.3 Malleolar clamp of tibial alignment device placed badly placed poor implantation 60

2.2.8 Oscillating saw used to make tibial cut poor cut poor implantation 60

2.3.4 Resection with oscillating saw poor cut poor implantation 60

2.3.5 Template used to drill holes inaccurate drilling

poor implantation 60

2.4.8 Cement applied to tibial implant poorly applied poor implantation 60

2.4.11 Cement applied to femoral interface poorly applied poor implantation 60

2.4.16 Implant fixed to patella poorly applied poor implantation 60

Table 2.4: Crit ical fai lure modes for total knee replacement operations

Task Failure Mode Effects Index 2.5.4 Cement injected retrogradely into femoral

canal with gun. Poor cement delivery Poor seating of prosthetic 72

2.5.5 Stem is inserted into the femoral canal. Poorly completed Poor seating of prosthetic 72

2.5.6 Stem is centralised in the cement mantle. Not completed properly Poor seating of prosthetic 72

2.3.4 Trial component is used to check size and fit Not completed properly Fitting problems 70

2.3.5 Movement of the hip joint is tested. Not completed properly Poor joint function 70

2.5.2 Intramedullary canal is plugged Not plugged Cement mixes with marrow 70

Table 2.3: Crit ical fai lure modes for total hip replacement operations The most critical failure modes in THR are those that either involve the seating of the

femoral prosthesis, or the checking of the fit. The seating of the femoral component has

important implications for the overall fit of the prosthesis – so risks minor injury to the

patient – while being difficult to detect.

Limitations of FMEA

The tasks in cardiac surgery are generally of higher criticality than those in orthopaedic

surgery, and the THR operation is generally of higher criticality than the TKR operation.

This would be in line with expectation. However, in order to achieve a task model that

incorporates the entire operative process, it has not been possible to develop the FMEA

to any great level of detail.

The FMEA process was developed in order to predict the down-stream effects that

failures of individual components might have on overall system performance. In

particular, it is based on having reasonable predictions about likelihood of failure, and

assumes a linear path between components. Though this general approach has been



applied successfully to human systems, the prediction of human reliability is a far less

accurate process, and viewing the human as a system component becomes less

appropriate the as the task becomes more removed from equipment or linear

procedures. This is because humans adapt to the environment in which they act;

machines do not. Thus, with highly human- and team-oriented tasks such as surgery, the

results from an FMEA are unlikely to be accurate.

A more detailed FMEA could provide greater accuracy, but was not been possible in the

current study, due to the complexity and unpredictability of surgery. Each individual – of

whom there are usually between five and eight directly involved in the procedure – will

have a range of tasks over a period of several hours. The requirement to document

everything that every team member does, and the interaction requirements at certain

points, makes a full FMEA unachievable at present. Furthermore, unlike engineered

systems, processes in an operation can be highly variable, and this variability is itself

likely to be a particular source of failure. Unlike engineered systems, human teams are a

dynamic and often unpredictable system component, both in the failures created, and

the failures detected. In addition, it is difficult to predict the effect that certain failures will

have on a patient, and that prediction may only be attempted when sufficient data has

been gathered from observation of each failure mode30. Thus, FMEA can provide some

assessment of potential weak areas in the process, but the results must be treated with

care when examining systems heavily reliant on humans. The range of and variation in

surgical procedures, coupled with the detail needed to achieve sufficient accuracy, and

the large number of potential effects on the patient, make FMEA in surgery suitable only

for discrete, equipment-oriented tasks.

For the purposes of the present study, the criticality index provides useful insight into the

more serious individual failures, but does not aid in the prediction of serious

intraoperative failures. As a result, for the subsequent observational studies of error, we

did not rely solely upon the FMEAs and task analyses. The task analyses in particular

proved useful for the development of the intraoperative score sheets, but both methods

proved insensitive to the large majority of recurrent minor failures in surgery, and to the

smaller number of more dangerous major failures described in section 6.



The processes described are much more complex and variable than can be completely

illustrated here, and consequently there will be failures that have either not been

assessed in the most representative way, or have been entirely omitted. The flow from

one task to another, or the accumulation of failures in the system, was not considered

aside from the heparinisation procedure, which is known to be safety-critical and thus

has a number of safety levels associated with it. A consideration of the accumulation of

failures would be particularly useful as FMEA does not lend itself well to failures in the

team, or the temporal co-ordination of individuals, that can lead to serious events.

Consequently, while this assessment yields results that might generally be in line with

expectation, this FMEA cannot currently be the only method by which failures in surgery

are modelled.

2.3 Summary

Through examination of the process and tasks required in surgery, it was possible to

derive task analyses and FMEAs for paediatric cardiac surgery, total hip replacement,

and total knee replacement operations. This allowed us to conclude the following:

• Task analysis is useful for:

o describing the process of surgery

o defining the bounds of error-related observation

o developing observational error capture methods

• Task analysis is limited by the variability and complexity of interactions in surgical

procedures

• FMEA is limited by the accuracy of the task analysis, and may not be suited to

healthcare tasks that are highly dependent upon interactions between individuals.



3. General Methods

3.1 Introduction

For the assessment of errors in surgery three methods were adopted; direct observation

by one or more expert observers trained to assess surgical performance; the use of

video to record and play back events; and the evaluation of the non-technical skills

apparent during each operation. By providing a structured approach to the identification

of system threats and human error from observations of undesirable events in the

operating theatre, it was hoped to measure the relative incidence of different sources of

failure within the system of surgery.

3.2 Ethics and Consent Issues

Since error in surgery is highly emotive for both patients and staff, the use of video

recordings of intraoperative events required particular sensitivity. To that end, a number

of ethical protocols were adopted. Firstly, as well as ensuring the data was recorded

anonymously, access to the video tapes was limited to the two observer researchers.

This was considered essential to maintain anonymity, particularly where other members

of the research team might have been asked to view the performance of their

colleagues, and was explicitly requested by the participating staff. Unfortunately, it

limited the use of the videos, since they could not be viewed by other technical experts

or used for debriefing purposes, as had been planned. Patient access to the video data

was not encouraged, since there was concern that they might cause distress, or

unwarranted medico-legal issues, and in the event no patient requested a copy of their

own data. A further protocol required that the video tapes were deleted upon completion

of the project. These safeguards ensured that the anonymity of both patient and staff

participants was protected. However, for optimal benefit of intraoperative video

techniques, we would recommend re-examination of these constraints. While patient

anonymity must be protected, that is already a duty of the healthcare team, and the

capacity of an individual to veto the taking and presentation of any record of their

involvement with patient care must be balanced by the wider public interest. This conflict

is not unique to medicine. Aviation, in particular, has developed methods by which the

fear of scapegoating and litigation, which may underpin inappropriate professional

sensitivity, is contravened by the use of intra-professional but independent review.



Given that personnel in operating theatres can change frequently, and there may be

additional entrants to operating theatres who had not been previously informed about the

study, there were particular difficulties in ensuring that all staff featured on the video had

given specific prior consent. Since this was a condition of this study, it was addressed in

several ways. Firstly, all nursing, surgical and anaesthetic staff were approached and

informed about the study prior to commencement. Secondly, wherever possible,

individuals who appeared in theatre unexpectedly were informed about the study as

soon as it was convenient to do so. Finally, all participants were given the option to

withdraw from the study should they wish to do so, which would result in cessation of the

recording and deletion of the video tape.

When attempting to obtain consent from staff, these safeguards were considered

necessary. However, concerns about the nature of the study remained, with particular

unease over the use of video. This was particularly acute in the orthopaedic centre,

where it was possible to recruit only one surgeon. Despite a range of presentations,

phone calls, letters and emails, only three consultant surgeons responded to our request

for participation. One was involved mostly with trauma, so was unsuitable for the study.

The second was happy to participate, but not to be video-recorded, and would not allow

the research team to approach his patients for consent. A single consultant surgeon was

willing and able to participate in these studies. Obtaining consent from theatre staff and

anaesthetists did not meet with any noteworthy concerns or resistance.

In the paediatric cardiac centre, all the surgeons agreed to participate, and while some

of the theatre staff in this centre were initially uncertain, all eventually agreed to

participate. However, the anaesthetists were uneasy with the study in this centre, and as

a result, observations with particular anaesthetists were not possible. This unease was

symptomatic of organisational difficulties between anaesthetists and surgeons, which

were also a source of recurrent problems in theatre. The difficulties experienced here

were indicative of a healthcare-wide culture that may see surgical error as something to

be ignored or feared, rather than to be studied and understood. It suggests a serious

misunderstanding of professional responsibility to the best interests of the patient, then

need for openness in performance matters, and a lack of commitment to the wider public



interest. This is of even greater significance when one considers that the outcome of the

patients in all the operations studied was good.

Patients were approached for consent in the orthopaedic centre during their pre-

operative assessment, which was conducted between one and three weeks before the

scheduled date of their operation. This consent process, which lasted four months,

identified 49 potentially suitable patients, of which 8 did not arrive for their pre-operative

assessment, and 8 declined to participate or were judged by the researcher to be

unsuitable for study. Consequently, 33 patients were recruited and consented to take

part, of whom 20 were eventually studied. This final attrition was due to the cancellation

or rescheduling of the operation and a range of other issues beyond the control of the

research team.

Parental consent was obtained for all observed paediatric cardiac cases. Since these

cases were of far greater risk than the orthopaedic cases, and far more distressing for

the parents of the patients involved, a great deal of sensitivity was required. An

information sheet was sent out with the admission instructions for elective cases, which

gave notice to parents that the research team might wish to approach them, and allowed

them to reject the approach at an early stage. Forty-six were returned, 42 of which

responded positively to our request. While this was effective for elective patients, only a

few patients were studied who had been prepared for consent in this way as it was not

suitable for the emergency or high-risk cases, such as ASO and Norwood operations.

The research team had little prior notice as to when operations would take place.

Instead, emergency cases were identified through examination of operative lists, and

regular contact with a range of hospital staff. Patients and their parents were then

located and approached with the permission of the nursing staff, with the total consent

process for each patient lasting between 2 and 4 hours. A total of 51 parents were

approached for consent, of whom 46 agreed to take part. Due to changes in the

operating lists, changes in the operating theatre, withdrawal of theatre team members on

the day, and the need to avoid MRSA cases, of 46 recruited patients it was possible to

study 25.

Overall, it was felt that many of the co-operation and consent issues – particularly the

refusal of some staff to participate and the constant changes in theatre lists – were



themselves symptomatic of some of the system problems and cultural handicaps

encountered in surgery. These issues pose significant difficulties for the study of error in

surgery and for any programme of improvement.

3.3 Observational Methods

General Approach

A single observer was present during each case, and utilised both implicit and explicit

data collection techniques40. Observations were classified into major and minor failure

categories, and minor failures were then classified according to one of 29 types. These

minor failure types were systematically associated with threats and errors through a

weighting technique described as the failure source model. This provided the method by

which systemic contribution to intraoperative failures and human errors could be

measured. In all cases a video recording was made of the operating theatre team from

two views, and in the paediatric cardiac cases the surgical field and video output from

the anaesthetic workstation were also recorded. Risk scores, explicitly based upon the

procedure performed, were assigned to each case, utilising the RACHS-1 method48 in

cardiac surgery, and a simple 3 level designation in orthopaedic surgery. Operative

duration (first incision to final closing suture), was recorded in order to calculate threat

and error rates, as well as offering a surrogate measure of intraoperative performance.

Cardio-pulmonary bypass (CPB), deep hypothermic circulatory arrest (DHCA), and

tourniquet times were also collected where appropriate, but were not used since there

were systematic problems with these measurements. In particular, patients subjected to

DHCA had time off CPB, which meant the two measures interacted. In orthopaedics, the

tourniquet was removed before skin closure in the more difficult cases, and not in the

easier cases, again allowing misrepresentation of tourniquet time as a measure of risk or

performance. Patient weight (for paediatrics) and age (for orthopaedics) and the

composition of the surgical team were used to provide qualitative detail to the case mix.

Expert Observers

It was planned that both observers should have suitable experience to observe both sets

of operations. Observer one was a human factors practitioner, experienced in

observational methods and human performance measurement. During the preparation

phase prior to data collection in cardiac surgery, technical knowledge was obtained by

studying text-book descriptions of the procedures, and observing 38 similar operations,



from which the task analysis and an explicit procedural-based error-capture checklist

were produced. The same process was adopted for orthopaedic surgery, though only 10

training operations were observed since these were technically less complex and more

proceduralised. Observer two was a researcher with prior experience of working in

operating theatres. He was familiar with specific elements of anaesthesia, though had

little previous observational or human factors experience. Both observers received

training in the evaluation of non-technical skills from professional pilots who had many

years of experience in observing and training flight crews in both technical and non

technical areas. These observation techniques were central to the assessment of the

surgical system, since it did not bias observation toward the technical skills of any

individual specialist. Observer two did not develop sufficient understanding of both

surgery and human factors after viewing 10 operations to make suitable observations7 in

cardiac theatres, though he had obtained a measurable and reasonable level of

observational ability following 10 training operations in orthopaedic surgery. The

specialty bias that observer two brought is also discussed further in section 5, and

particular difficulties that this observer encountered in assessing non-technical skills are

examined in section 7.

Observational competency was assessed by examining the degree of agreement

between observations of time-marked events, and through the ongoing assessment of

intraoperative reports and vignettes by technical experts. The former is reported in

chapter 5, the latter in chapter 6. The ethnographic nature of the studies also allowed the

observers frequently to check and develop their technical understanding through

discussions with healthcare professionals, on an ad-hoc basis, and through participation

in clinical governance forums.

Observation Technique

A summary data sheet was completed for each operation, including patient details,

procedural details, team composition, and a range of other case-specific variables. From

the task analysis, explicit procedure-based observational error-capture checklists were

produced for TKR, THR and paediatric cardiac surgery. These checklists, which included

specific time-marked events, allowed structured error-capture observations using

surgical task descriptions coupled with specific local practices and protocols and more

general elements of safety, culture and teamwork. The checklist for cardiac surgery was



modular, consisting of sequences that were similar across different procedures, but

allowing for adaptations by the surgical team (such as preferred cannula configuration).

For example, procedures for midline sternotomy, cannulation for CPB, initiation of CPB,

wean from CPB and chest closure were similar across all cardiac operations, but some

modules required procedure, patient or surgeon specific variations. As far as possible

these alternatives were included in the modules. Other modules were specific to the

planned surgical procedures. The THR and TKR orthopaedic checklists were generally

invariant for elective cases, but could not provide sufficient flexibility for the less

predictable revision operations. Since surgery must be adapted to the needs of the

patient, and the preferences of the team, the observer needed to be flexible, and did not

always strictly adhere to the observation checklist. Rather, it provided a guide as to what

should happen when, and allowed the observer easily to keep track of each operation

and record the context in which unplanned events might be observed. A complete data

collection pack for TKR operations and Arterial Switch operations can be found in

Appendix 4. From our pilot studies it was apparent that this explicit error capture

technique would not be sufficient, so it was also necessary to make detailed notes of

activities and communications. This produced descriptions of events in theatre and the

time at which those events occurred. Case reports were produced for the more complex

paediatric cardiac cases, and were recorded using spreadsheets for the less complex

orthopaedics cases.

Two levels of observation were used. Major failures were described as events that came

close to an incident or accident42. Minor failures were identified where observed events

were judged to have had small negative effects on the duration or difficulty of the

operation, the risk to the patient, or the demand for resources. Where major failures or

unusual or complex minor failures occurred, brief reviews were conducted with the

individuals involved within 24 hours of the operation to ensure that the appropriate

specialist information had been recorded. Minor failures will be discussed in more detail

in section 3.4.

Video Configuration

In all but one of the 45 cases studied a video recording was made of the events in the

operating theatre. In one case the equipment failed, and was disregarded in subsequent

analysis. The installation of the video equipment in the operating theatre at each site



required the approval of the theatre manager, nursing theatre teams, infection control,

and biomedical engineering. This process took several weeks, and meant that it was

possible to use the video equipment in only one operating theatre in each centre. On the

day of the operation, the equipment was installed in the operating theatre prior to or

during anaesthetic induction, and was uninstalled and removed from the theatre after all

study operations for that day were complete. A careful balance had to be struck between

the need for obtaining good visual and audio fidelity, and the potential hazard and

general disruption caused by the presence of the video equipment in the limited space of

both operating theatres. The solutions for each operating theatre can be found in figures

3.1 and 3.2.

Figure 3.1: Video equipment configuration for paediatric cardiac surgery

S

1A

SN

P

AC CN

2A

AR

Pumps & Drips

Perfusion Machine

Anaesthetic Workstation

2

1

3

Video Equipment

Stack

Video Cameras 1 = Wide-angle 2 = Narrow angle 3 = Lamp camera 4 = Data feed

Operating Theatre Team S = 1st Surgeon SN = Scrub Nurse 1A, 2A = Assistant surgeons CN = Circulating nurse AC = Anaesthetic Consultant P = Perfusionist AR = Anaesthetic Registrar

4



Figure 3.2: Video equipment configuration for orthopaedic surgery Figure 3.3: Output from the video recording. The date of the operation has been deliberately obscured.

S

1A

SN AC

CN

Drip

Anaesthetic Workstation

2

1 Video

Equipment Stack

Video Cameras 1 = Wide-angle 2 = Narrow angle

Operating Theatre Team S = 1st Surgeon SN = Scrub Nurse 1A = Assistant surgeon CN = Circulating nurse AC = Anaesthetic Consultant

Pump



Figure 3.4: Images of video equipment installed in the paediatric cardiac operating theatre. For the paediatric cardiac study, the video recordings were used to examine each

operation in order to check, and provide further detail to, in-theatre observations. The

operating theatre already had a lamp camera installed, and the video recording

equipment interfaced easily with the data feed from the anaesthetic workstation. The

recording of audio was particularly useful in paediatric cardiac surgery, since the

placement of the microphone and the noise generated by the theatre equipment meant

that some conversations were more easily audible on the video and others were more

easily picked up by the observer in situ. There was considerable individual variation in

the audibility of voices on the tape, and simultaneous speakers were sometimes very

difficult to distinguish, but the obvious advantage of video was that the recording allowed

word-for-word transcription of the events. Though this detailed analysis of the video

tapes obtained in paediatric cardiac theatres was time-consuming, it was also rewarding

as it provided the means to identify the antecedent causes of subsequent failures, as

well as the ability to confirm the in-theatre observations. It was also used to perform the

NOTECHS non-technical skills evaluation in paediatrics. The video data in the

orthopaedic operations was less useful since the orthopaedic operations were far less

complex, and involved two observers, rather than the one in paediatrics. In addition, the

lamp camera view was not available, and the anaesthetic workstation could not be



interfaced with the video recording equipment. Consequently the video data was used

only to monitor or check the results of the observers in the orthopaedic operations.

3.4 Analysis Methods

Minor and Major failures

Following the completion of the cardiac operations, the detailed event descriptions were

used to group the minor failures into 29 different types that provided equivalence across

all the studied procedures. Careful definitions for each of the minor failure types (table

3.1), and re-classification following changes to the classification system, ensured

consistency. Using the videos, observer notes, and reported events, each of the major

failures was also deconstructed into the minor failures from which it was formed. This

allowed the description of events in terms of the number and types of failures that could

be observed, allowed the identification of failure sequences that were present, and

provided a consistent way in which to observe error-related behaviours. It also ensured

that the appropriate medical perspectives had been assimilated. For orthopaedic

surgery, minor failures were classified according to the 20 types already defined in

paediatric cardiac surgery. Nine failure types found previously were not observable in

orthopaedics, five of which were related specifically to cardiac surgery. The remaining

four failure types (fatigue, external pressures, fault resolution failure and failure to

address known problem) were not observed. All but one major failure occurred in the

paediatric cardiac cases. These were documented as vignettes for section six, which

also details how these major events were deconstructed into the minor failures from

which they formed.



Failure Description & Example Absence Lack of personnel when required.

Example: Circulating nurse is absent when scrub nurse needs more suture material.

*Cannulation difficulties

Problems encountered during arterial or venous cannulation. Example: Arterial cannula is not cleanly inserted first time, and some blood loss results.

Co-ordination / communication failure

Failures in task co-ordination and communication between individuals. Cardiac example: Ventilation is not switched off before sternum is split with saw. Orthopaedic example: Surgeon asks for the drill, but the scrub nurse is busy doing something else.

Decision-related surgical error

Surgeon fails to make the appropriate decision. Cardiac example: Surgeon asks for a delay in giving the heparin when heparin is required. Orthopaedic example: The surgeon decides the tibial cut is satisfactory, then later finds it is not.

Distraction Disturbance from external sources not related to current case during a critical period. Examples: (i) Mobile phone rings in theatre; (ii) Another consultant enters theatre and distracts the surgeon.

Equipment / workspace management failure

Failures in the organisation of workspace and equipment. Examples: (i) the defibrillator is unplugged when required for; (ii) infusion lines become caught on clothing.

Equipment configuration failure

Failure to set-up or operate equipment appropriately. Cardiac example: Transducer is not zeroed correctly, leading to false readings. Orthopaedic example: Intramedullary rod not inserted far enough into femur

Equipment failure Interoperative equipment failure. Examples: (i) Sutures break; (ii) faulty transducer

Expertise / skill failure

Failures associated with lack of expertise or skill. Example: (i) consultant surgeon captures error made by trainee surgeon; (ii) assistant surgeon doesn’t know how to use the saw correctly;

**External pressures Requirements of the external organisation that impact upon the operation Example: team rushing to complete current case in order to start the next case

External resource failure

Failures in elements of the external organisation to provide equipment or human resources. Examples: (i) Lack of blood products; (ii) Piece of equipment is missing from the standard set

**Fatigue Clear indication that individuals in theatre are suffering from lack of sleep. Example: anaesthetist falling asleep during CPB

**Fault resolution failure

Failure to identify sources of problems. Example: Anaesthetist and perfusionist unsure why blood pressure is low, and have only minimal discussions.

Patient-sourced procedural difficulties

Features of the patient that make the planned procedure more challenging to carry out than would be expected from the pre-operative diagnosis. Cardiac example: LeCompte manoeuvre impossible in arterial switch operation due to anatomical restrictions Orthopaedic example: Not enough femur remaining to make box cuts.

**Failure to address known problem

Failure to address known risk to operative success. Example: Operating team know of serious lim itation in current provision of resources but do not consider means of reducing occurrence or potential effects.

*Perfusion difficulties: non-technical

Problems with management of CPB or cardioplegia, not sufficiently addressed. Example: The perfusionist cannot keep the surgical field clear of blood due to anaesthetic induced vasodilatation.



Failure Description & Example *Perfusion difficulties: technical

Difficulties with perfusion sufficiently discussed. Example: Cause of low pressure identified as being due to aortic cannula obstruction.

Planning failure Failure to anticipate or discuss future task requirements. Example: Surgeon consults patient notes after the start of the operation. Cardiac example: Haematocrit target is agreed after modified ultrafiltration has started.

Pre-operative diagnosis failure

Failure to provide accurate diagnosis prior to operation. Cardiac example: Undiagnosed intramural coronary arteries found upon anatomical inspection Orthopaedic example: Surgical team have decided upon an implant from one manufacturer, but the x-rays show the previous implant to be from another manufacturer.

Procedure-related error

Procedural errors by surgeon, assistant surgeon, anaesthetist, perfusionist, scrub nurse, or circulating nurse. Cardiac example: filtration finishes before haematocrit is checked. Orthopaedic example: cement mixing time is not recorded.

Psychomotor error (general)

Handling errors. Example: retractor is dropped

Psychomotor-related perfusion error

Technical manipulation errors by perfusionist Example: cardioplegia pump is run too fast

*Psychomotor-related surgical error

Technical manipulation errors by surgeon Examples: (i) incorrectly placed sutures are removed and suturing starts again; (ii) assistant surgeon gets forceps tangled in sutures.

Resource management

Failures in the organisation of available people or things in the operating theatre Example: consultant leaves theatre to find cardiologist, unnecessarily removing senior member of the team; surgeon leaves assistant surgeon to close without confirming ability to do so.

Safety consciousness

Failures to observe basic elements of patient safety Examples: (i) Mask is not fitted on entry to theatre; (ii) student chewing gum in theatre

Team Conflict Team members have differing opinions, or give conflicting commands, that are not resolved Example: scrub nurse and assistant surgeon argue over procedural requirements.

*Temperature control difficulties

Problems with management of patient temperature Examples: (i) patient temperature overshoots to 37°; (ii) patient temperature falls unnoticed.

Unintended effects on patient

Problems arising with the patient, as a result of the treatment, that are unplanned Example: Unexpected post-treatment bleeding; blood pressure increases unexpectedly after bag of fluids is changed

Vigilance / awareness failure

Failures to notice immediately important aspects of the task or the patient Example: (i) surgeon fails to spot bleeding sites; (ii) anaesthetist does not notice a drop in blood pressure.

* Failure exclusive to cardiac surgery ** Failure possible in orthopaedic surgery, but observed only in cardiac surgery Table 3.1: Descript ions and examples of minor fai lure types.



Failure source model

Minor failures formed a level of quantifiable undesirable events observed in the

operation that sometimes reflected:

• Individual errors

• Failures in group processes

• Threats outside theatre that affected events in theatre

• Combinations of all three.

As most failures arose from an interaction between several threats and errors, a single

classification for each observable failure was insufficient. Furthermore, sometimes these

threats or errors were not directly observable. For example, a difficult cannulation arises

because of the interaction between the task requirements for cannula sizing and

insertion, the anatomy of the patient, and the technical ability of the surgeon to

compensate for the limitations of both. Following extensive examination of the video

tapes, detailed considerations and further discussions with a range of medical and

safety-related practitioners, each of the 29 minor failure types was associated with one

or more threats and errors, according to a failure source model. Table 3.2 provides full

definitions of the threat and error types, and their association with different types of

minor failure, which formed the failure source model11;18;35;43. Patient threats and cultural

/ organisational threats were both associated with 7 different minor failure types, task

threats were associated with 16 different minor failure types, and environmental threats

were associated with 4 different minor failure types. Technical errors were associated

with 11 different minor failure types, and non-technical errors were associated with 14

different types of minor failure. As it was not always possible to observe the specific

source of the failure, the model assumed that all potential sources of error in each failure

had contributed in equal proportions. Applying this failure source model allowed the

generation of a threat and error profile for each operation43.

The failure source model was applied in the same way for both paediatrics and

orthopaedics. This allowed the description of events in terms of the systemic threats and

human errors that they represented, and allowed direct equivalence between operations.

This provided the final systems analysis of intraoperative failures in surgery.



Failure Source

Component Definition Associated failure types

Cultural and Organisational Threats

Threats that arise due to aspects of the organisation or culture.

Absence; Distraction; External pressures; External resource failure; Fatigue; Known problem; Safety consciousness

Patient Threats Threats relating to patient anatomy and physiology beyond those specified pre-operatively.

Cannulation difficulties; Patient-sourced procedural difficulties; Perfusion difficulties: non-technical; Perfusion difficulties: technical; Pre-operative diagnosis failure; Temperature control difficulties; Unintended effects on patient

Tasks Threats Threats arising from the processes, protocols or techniques employed to complete the operation, beyond those specified pre-operatively.

Cannulation difficulties; Equipment / workspace management failure; Fault resolution failure; Patient-sourced procedural difficulties; Perfusion difficulties: non-technical; Perfusion difficulties: technical; Planning failure; Procedure-related error; Psychomotor-related perfusion error; Psychomotor-related surgical error; Pre-operative diagnosis failure; Temperature control difficulties; Unintended effects on patient

Threats

Environmental Threats

Threats that arise from deficiencies in equipment, workspace & resources

Equipment / workspace management failure; Equipment configuration failure; Equipment failure; External resource failure

Technical Errors

Human errors associated with knowledge, technical skill or expertise.

Cannulation difficulties; Decision-related surgical error; Equipment configuration failure; Expertise / skill failures; Fault resolution; Perfusion difficulties: non-technical; Perfusion difficulties: technical; Psychomotor error (general); Psychomotor-related perfusion error; Psychomotor-related surgical error; Temperature control difficulties/

Errors

Non-technical Errors

Human errors associated with team working and general cognitive skills.

Co-ordination / communication failure; Decision-related surgical error; Equipment / workspace management failure; Fault resolution failure; Perfusion difficulties: non-technical; Planning failure; Procedure-related error; Resource management failure; Team conflict; Temperature control difficulties; Vigilance / awareness failure

Table 3.2: Sources of fai lure. This defines the threat and error types, and shows how they were associated with observed failures in the failure source model.



Non-Technical Skills Analysis

In order to focus on the influence of teamwork and general cognitive skills of the theatre

teams, two separate non-technical skills assessments, an error-based assessment

method, and a global scoring method, were conducted. Both were based on the

NOTECHS behavioural marker system currently in operational use in the airline industry,

which had been carefully developed to allow consistent assessment across different

national and organisational cultures49. NOTECHS consists of four dimensions, with

several elements within each. Table 3.3 summarises the elements and dimensions of the

Surgical NOTECHS system. These elements can be identified by a trained observer,

and are provided in generic form, which allows broad application of the scale. It was this

broad applicability, demonstrated utility49;50, operational validation51 and the success with

which the scale has been adapted to specialties in medicine previously46;52 that we

selected this behavioural marker methodology. NOTECHS was adapted for use with

surgical teams following consultation with two aviation human factors trainers, two

cardiac surgeons, one vascular surgeon, and one orthopaedic surgeon. The surgical

NOTECHS scale can be found as part of the TKR data collection pack in Appendix 4.

Dimensions Elements Leadership & Management Leadership

Maintenance of Standards Planning & Preparation Workload Management Authority & Assertiveness

Teamwork & Co-operation Team building & Maintaining Support of others Understanding team needs Conflict solving

Problem Solving & Decision Making

Definition & Diagnosis Option Generation Risk Assessment Outcome Review

Situation Awareness

Notice Understand Think Ahead

Table 3.3: Surgical NOTECHS dimensions and elements . Copies of the complete scale can be found in the TKR data collection pack in Appendix 3. From the data produced by the minor failures and failure source model, it was possible

to identify the non-technical errors that had been made by the surgical teams. For the

purposes of understanding exactly the non-technical deficiencies found in the 24



operations, each of these errors were assigned to an element on the NOTECHS

instrument. This provided a diagnostic assessment of the deficiencies in non-technical

performance found across the different operations studied. However, it was also useful

to allow an assessment of positive behaviour. Consequently, the NOTECHS scale was

applied to each team by and assigning a score from 1 to 4 to each of the four NOTECHS

dimensions (table 3.4). This was conducted once each for three pre-defined periods of

the operation. The first phase, described as the access phase, started with the first

incision, and lasted until the patient was successfully on cardio-pulmonary bypass. The

second phase, known as the treatment phase, followed directly afterwards, and lasted

until the final suture in the planned treatment for the patient was finished. The final

phase, known as the closure phase, lasted from this moment, to the moment that the

final suture in the closure of the chest was tied off. The observer recorded his

observations on the scoring sheet, in order to promote consistency and balance of

judgement intraoperatively and interoperatively. For paediatric cardiac surgery, the

global NOTECHS evaluation was conducted entirely from the video-taped operation53. In

orthopaedic surgery, it was conducted by the observers in situ. Operations were then

ranked from best to worst according to the number of “below standard” scores obtained

in each operation, then the number of “basic standard” scores, then “standard” and

finally “exceed” scores. This gave an overall order to each operation purely in terms of

the positive and negative non-technical behaviour observed.

Below Standard

(1) Basic Standard

(2) Standard

(3) Exceed

(4) Behaviour directly compromises patient safety and effective teamwork.

Behaviour in other conditions could directly compromise patient safety and effective teamwork.

Behaviour maintains an effective level of patient safety and teamwork.

Behaviour enhances patient safety and teamwork. A model for all other teams.

Table 3.4: Scoring categories for surgical NOTECHS behavioural marker scale

3.5 Summary

• The study of error in healthcare raises some considerable cultural difficulties,

particularly in access to and ownership of data.



• The video recording of operations was useful for examining the aetiology of failures

that might have begun a considerable time before the failure became apparent.

However, scrutiny could result in further professional sensitivity and the value of this

technique may therefore be incompletely realised.

• Expert observers were prepared through study, observation, and the process of task

analysis, and were assessed throughout the course of study.

• The observation methodology required a battery of measures:

o Error-capture checklist, based on the prior task analysis

o Unstructured note-taking

o Non-technical skills evaluation

o Post-hoc discussions with participants and other technical experts

o A standard range of intraoperative parameters

• The intraoperative reports, videos, and subsequent failure, threat and error analyses

provided an immensely rich source of data.

• Structure and consistency was achieved in the analysis by distinguishing minor

failures from major failures, and applying a failure source model that made

assumptions about the threats and errors that underlie the causes of failure.

• Non-technical skills were evaluated specifically using a scale derived from the

NOTECHS technique employed in civil aviation.



4. Systems Failures in Paediatric Cardiac Surgery

4.1 Introduction

The first surgical domain studied was paediatric cardiac surgery. The aim was to develop

a method to measure systemic deficiencies in the surgical system, to examine the

frequency and types of failure in cardiac surgery, and investigate the propagation of

minor to major failure. Since errors are more likely when systemic demands are already

high54, we hypothesised that high risk cases would have more minor and major failures.

We expected a similar effect in longer operations since operative duration is closely

related to operative risk and intraoperative complications55-57. Finally, we hoped to apply

the methodology to identify systems solutions that can lead to improvements in patient

safety in paediatric cardiac surgery. Table 4.1 illustrates a typical paediatric cardiac

procedure.

Key Phases of Surgery in the Arterial Switch Operation The arterial switch operation corrects an otherwise fatal congenital heart abnormality known as transposition of the great arteries, where the child is born with aortic and pulmonary arteries connected to the heart in the opposite way to a normal heart. In the operation, which takes between five and seven hours, and is performed within the first two weeks of life, the two arteries are swapped to provide the anatomical correction.

1. The patient is anaesthetised in the anaesthetic room which adjoins the operating theatre.

2. The patient is transferred to operating table in the operating theatre.

3. Once the patient is ready, the first incision is made.

4. The sternum is split and the ribs are spread.

5. Heparin is given to the patient. This increases the clotting time of the blood to allow cardio-pulmonary bypass.

6. Cannulae are inserted into the ascending aorta and the atrial appendage, and attached to the bypass machine.

7. Cardio-pulmonary bypass is initiated, and ventilation is switched off

8. Patient is cooled to between 18°C and 28°C to protect against neurological damage.

9. Cross-clamp is applied to the aorta, and cardioplegia (a potassium based solution) is infused into the coronary arteries to prevent the contraction of the heart muscle during the operation. This is re-applied approximately every 20 minutes for the duration of the treatment.

10. The aorta and pulmonary artery are transected, the coronary arteries are mobilised from the aorta.

11. The aorta is stitched back to the great vessel arising from the left ventricle.

12. The coronary arteries are connected back to the new aorta

13. The great artery arising from the right ventricle is reconstructed, and stitched to the pulmonary artery.



14. If present, the ventricular septal defect is closed.

15. The cross-clamp is removed.

16. The patient is slowly warmed and the heart is observed to beat normally.

17. The patient is weaned from bypass.

18. Electrocardiography is used to assess the stability of the patient. An echocardiogram may be conducted to examine the success of the repair.

19. Protamine is introduced to the patient to reverse the effects of the heparin, and bleeding is controlled. Various drugs, drains, and cardiac pacing may also be utilised to aid the recovery.

20. The chest is closed and the patient is transferred to the intensive care unit. Tab le 4.1: Surgery to correct congenital heart defect. A brief description of one type of operation to correct a congenital heart defect.

4.2 Method

An opportunity sample of 25 paediatric cardiac cases was studied between October

2003 and July 2004. The general method adopted has been described previously.

Observer one was present for all cases, and one case was discarded because the video

equipment failed. The data were grouped into low risk (Jenkins48 levels 1 and 2) and

high risk (Jenkins48 levels 3, 4 and 6) categories, and threat and error rates were

calculated for each operation using operative duration as the denominator.

4.3 Results

Case mix

None of the operations studied were considered to be outside normal system function, in

terms of the individuals that were involved in the operations, the procedures that were

planned, or the condition of the patient as was understood by the surgical team prior to

surgery. Composition of the surgical teams varied considerably, and though the same

members were frequently involved, all members and roles were not identical. Operative

duration ranged from 110 mins to 369 mins, with a mean of 236 mins and a standard

deviation of 74 mins. One operation was level 1 risk, thirteen were at level 2 risk, three

were at level 3 risk, four were at level 4 risk, and three were at level 6 risk48. The age of

the patients ranged from two days to nine years and weights ranged from 2.5 kg to 19.3

kg. No patient died within 28 days of the operation, and no failure was deemed worthy of

further investigation either by the individuals involved, or by the hospital.



The only mortality during the study period was known to be a challenging clinical

problem prior to our observations and this patient died several months later following

several additional operations. The operation studied appeared to be more successful

than expected. Since, coincidentally, this was also the case in which the video

equipment failed, no further assessment was made, and this patient was omitted from

subsequent analyses.

Minor and major failures

Figure 4.1: Minor and major failures by operation and risk level. Operations are organised on the y-axis by risk level (Jenkins et al. 2002), and then by increasing numbers of minor failures. Shaded areas indicate those minor failures sequences that were implicated in major failures. The “low” and “high” risk categories that were used in the analysis are also indicated here. In the 24 operations studied, a total of 366 minor failures were identified. Figure 4.1

shows the distribution of minor failures in the study which varied between 2 and 34 in

each operation, with a mean of 15.3 and a standard deviation of 7.3. A two-tailed

Spearman’s rho correlation showed a moderate relationship between operative duration

and the number of minor failures in an operation (rho = 0.647, n = 24, p<0.001). Seven

operations contained major failures which came close to an incident or accident. These

are indicated in figure 4.1, and are described in more detail in section 6. The operations

0

5

10

15

20

25

30

35

40

Operations by Risk Level

Min

or

Fai

lure

s O

bse

rved

Level 2 Level 3 Level 4 Level 6Level 1

"Low" Risk Operations

"High" Risk Operations

Minor failures implicated in major failure



where major failures occurred generally had more minor failures than those operations

that did not feature any major failure. A Mann-Whitney U test between these two groups

showed a significant difference between them (u = 14.0, n = 24, p<0.005), though this

difference disappeared when those minor failures were removed that were directly

associated with the major failure. In terms of operative risk, the proportion of operations

experiencing major failures increased in the higher risk operations. Grouping the minor

failures into low risk (levels 1 and 2) and high risk (levels 3, 4 and 6) categories (Jenkins

et al. 2002), a Mann-Whitney U test (u = 29.5, n = 24, p<0.05) showed a significant

difference between the numbers of minor failures found in each group. Thus, the number

of minor failures and major failures encountered in each operation related both to risk

and operative duration.

Figure 4.2 shows the mean number of observations of each type of minor failure per

operation. The minor failure type of highest observed frequency – co-ordination (1.95 per

operation) – related to problems either in the distribution of important information within

the surgical team, or to the timely execution of procedural sequences requiring

interaction between individuals. Though process-related, and therefore not always of

direct threat to the patient, this type of failure might have serious consequences at

certain points in the operation, and was a feature of all but two major failures. Absences

that had some form of negative effect on the procedure were also frequent (1.67 per

operation), as were equipment problems, either in terms of a direct failure (such as non-

serviceable instruments or malfunctioning equipment; 1.21 per operation) or in terms of

difficulties in the operation of equipment (1.04 per operation). Though this was non-

critical equipment, the failure of which could usually be compensated for, it was often the

same pieces of equipment that created problems in different operations. As might be

expected with this type of surgery, patient-related difficulties, usually anatomical in

nature, appeared on average once per operation, requiring a more complicated surgical

strategy and greater surgical skill to complete successfully.



0 0.5 1 1.5 2 2.5 3 3.5 4

Decision-related surgical errorKnown problem

FatigueExternal pressures

Psychomotor-related perfusion errorResource management

Fault resolutionPsychomotor Error (general)

Pre-operative diagnosis failureTemperature control difficulties

Planning failureExpertise / skill failure

Perfusion difficulties: technicalTeam Conflict

Vigilance / awarenessProcedure-related Error

Cannulation difficultiesExternal resource failure

Psychomotor-related surgical errorPerfusion difficulties

DistractionEquipment / Workspace management

Unintended effects on patientSafety consciousness

Patient-sourced procedural difficultiesEquipment Configuration failure

Equipment failureAbsence

Co-ordination / communication

Mean number per operation Figure 4.2: Mean rate per operation of minor failure types. Minor failure types are organised with the most frequent at the top, and least frequent at the bottom. Error bars show mean plus one standard deviation. Minor failures that appeared on average in a quarter of operations or more, but less than

once per operation, were a mixture of procedural, perfusion, skill, teamwork and cultural

difficulties. On average, in most operations failures were observed that suggested

reduced safety consciousness (0.92 per operation), usually relating to mask discipline in

theatre. The failure to control distractions was similarly related to culture (0.75 per

operation). The patient reacted to the treatment in a way that complicated the procedure

and creating a greater challenge for the team on average every 0.83 operations.

Perfusion management problems (0.29 per operation) arose through the tolerance of the

patient to CPB, the CPB requirements for a given procedure or task, and technical errors

in the management, or configuration of the cannulae, which were often difficult to insert

(0.46 per operation). Perfusion problems were more often than not compounded by



problems in co-ordination between surgeon, anaesthetist, and perfusionist (0.625 per

operation), which sometimes made it difficult to ascertain the precise nature of the

problem. Control of patient temperature is a similar task, failures were considerably less

frequent (0.25 per operation). Conflicts either in task or understanding occurred in nearly

one-third of cases (0.29 per operation). Problems with the management of equipment

and workspace – ensuring everything was in the right place when needed – were among

the more frequent failures (0.79 per operation), as was the provision of equipment,

supplies and personnel external to theatre (0.5 per operation) whilst the direct impact of

planning failures was less apparent (0.29 per operation). Given the length and technical

demands of the operation, surgical errors (0.5 per operation), procedural errors (0.37 per

operation), vigilance and awareness (0.33 per operation) and other expertise or skill

failures (0.29 per operation) were relatively infrequent, though they were also difficult to

observe.

The minor failures which appeared on average in less than one quarter of cases were

also mixed, with failures to address a known problem and in decision making (both 0.04

per operation, 1 each observed) the least frequent, while evidence of fatigue, external

pressures and perfusion control failures were only slightly more common (all 0.08 per

operation, 2 each observed). Three observations each were noted for general

psychomotor errors, fault resolution failures, and resource management failures (0.125

per operation). Finally, four pre-operative diagnosis failures were observed (0.166 per

operation), one of which was a particularly difficult undiagnosed intramural coronary

artery pattern.

Threat and error failure source model

From the failure source model it was possible to examine the systemic sources of threat

and error. Figure 4.3 shows the 366 minor failures grouped into the types of threats that

were present. Cultural and organisational threats were the most frequently encountered

single type of threat. Task threats were the most common, frequently appearing in

combination with patient and environmental threats, though not as frequently on their

own. Patient threats always appeared in conjunction with task threats. Figure 4.4 shows

the minor failures in terms of the types of errors that they reflect. The most numerous

error was non-technical, but nearly half the failures did not feature any observable error



by the team. Only a small proportion showed evidence of both technical and non-

technical error.

Figure 4.3: Sources of threat and error. This shows the numbers and proportions of threats and errors identified from the 366 failures using the failure source model. The left panel shows threats, the right panel shows errors. Using this information, the final stage examined the individual instances of threats and

errors associated with the minor failures. From the 366 minor failures, a total of 406

threat instances were identified, with a mean of 16.9 per operation and a standard

deviation of 8.2. A total of 218 errors were identified, with a mean of 9.1 per operation

and a standard deviation of 5.2. Operative duration correlated moderately with the

number of intraoperative threats (rho = 0.616, p<0.005) and weakly with the number of

intraoperative errors (rho = 0.410, p<0.05). Task-related threats were judged to be the

most numerous, with a total of 139 instances. Cultural and organisational threats

accounted for just under a quarter of the total number threats with 97 instances, with

patient- and environment- related threats reflecting similar numbers of instances (87 and

85 respectively), each accounting for just over one fifth of the total number of threats. In

both high-risk and low risk groups, task threats occur at a higher rate than other types of

threats, as can be seen in Figure 4.4. There is little difference between the rate of threat

occurrence in high-risk and low risk operations. Of the errors observed in this study, 125

were non-technical errors, and 92 were technical. Non-technical errors occurred at a

Environmental and Cultural /

Organisational, 12, 3%

No Threat(Error only),

76, 21%

Cultural / Organisational, 85,

23%

Patient and Task, 87, 24%

Task, 33, 9%

Environmental, 54, 15%

Task and Environmental, 19,

5%

No Error (Threat Only), 174, 48%

Technical Errors, 67, 18%

Non-Technical Errors, 100, 27%

Technical and Non-Technical Errors, 25,

7%

Threats Errors



slightly higher rate than technical errors (figure 4.5), though there was little difference

between high risk and low risk operations.

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

Cultural /Organisational

Threats

Patient Threats Task Threats EnvironmentalThreats

Th

reat

Rat

e (T

hre

ats/

min

)

Low Risk

High Risk

Figure 4.4: Threat rate, threat type, and operative risk. This figure displays the mean rate of each threat type over the 24 operations, grouped into high and low risk categories. The vertical bars indicate the mean ± one standard deviation.

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

Technical Errors Non-technical Errors

Err

or

Rat

e (E

rro

rs /

Min

) Low RiskHigh Risk

Figure 4.5: Error rate, error type, and operative risk. This figure displays the mean rate of each error type over the 24 operations, grouped into high and low risk categories. The vertical bars indicate the mean ± one standard deviation.



4.4 Discussion

In this study, minor failures that resulted in small increases to the duration or difficulty of

the operation, risks to the patient, or demands for resources were frequent and showed

considerable variability in type and number between operations. Co-ordination and

communication problems, equipment problems, elements of safety culture, patient-

related problems and perfusion-related problems were most frequent, with a smaller

number of skill, knowledge and decision making failures. They derived from many

different sources, were not considered to be important enough to warrant recording in

any incident reporting or post-operative review mechanism, and did not appear to affect

patient outcome. All the healthcare professionals studied were highly skilled and suitable

for the roles they filled. Since there was considerable variation in the composition of the

surgical teams across the 24 operations, these failures largely reflect elements of the

system, rather than the unique failings of individuals. By examining these events in more

detail than has been attempted before, it is possible to suggest ways in which threats

predispose human errors, and how these can cascade to form more serious events.

Longer and more risky operations were likely to experience a greater number of minor

failures than shorter and lower risk operations. Increasing demands on the team

increases the chance for human error6, though as longer operations are usually higher

risk, they may feature more failures simply because there is more time and opportunity

for them to appear. More interestingly, operations where a high number of minor failures

were observed also tended to contain a major failure. While it is possible that this arises

because the observer was more likely to note minor failures in those operations

containing major failures, it is the composition of the major failures that is most

important. One major failure (chapter 6, table 6.1, item 6) occurred as a result of an

unusual error outside theatre, but the remaining six major failures contained commonly

occurring minor failures. The most serious major failure (chapter 6, section 6.6 and table

6.1 item 5), a potentially catastrophic bleed, was triggered by a single event that is an

acknowledged potential complication with that particular procedure58 but was then

compounded by a subsequent sequence of common minor failures occurring under

considerable time pressure, including the only observed decision making error in the

study, absences, team conflicts, and equipment problems. The major failure that

otherwise had the most impact on the patient interoperatively (chapter 6, section 6.2 and

table 6.1, item 1), arose through a window of opportunity that presented itself two hours



before the sequence of events themselves. The patient anatomy, which required an

unusual surgical approach, resulted in more bleeding than normal in the post-bypass

period. A failure in the management of the bleeding, which featured absence, vigilance

and expertise failures, allowed a swab to prevent blood flow to the heart, and lead to an

increasingly unstable patient. Despite excellent co-ordination, the team found it difficult

to understand what was happening because the situation was not a recognised

complication. In both cases the team mitigated the effects of the failures and prevented

adverse patient outcomes, but during the analysis of these data it became apparent that

a patient not involved in this study had experienced a serious event in theatre that had

adversely affected outcome. Discussions with the surgeon revealed that three of the five

failures observed most frequently in our studies (equipment configuration, equipment

failure, and co-ordination and communication) directly contributed to this event. Though

the resultant complication was specific to that particular type of operation, the causative

errors were far from unique to the case, to the particular type of operation, or to the

group of individuals involved on that day. It was the accumulation of errors predisposed

by properties of the system, rather than the isolated failings of individuals, that had led to

an undesirable outcome.

Task threats were the most salient source of failure in spite of the fact that only one of

the most frequently occurring major failure types was associated with task threats. This

illustrates the importance of adopting a threat and error analysis methodology which

assesses the sources of intraoperative failure, rather than just the failures. Task threats

underlie many different types of minor failure and were often co-incident with patient

threats. Patient threats never occurred without a corresponding task threat because

unexpectedly difficult or unusual patients require the completion of extra tasks which

introduce their own additional risks. Surgical teams must frequently trade transient

increases in risk for benefits it the longer term. Task threats reflect this option to increase

risk, and the observed failures are where the team had not directly compensated for the

corresponding increase in difficulty. Cannula selection is an excellent example of this,

where a wider aortic cannula bore will help perfusion during CPB, but will increase the

difficulty of inserting the cannula into the aorta, providing a task threat during

cannulation. Greater understanding of failures in risk/benefit trade-offs will help to

enhance the certainty of these decisions, and offer ongoing improvements to medical

practice beyond the traditional specialist technical developments59. In contrast to task



and patient threats, environmental threats only relate to a small number of frequently

recurring failures. Addressing these failures through improvements in design and

maintenance of equipment may therefore be especially worthwhile60-62, even before the

development of technologies that support the surgical process63;64. Finally, organisational

and cultural threats were generally independent, being unrelated to, but at least as

frequent as, other threat types. These threats, which were generally related to discipline

or nosocomial infection risks, could be addressed through improved management

practices, for example by ensuring that masks are worn at all times in the operating

theatre, and that team members need not be absent from theatre during an operation

(see Kennedy16, p. 193).

Given that the individuals studied were generally extremely experienced and competent,

the technical errors observed here are not expected to be the result of technical

deficiencies in any specialty. Certainly the management of trainees in the operating

theatre might warrant further consideration, but there is also an argument for partial

cross-training of specialist skills in the operating theatre65-68, particularly for improved

management of perfusion. Perfusion can be seen as the sole responsibility of the

perfusionist, but since it affects, and is dependent upon with the actions of the surgeon

and the anaesthetist13, it needs to be considered across specialties16. Improving

understanding between specialties could reduce the perfusion-related failures and may

also reduce blame and improve responses when things go wrong69. A breakdown in this

relationship was directly responsible for two major failures (chapter 6, sections 6.4 and

6.5; table 6.1 items 3 and 4). Clearly, shared tasks are also dependent upon non-

technical skills, which accounted for the majority of the human errors observed here.

These are examined in detail in section 7.

The evidence offered through this study supports the application of a threat and error

assessment approach to the prospective identification of threats to patient safety in

surgery. Adverse events in surgery are likely to be associated with a number of recurring

and prospectively identifiable co-incident and cumulative human errors predisposed by

threats residing in the system, rather than individual incompetence or negligence.



4.5 Summary

• In 24 operations, 366 minor failures and seven major failures were identified. The

minor failures were of 29 different types, with communication and co-ordination,

absences from theatre, and equipment problems being noted as most frequent. Task

threats were the most salient sources of failure, and non-technical errors were

observed more frequently than technical errors.

• Commonly recurring minor failures in operating theatres are frequent, usually

tolerated, and almost never reported.

• They are prospectively identifiable, more common in higher risk operations, and can

be associated with human errors predisposed by threats residing in the system,

• Adverse events in surgery are likely to be associated with a co-incidental

accumulation of number of these minor recurring failures, rather than individual

incompetence or negligence.

• Improved training in non-technical skills might be used to reduce the incidence and

impact of errors in cardiac surgery.



5. Systems Failures in Orthopaedic Surgery

5.1 Introduction

Following completion of the work in paediatrics, our studies focused on elective

orthopaedic operations. This formed a useful contrast with our previous work because it

is a highly proceduralised, relatively invariant, and lower risk form of surgery. Table 5.1

gives the key phases in a TKR operation. An estimate based on current failure rates4;39

and projected surgical volume70 suggests that in 2010, between 2000 and 9000 patients

will experience some form of adverse event following a TKR. A smaller number of TKR

revisions were also examined which are less frequent, more complex, more

unpredictable, and required a larger array of instruments than a primary TKR. Since TKR

revision operations are also longer and of higher risk71 this provided the possibility of

examining similar risk/duration effects as had been found previously. The previously

defined minor failure and failure source model methodologies could be employed here,

and the lower technical complexity of this type of surgery also meant that dual observers

could be more usefully employed, which allowed inter-observer assessments.

The purpose of this work was therefore to test the application to orthopaedic surgery of

the previously defined prospective failure evaluation methodology. We aimed to measure

the frequency and types of failures encountered, to examine whether a similar

relationship between risk, minor and major failure could be found in orthopaedics as had

been found in paediatric cardiac surgery. Finally we planned to use this to identify

systems solutions that can lead to improvements in patient safety in orthopaedic surgery.

Key Phases in the Total Knee Replacement Operation The Total Knee Replacement operation replaces damaged and painful knee joints with a completely artificial joint prosthesis

1. The patient is anaesthetised in the anaesthetic room which adjoins the operating theatre.

2. The patient is transferred to operating table in the operating theatre.

3. The treatment site is cleaned, and a tourniquet is applied to the thigh to be treated.

4. The first incision is made, and carried to the knee capsule.

5. The knee is dislocated, and an intramedullary rod inserted to fix the femoral cutting block.

6. The femur is cut distally with an oscillating saw. Following appropriate sizing, a second cutting block is used to make anterior posterior and chamfer cuts.

7. The tibial cut is made following alignment with the lower leg, and configuration of the tibial cutting block.



8. Trial tibial and femoral prostheses are used to test the fit and confirm sizing of the implant.

9. The patella is resurfaced if required.

10. Cement is prepared and applied to tibial and femoral prosthetic components, which are firmly seated. The leg is again checked for fit.

11. Once the cement has cured, the wound is washed and closed.

12. The tourniquet is removed, tourniquet time is recorded, and the patient is transferred to the recovery suite.

Table 5.1: Total knee replacement s urgery. A brief description, in simple terms, of a total knee replacement operation.

5.2 Data Collection

The general methods have been described previously. A total of 33 patients agreed to

participate in the study, of whom 20 were studied in a single operating theatre at a UK

teaching hospital. Observer one, present for all 20 cases, was a human factors

practitioner, experienced in observational methods, human performance measurement,

and in the evaluation of non-technical skills. Observer two, present for 15 of those cases,

was a researcher with prior experience of working in operating theatres, and with some

prior knowledge of non-technical skills. Though preparations were made to study both

TKR and THR operations, the case load of the sole participating surgeon meant our

studies focused almost exclusively on the TKR and TKR revisions.

One of three risk levels were assigned to each procedure, based on the complexity of

the operation. Level 1 risk was used for arthroscopies, level 2 for primary TKR or THR

procedures, and level 3 for TKR revisions. One operation was a TKR revision, but was

classed at risk level 2 as it involved only the removal of the existing prosthesis. One

operation was a primary knee replacement, but was classed as a level 3 risk as it

required revision instruments and prostheses.

5.3 Results

Overview and Case Mix

No operation studied was considered by the surgical team to be outside normal system

function, in terms of the individuals who were involved, the procedures that were

planned, or the condition of the patient. The first surgeon was always a consultant or his

specialist registrar. Fourteen of the 20 operations featured the same anaesthetist. The

scrub nurse and circulating nurses came from a pool of four individuals who regularly



interchanged roles. This meant that while the team composition was not always identical,

operations usually featured the same individuals, often in the same roles. Thirteen

operations were TKR, five were TKR revisions, one was a bilateral arthroscopy, and one

was a THR. Prosthetic implants from a range of manufacturers were studied. Mean

operative duration was 108 mins, with a standard deviation of 45 mins. Average

tourniquet time was 114 mins, with a standard deviation of 32 mins (n=18, since the

arthroscopy and total hip replacement did not utilise a tourniquet). The patients ranged

from sixty to eighty four years of age. All operations were successful, and no observed

failure was deemed worthy of further investigation either by the individuals involved, or

by the hospital.

Figure 5.1 shows the distribution of the 421 minor failures identified in the study by the

two observers. These varied between 1 and 50 in each operation, with a mean of 21.0

and a standard deviation of 13.1. One operation was considered to have a major failure

sequence, where the tibial cut was misaligned, which was identified and resolved by

making further tibial cuts. This was considered a major failure because it demonstrated a

major process breakdown, and resulted in the removal of more bone than intended. This

event was also indirectly precipitated by a number of other issues, including uncertainly,

high workload, task requirements, non-technical errors, and the introduction of new

technology into the operating theatre. There appears to be an effect of operative risk on

the number of failures encountered in each operation, with three exceptions; the hip

replacement operation, the training case that included both an inexperienced assistant

surgeon and an inexperienced scrub nurse, which both demonstrated more failures than

would otherwise be encountered, and the level 3 risk operation that demonstrated a far

lower number of failures than the other three high risk operations.

Focusing on operations which featured dual observers, and discarding the single total

hip replacement, since it was an unusual operation for this team and was unrelated to

the rest of the sample, there was a mean of 23.3 failures per operation, with a standard

deviation of 13.6. An independent samples t-test, assuming unequal variances, showed

a significant difference between minor failures in risk level 2 and risk level 3 groups (t = -

6.53 , df = 3.6, p<0.005). Using Spearman’s rho correlations, minor failures also showed

a moderate relationship with operative duration (rho = 0.678, n = 14, p<0.01) and

tourniquet time (rho=0.788, n=14, p<0.005).



Figure 5.1: Minor and major failures by operation and risk level. Operations are organised on the y-axis by risk level. Shaded areas indicate those minor failures sequences that were implicated in major failures.

Dual Observations

The 17 time markers for each total knee replacement operation allowed an assessment

of the level of understanding that the observers had of the basic stages of the operation.

Correlation between time markers in the 10 dual-observed total knee replacement

operations was almost perfect (r = 0.998, n = 151, p<0.001). Though all seventeen times

were not always recorded in each operation, this suggests that both observers were in

close agreement about when each stage of the operation occurred.

Of the 361 different failures recorded in the dual-observer operations, 77 were recorded

by both observers, 206 were recorded by observer one only, and 78 were recorded by

observer two only. Figure 5.2 shows these data by operation. Actual overlap between

observations was often low, accounting for between 0% and 42% of the total number of

failures observed in each operation. A Bland-Altman plot (72) was used to examine inter-

observer agreement, and can be found in figure 5.3. Kendall’s tau-b correlation between

ratio and mean data (tau = -0.069, n = 15, p = 0.727) suggests that there was no

relationship between measurement error between observers and the magnitude of

errors. Observer 1, the experienced human factors practitioner, generally saw the same

0

10

20

30

40

50

60

Min

or

Fai

lure

s

Risk level 3Risk level 2Risk

level 1

Minor failures directly implicated in major failure

Hip replacement

“Training” Case

Computer Navigated Procedure

Single observer only



or twice as many failures as observer 2, who had minimal human factors training. There

were three outliers, where the number of failures noted by observer 1 was considerably

more than the number noted by observer 2. One was the computer navigated operation,

which was a technique unfamiliar to observer 2. The other two outliers appear to show a

floor effect, where the number of intraoperative failures was generally small. One outlier

was found in the operation with the lowest number of failures, where observer 2 did not

see any failures independently, while in the other outlying operation there was no

overlap between observations, suggesting that both observers had a different focus of

attention. This could be expected in operations where the number of intraoperative

failures was small. Though both observers viewed events differently, and both under-

reported the total number of observable intraoperative failures, both demonstrated a

good understanding of the process, and the observations from both remained similarly

proportional across the range of operations viewed. Consequently, judgement between

observers about the success of each operation was generally consistent.

Figure 5.2: Minor failures in each dual-observed operation by risk level and observers. Operations are organised on the y-axis by risk level, and then by increasing numbers of minor failures.

0

10

20

30

40

50

60

Min

or

failu

res

per

op

erat

ion

Observer 2 only

Both Observers

Observer 1 only

Risk Level 2Risk Level

3

Hip replacement

“Training” Case




0

2

4

6

8

10

12

0 5 10 15 20 25 30 35Mean of Observed Failures [(Observer 1 + Observer 2)/2]

Rat

io o

f O

bser

ved

Failu

res

[Obs

erve

r 1/

Obs

erve

r 2]

Figure 5.3: Bland-Altman plot for assessing measurement agreement between minor failure observations.

Minor Failure Types

Utilising data from all 20 operations again, the 421 failures were classed into 20 failure

types. The mean number of each type per operation is shown in Figure 5.4.

Approximately half the minor failures observed were distributed among three failure

types, reflecting either aspects of theatre culture and environment, or the more difficult

elements of the equipment and procedure. Distractions were the most numerous failure,

accounting for nearly twice the number of the next most frequently observed failure. The

greatest distracters were communication devices, including 16 wrong numbers or calls

for people not in the theatre, 29 other non-case related telephone calls (case-related

calls were excluded from the data), 12 mobile phone calls and 15 instances of

bleepers/pagers sounding in theatre. The demands for intraoperative teaching, as well

as the presence of newspapers and other non-case related reading material during the

operation also contributed to this high level of distraction. Several instances were

observed where these distractions coincided with parts of the procedure particularly

susceptible to interference, such as equipment counts and the mixing of bone cement.




0 1 2 3 4 5 6 7 8 9 10

Decision-related surgical error

Pre-operative diagnosis failure

Unintended effects on patient

Resource management failure

Psychomotor-related surgical error

Team Conflict

External resource failure

Planning failure

Psychomotor Error (general)

Absence

Vigilance / awareness failure

Equipment failure

Patient-sourced procedural difficulties

Equipment Configuration failure

Procedure-related Error

Expertise / skill failure

Co-ordination / communication

Safety consciousness

Equipment / workspace management failure

Distraction

Mean number per operation

Figure 5.4: Mean rate per operation of minor failure types. Minor failure types are organised with the most frequent at the top, and least frequent at the bottom. Error bars show mean plus one standard deviation.

The demands to organise, prepare, and exchange different pieces of equipment (jigs,

drill bits, saw blades, cutting blocks, implants) during the operation is reflected in the

high number of equipment and workspace management failures where equipment was

needed and was not immediately available (such as finding the drill not plugged into the

air line when attempting to use it), or where it was placed in a physical configuration that

compromised the operation of the equipment (such as tangled diathermy wires). There

were also more than 40 instances suggesting reduced safety consciousness, mostly

relating to mask discipline – failures to wear masks appropriately at all times – in theatre.

Other failures classed under this type included failure of the surgical team to wear eye

protection, inconsistent glove-changing protocols, and one event where the scrub nurse



left the theatre and returned while still scrubbed. Co-ordination and communication

problems were frequent, suggesting problems either in the distribution of important

information within the surgical team, or to the timely execution of procedural sequences

requiring interaction between individuals. Procedure-related errors, expertise / skill

failures, and equipment configuration failures also reflect the close-coupling of

procedures, equipment, and expertise in this type of surgery. In particular, less

experienced surgeons and scrub nurses suffered expertise or skill failures because of

tightly coupled tasks, especially in the more demanding TKR revision surgery.

Only three minor failures are found with less than 21 but more than 10 observations,

relating to failures in equipment or vigilance and awareness, and absences that had an

impact on the operation. With 10 or less observations, technical skill based issues

predominate – decision-related error, diagnosis failures, an psychomotor errors – along

with more undesirable elements of the task, notably unintended effects on the patient,

and team conflicts.

Threat and Error Profile

From the failure source model it was possible to generate a profile of the systemic

sources of threat and error. Figure 5.5, left panel, shows the 421 minor failures grouped

by threat type. Cultural and organisational threats were the most frequently encountered.

Patient threats always appeared in conjunction with task threats, by definition (table 3).

Approximately a quarter of the failures appeared to be unforced errors, so had no threat

associated with them. Figure 5.5, right panel, shows the minor failures in terms of human

errors. The most numerous were non-technical, and only a single failure showed

evidence of both technical and non-technical error. Half the failures did not feature any

observable error by the team.

Using this information, the final stage examined the dual-observed threat and errors per

operation for risk levels 2 and 3. Though the number of threats and errors per operation

is far higher in the level 3 risk operations when compared with level 2 risk operations,

this difference is negated when operative duration is taken into account. This data is

shown in figures 5.6 and 5.7, and suggests that there is little difference between level 2

and level 3 risk operations in terms of threat and error rates. Though the rate at which



Technical and Non-Technical Error, 1, 0%

Technical Error, 59, 14%

Non-Technical Error, 150, 36%

No Error, 211, 50%

Environmental and Cultural /

Organisational, 7, 2%

Patient and Task, 23, 5%

No threat, 96, 23%

Environmental, 40, 10%

Task and Environmental,

60, 14%

Task, 32, 8%

Cultural / Organisational,

163, 38%

0.00

0.02

0.04

0.06

0.080.10

0.12

0.14

0.16

0.18

0.20

Cultural /Organisational

Threats

Patient Threats Task Threats EnvirnomentalThreats

Mea

n th

reat

rat

e (t

hrea

ts/m

in)

Risk Level 2Risk Level 3

organisational and cultural threats occur seems to be higher in level 3 risk operations,

there was no statistically significant difference.

Figure 5.5: Sources of threat and error. This shows the numbers and proportions of threats and errors identified from the 421 failures using the failure source model. The left panel shows threats, the right panel shows errors. Figure 5.6: Threat rate, threat type, and operative risk. This figure displays the mean rate of each threat type over the 15 dual observed operations, grouped into level 2 and 3 risk categories. The vertical bars indicate the mean ± one standard deviation.

Threats Errors



0.000.020.040.060.080.100.120.140.160.180.20

Technical Error Non-technical Error

Mea

n e

rro

r ra

te (e

rro

rs/m

in) Risk Level 2

Risk Level 3

Figure 5.7: Error rate, error type, and operative risk. This displays the mean rate of each error type over the 15 dual observed operations, grouped into high and low risk categories. The vertical bars indicate the mean ± one standard deviation.

5.4 Discussion

This study provided further evidence of the value of expert observers in identifying

threats to patient safety. While each individual observer did not observe all intraoperative

failures, there was general consensus between the observers about the success of each

operation. Both were able to identify a considerable number of failures, with the small

overlap between observations suggesting differences in experience and focus of

attention between observers. As in found in paediatric cardiac surgery, intraoperative

events in orthopaedic surgery that resulted in small increases to the duration or difficulty

of the operation, risks to the patient, or demands for resources were frequent, and varied

considerably between operations. Longer, higher risk operations were more likely to

include a greater number of errors than shorter, lower risk operations. This confirmed the

findings in paediatric cardiac surgery. Distractions, equipment management problems,

safety consciousness and co-ordination and communication problems were the most

frequently occurring types of failure. Decision-making and diagnostic errors were the

least frequent type of failure, reflecting the skill and competence with which the

individuals studied carried out the technical demands of their tasks. Several team

members suggested that operations were more frequently problematic with other

surgeons, and it was also apparent that the operative sequence was less smooth when

key individuals were temporarily replaced with less experienced team members.



Consequently, we suspect that the measures of failure presented here are likely to

represent a conservative estimate of the likely incidence of intraoperative errors in

orthopaedic surgery across the UK.

The high incidence of distracters suggests that both the opportunity for and tolerance of

external interference was high. Instances of reduced safety consciousness were also

high, which was surprising considering the well recognised and highly undesirable effect

that nosocomial infection can have on the outcome of joint replacement surgery73.

Though equivocal evidence has been offered for the effectiveness of mask discipline for

control of nosocomial infection74;75 protocols should be applied consistently and

particular attention should be paid to prosthetic implantation76. Controlling the use of the

telephone in theatre and reinforcing safety procedures are simple and direct solutions

that could and should be applied. On a number of occasions senior team members

attempted to maintain standards with little success, and in one instance an individual not

involved with the operation entered the theatre during the procedure, initially without a

mask, made a loud and argumentative phone call without the permission of anyone in

the operating team, then left theatre violating mask protocol on the way out. Failures to

control distraction and safety are therefore symptomatic of the impact of culture on

surgical quality77, and rather than blaming individuals, the aetiology of this problem

should be considered at all levels, from hospital management, through professional

societies and associations, to national policy.

Unlike other types of surgery, which may be reliant on the scalpel-and-suture skills of the

surgeon, TKR and TKR revision operations are reliant on the appropriate use of

procedure-specific instruments. Achieving the correct equipment configuration is a key

technical skill for this type of surgery. Though the instruments are usually reliable, the

wide range of equipment required in a short space of time placed pressure on the

operating team. For example, the 10 minute period from the first incision to completion of

the femoral cuts for a PFC implant requires the use of five specialist instruments

(intramedullary rod, femoral locating device, distal femoral cutting block, femoral sizing

guide, and A/P chamfer cutting block), as well as a range of other incidental equipment

such as drills, saws, retractors, pins, hammers, scalpels, swabs, diathermy and suction.

A surgeon cannot perform the operation without the scrub nurse to provide the

appropriate instruments, and a scrub nurse cannot support the surgeon effectively if they



are overloaded, distracted or unable to keep up with and anticipate the surgical tasks.

Given the skill required for both roles, the mutual trust required, and the authority

gradient between them, this relationship is brittle. It was encouraging that in the present

study the interaction between surgeons and scrub nurses was mutually supportive

almost without exception, and remained solution-focused when things did not go

according to plan. In the more demanding TKR revisions the active participation of the

prosthetic manufacturer sales representative helped the theatre and surgical staff to

appropriately manipulate the more complex and less familiar instrumentation, as well as

providing an additional resource for aiding co-ordination and error capture. However,

while this can have exceptional safety benefits, it can also have negative effects on the

rest of the team. Moreover, this reflects a deficiency in the design of the instruments

which seem too complex for surgical teams to use without a considerable amount of

additional, manufacturer-specific training. It also raises ethical concerns. Overall, though

patient-related difficulties were frequent, the success of the operation was more

dependent upon the design and management of the equipment and the co-ordination of

the team.

The most frequent failures may not always demonstrate the greatest threat to operative

success or system function, but they can provide the environment in which major failures

and adverse events are more likely. Examination of the fundamental properties of the

system that these minor failure types display allows the identification of a small number

of error reduction strategies that address the problem at the source. This is more

advantageous than providing defences to a large number of unique deficiencies, since

safety systems themselves can become brittle, ineffective and add to the problems in the

system if too many defences are implemented6. The failure source model provided a

diagnosis of the likely source of threats and errors in this operating theatre and in this

type of surgery. By taking a prospective approach to the observation of system failures

and human error in orthopaedic surgery, this work identified problems with the

organisation and culture, problems with the design and organisation of the equipment

which were amplified in more difficult operations, and the potential for improved non-

technical performance in the operating team.



5.5 Summary

• In 20 cases, 421 minor failures and one major failure were observed. The most

frequent types of minor failure were distractions, equipment management failures,

safety consciousness failures, and co-ordination and communication failures. The

most frequently observed sources of threat were cultural and organisational

problems, and non-technical errors were more frequent than technical errors.

• As in cardiac surgery, minor failures are frequent, and usually tolerated or ignored.

• Sequences of commonly occurring minor failures accumulate in orthopaedic

operations much more rarely, and are of considerably less risk when they do.

• The effect of operative risk on the number of major failures may be apparent in this

data set, but the small number of high risk operations limit the confidence which can

be applied to this association.



6. Major Failure Vignettes

6.1 Introduction

Eight major events were identified during the course of the 45 operations. Here, they are

presented in narrative vignette form, with a brief commentry supplied after each vignette

to add further quality to the observations. The analysis technique that was applied has

been illustreated in section 6.10 to demonstrate the methodology adopted to assess the

constituent minor failures.

6.2 Mastoid strip causes compression of right coronary artery.

A 7 day old male with transposition of the great arteries and large ventricular and atrial septal defects was anaesthetised, placed on cardio-pulmonary bypass and cooled to 28ºC. Two hours and seven minutes into the arterial switch operation, the surgeon noted that the anatomy of the patient did not allow the LeCompte manoeuvre, where the pulmonary artery is passed behind the aorta. Instead, the surgeon chose to bisect and re-anastomose the median pulmonary artery at the bifurcation, knowing this would be a suitable alternative strategy, but would lead to extra bleeding. While preparing to come off cardio-pulmonary bypass, the surgical team noted that ventilation was difficult, due to pooling of blood in the lungs, requiring the anaesthetist to balance the oxygen saturation levels using manual control of ventilation. It was not possible to obtain a good visualisation of the surgical repair using an epicardial echocardiogram, and the ventilation problems prevented a transoesophageal echocardiogram. Neither ventilation or echocardiogram problems were unusual, and the surgeon was satisfied with the success of the procedure. The surgical team weaned the patient from CPB, and expecting a quick resolution to the case, the anaesthetist asked for the next-patient to be prepared for theatre. Four hours and six minutes after the start of the operation, the consultant surgeon left his experienced assistant surgeon to control the bleeding and close. The delivery of platelets from the blood bank was delayed due to a shortage of porters, so the expected bleeding was managed through the use of swabs and mastoid strips. The assistant surgeon placed an extra suture in the SVC to address one source of bleeding. Five hours and one minute after the start of the operation, the team observed that oxygen saturation was falling, and that the heart was in and out of sinus rhythm and looking purple. The consultant surgeon returned, and pacing and defibriallation were used to control the cardiac rhythm. Four minutes later, the oxygen saturation was observed to be at 61%, and from diagnostic discussions it became clear that the team did not understand what was happening. A second echocardiogram and nitric oxide ventilator were ordered. Upon inspection of the right coronary artery, the assistant surgeon did not see anything problematic, and he and the consultant decided that the extra suture may be the cause of the problem. As ETCO2 fell and the heart continued to worsen, the anaesthetist urgently prepared a dose of heparin that would have allowed the patient to be quickly placed back on bypass. During this period, the temperature of the patient rose to 37ºC. Finally, thirteen minutes after the problem was first noticed, the surgeon inspected the region of the right coronary artery and removed a small section of mastoid strip. Immediately oxygen saturation and ETCO2 rose, to the relief of the surgical team. Delaying sternal closure, and patient was received into the intensive care unit forty-five minutes later. The second patient was cancelled, and re-scheduled for the following day, which was a Saturday.



Minor difficulties within an operation accumulated over the perioperative period to create

an unstable situation. The anatomy of the patient required an unusual approach that

resulted in additional bleeding. Two hours later, the management of the bleeding that

was required, exacerbated by the lack of available blood products and changes in

surgeons at the table, created the window of opportunity for the subsequent events.

Since this type of complication is not well documented, the surgical team had

considerable difficulty in identifying the source of the problem even though they quickly

identified the right coronary artery as being a likely source. The uncertainty resulting

from the sutures placed by specialist registrar and the difficulty in correctly assessing the

surgical repair due to echocardiogram and ventilation problems, compounded this

diagnostic difficulty.

6.3 Omission of surgical step.

Often it is the unknown complications and difficulties that cause errors in surgical

procedures. In this case, the team relaxed slightly too soon after a difficult surgical

procedure. Closing the atrial septostomy would usually be done just after the removal of

the cross-clamp. However, in his haste the surgeon did not announce it in the usual way,

deviating from his usual practice and providing the window of opportunity for the error.

The teaching discussions immediately afterward served to further distract the surgeon

from his usual practice. Seemingly, none of the operating team realised the omission.

The absence of the consultant anaesthetist was critical, because usually he would

closely monitor the progress of the operation and offer support to the surgeon. However

A 12 day old male with transposition of the great arteries was found upon inspection to have a previously undiagnosed intermural coronary pattern. During the operation the team struggled with a perfusion airlock which resulted in excessive blood in the surgical field. The consultant anaesthetist was periodically called away from theatre, and his registrar, who was unfamiliar with this type of operation, made a number of errors in communication and equipment operation. The operation was therefore considerably more difficult than expected. Nearing the end of the operation, rewarming of the patient began. Since the operation had been particularly difficult, the surgeon was keen to see whether his repair had been successful, and removed the cross-clamp without the usual verbal announcement. As the heart began to beat the consultant anaesthetist returned to theatre, and a non-operating surgical registrar entered to discuss the case with the surgical team. Six minutes later, the surgeon realised he had forgotten to close the atrial septostomy and immediately moved to halt rewaming and re-configure perfusion to allow him to complete the procedure. Ten minutes later, the septostomy was closed and the operation was completed without further incident.



in this case he was unaware that the septostomy was not closed. Cross-checking with

other members of the team and a procedural checklist would have reduced the chance

of this type of error happening. Moreover, an understanding of the increased likelihood

of human error in these difficult circumstances might have raised the awareness of the

team, and heightened their vigilance.

6.4 Ex-sanguination mitigated fortuitously

The period during and immediately following cardio-pulmonary bypass is of particularly

high workload, especially when it follows a high risk or complex set of surgical

procedures. It requires close co-ordination between the team members, their equipment,

and the procedures that need to be carried out to ensure the successful recovery of the

patient. The interaction between the surgeon, anaesthetist and perfusionist is key, and

can sometimes break down. In this case, because a target haematocrit was not properly

agreed before coming off CPB and starting MUF, the initial plan was made without

including the anaesthetist, and was then re-planned without the appropriate involvement

of the surgeons. In the flurry of communication, the surgeon did not properly

communicate the removal of the MUF line until some time after it had been done.

It is always better to plan for periods of high workload during periods of low workload. A

haematocrit plan appropriately agreed prior to coming off cardio-pulmonary bypass

would have avoided this undesirable situation.

A three day old female underwent the Norwood Stage 1 procedure for Hypoplastic Left Heart Syndrome. Towards the end of the operation, immediately after weaning from cardio-pulmonary bypass, modified ultrafiltration (MUF) began. The perfusionist and assistant surgeon then agreed that the target haematocrit level should be 40. Four minutes later, upon reaching a haematocrit of 40, the perfusionist announced that he had finished the MUF. Thinking that a higher haematocrit might be achievable, the anaesthetist asked the perfusionist to continue the MUF. The assistant surgeon was involved in making the new plan, but thought the decision was to concentrate the blood in the bypass reservoir, not to continue the MUF. Meanwhile, the surgeon had taken the MUF line out of the atrial appendage. As filtering continued, blood was taken from the patient via the aortic cannula, and passed back down the MUF line which should have been placed in the atrial appendage, but had now been removed, slowly exsanguinating the patient. The operating team then concentrated on the bloodgas readings and control of blood acidity. Two minutes after removing the atrial appendage line, the surgeon mentioned to the perfusionist that he had done so, and the error was spotted. Fortunately, only a small amount of blood volume had been removed and a new plan was correctly co-ordinated. The problem was resolved quickly.





6.5 Breakdown in team co-ordination

Management of the patient immediately after cardio-pulmonary bypass requires

substantial co-ordination between the consultant surgeon, the consultant anaesthetist,

and the perfusionist. In this case it completely broke down. The surgeon did not co-

ordinate the initial wean from bypass with the anaesthetist, and as a result both gave

conflicting commands to the perfusionist. The perfusionist did not sufficiently indicate the

difficulties he was encountering with heamofiltering, or the blood loss he was observing.

The anaesthetist ignored important observations made by the perfusionist and blamed

him and the blood bank for the difficulties with the patient.

All surgical procedures require close co-ordination between a number of individuals. The

consultants are always seen as the leaders of an operating team, but they must listen

and encourage the participation of other members of the team, especially when those

other members are experienced specialists. Furthermore, where two consultants are

A two day old male underwent the Norwood Stage 1 procedure for Hypoplastic Left Heart Syndrome. Following successful completion of the major surgical tasks, the surgeon made the decision to wean from bypass, even though the anaesthetist felt that the patient was not ready . Once off bypass, the surgeon asked the perfusionist to raise the CVP, while the anaesthetist asked the perfusionist to keep the CVP stable, but filter and transfuse the blood. Eight minutes after coming off bypass, the perfusionist reached the previously agreed haematocrit level of 30 and announced that he had finished filtering. The anaesthetist was surprised, and approximately two minutes later asked the anaesthetic registrar for the haematocrit level, which was found to be 29. Six minutes later, the CVP had fallen to 8mmHg and temperaturehad fallen to 33ºC, so the anaesthetist switched on the heating blanket and asked the perfusionist to raise the CVP to 10mmHg. Ordered platelets were taking longer than expected to arrive from the blood bank. The perfusionist indicated to the anaesthetist that a lot of blood was being lost, but this was not heard. The surgeon then asked the perfusionist to transfuse more blood, but the anaesthetist indicated that no more blood should be given, and discussed the possibility of further heamofiltering. The surgeon then asked the anaesthetist if the wean from bypass was too soon, but the anaesthetist indicated that he believed the perfusionist was to blame for the instability of the patient. Twenty four minutes after coming off bypass, the anaesthetist and surgeon discussed the reasons for the low haematocrit. The surgeon then asked if there is a lot of bleeding and the perfusionist replied that 10mls/minute were being lost, and that he had said this some time ago. The surgeon idenitified the bleeding site and placed an extra suture. With the patient stabilised, protamine was given. Forty two minutes after coming off bypass, heamatocrit was found to be 40, and the platelets arrived. The patient was received into the Intensive Care Unit forty minutes later.



required to work closely together, they must ensure that they co-ordinate in a way that

ensures the combined experience is shared and used in the most effective way.

6.6 Aortic homograft ruptured during sternotomy.

Once catastrophic bleeding has started, time becomes a limiting factor. The absence of

consultants, the awareness failure by the surgeon and lack of blood products all

elongated the response time to the initial error. Though clearly the assistant surgeon

made the initial error by rupturing the homograft, both the patient, who had particularly

adherent arteries and the procedure, for which this is a known complication, predisposed

that error. His subsequent decision, made with the approval of the consultant

anaesthetist, to overrule the wishes of the consultant surgeon, was critical to the

eventual outcome. Fortunately, the team worked exceptionally well together once the

management of the patient had begun, appropriately managing the resources available

and prioritising their responses to the situation. It is difficult to know how the bleeding

A six month old male underwent the Norwood Stage 2 procedure for Hypoplastic Left Heart Syndrome. As the consultant anaesthetist prepared for the operation, he realised only two units of blood had been ordered and left the anaesthetic registrar in theatre while he went to arrange for more. During the initial sternotomy, conducted by the experienced assistant surgeon, the heart and great arteries were found to be adherent to the sternum, and an unusual bleeding site was noted. Placing pressure on the bleed, the assistant surgeon asked for the consultant anaesthetist and consultant surgeon to be called, since neither were present in the operating theatre. Both consultants arrived within two minutes, but in that time bleeding had increased severely. The assistant surgeon asked for heparin to be given to the patient, which would allow rapid initiation of cardio-pulmonary bypass and re-use of the blood being lost. The consultant surgeon said no to heparin, and went to scrub up. By this time, heart rate, arterial pressure, CVP, oxygen saturation and ETCO2 had all fallen drastically. Seeing the bleeding get worse, the assistant surgeon again asked the consultant anaesthetist for the heparin, and the anaesthetist agreed. Looking up from scrubbing, the surgeon stated that he had asked for no heparin. However, on viewing the anaesthetic monitor and the surgical field, he realised the seriousness of the situation. The aortic homograft from the previous surgery had been ruptured during the sternotomy. Vigorous cardiac massage was given, and with considerable difficulty an aortic cannula was placed. Adrenaline and blood volume were given, but twenty minutes after the identification of the bleed defibrillation was required. Fortunately, sinus rhythm was regained, and six minuteslater the heart rate began to rise again. The patient was placed on cardio-pulmonary bypass 34 minutes after the initial bleeding, by which time the heart had been bradycardic for 22mins, the blood had a pH of 6.9, and haematocrit was at 19. There were a number of further difficulties in re-configuring the hastily initiated bypass for the surgical intervention, but the surgery was successfully completed and the patient received into ICU after a 279 minute operation. A post-operative neurological examination showed no adverse effects from the incident.



A five day old male underwent the Norwood Stage 1 procedure for Hypoplastic Left Heart Syndrome. The patient was anaesthetised and the surgeon began the midline sternotomy. Four minutes after the start of the operation, the donor homograft was unpacked by the scrub nurse and bathed in saline. The patient was placed on bypass and began cooling to a target temperature of 15ºC. Thirty four minutes after the start of the operation, the donor homograft was inspected by the surgeon, and found to be a section of pulmonary tissue rather than the required aortic tissue. The labelling on the packaging was incorrect. The nursing team mentioned that this was the second time in two weeks that this had happened. The surgeon decided that he could not use the pulmonary homograft, and asked for another aortic section to be sent from the supplying hospital. Cooling was halted at 22ºC, and the patient was kept stable on cardio-pulmonary bypass for 40 minutes until the new donor tissue arrived. This new homograft had been labelled correctly, and the operation was completed successfully with a cardio-pulmonary bypass time of 216 minutes.

A 7 day old female with transposition of the great arteries and pre-operatively diagnosed intramural coronary artery pattern was prepared for an arterial switch operation. Twenty two minutes after the first incision, as the surgeon prepared to cannulate for cardio-pulmonary bypass, he asked for the consultant anaesthetist to give heparin. The consultant anaesthetist replied clearly in the affirmative. Four minutes later, the consultant anaesthetist took a blood sample and placed it in the Hemochrom Activated Clotting Time (ACT) measurement device. There was a shortage of disposable cartridges for the preferred ACT machine, forcing anaesthetist to use an older, slower machine. Seven minutes later, the anaesthetist indicated to the perfusionist, who had been distracted, that the ACT measurement had finished. It read 183. Five minutes later, a second sample stopped at 191. Unsure as to why the first dose had not worked, the perfusionist and anaesthetist decided more heparin was needed. Just after giving the second dose, the anaesthetist was called away to speak on the phone outside the operating theatre. He returned three minutes later and took more blood for an third ACT reading.

could have been avoided, but with better preparedness the response would have been

more timely and effective.

6.7 Incorrectly Labelled Donor Tissue.

Surgical teams do not always have control over critical components of the operation.

Once the operation had started, there was nothing that could have been done to avoid

the incorrect homograft. However, anticipating potential problems can help surgical

teams compensate for the unavoidable problems they are faced with. Given that a

similar event had recently happened previously, pre-operative planning with the whole

team could have identified this as a potential risk. The precaution of inspecting the

homograft earlier in the operative course could have been introduced for this case, and

could have avoided the delay and the long bypass time.

6.8 ACT management difficulties in high-risk case.



The perfusionist and anaesthetist discussed the possibility that the batch of heparin was at fault, and changed to a new batch. This time the ACT stopped at 292. The surgeon, anaesthetist and perfusionist discussed the problem, and a third dose of heparin was given. The perfusionist and anaesthetist agreed to try another Heamochron machine. Again the consultant anaesthetist was called away to the phone. After waiting seven minutes, the surgeon, who was clearly impatient to start, asked whether he could cannulate. The perfusionist replied that ACT was still not above 300, then left the theatre for approximately one minute.At this point, neither the perfusionist nor the two anaesthetists were in the operating theatre. Upon returning, the surgeon again asked the perfusionist whether he could cannulate, and finding an ACT of 370, the perfusionist confirmed that cannulation could commence. The patient was successfully placed on cardio-pulmonary bypass (CPB) and the operation continued without incident for three hours. Following a successfull wean from CPB, just as the protamine was given to lower the ACT again, the left atrial pressure rose, and the heart became ischemic. The operating team identified a possible compression of the left intermural coronary artery. A dose of adrenaline was given, and the anaesthetist asked his registrar to draw up a double dose of heparin in case there was a need to rapidly re-initiate CPB. ACT following the first dose of protamine was found to be greater than 999. After a brief discussion with the surgeon, the anaesthetist gave a second dose. Again left atrial pressure rose, and again the heart became ischeamic. The surgical team then controlled and monitored the stability of the patient for 25 minutes. Though ACT was still greater than 999, the surgeon was not keen to risk ischemic incident, so decided to delay sternal closure and return the patient to ICU. The surgeon remarked that he has never sent a patient back to ICU who is still heparinised, but just before being moved from the operating theatre, a final ACT reading showed 239. Upon arrival in ICU the patient almost immediately deteriorated severely, and required a return to theatre the same evening. This second procedure was successful.

Since it is crucial to the safety of the patient that heparin is applied appropriately before

cannulation for cardio-pulmonary bypass, there are a range of safety and communication

protocols. In the present case, the absence of the perfusionist and both anaesthetists

was a compromise of these protocols at the most critical moment, unnecessarily adding

risk to a case that was already challenging. The intramural coronary artery pattern posed

a particularly difficult problem for the surgeon, and undoubtedly was the primary reason

for the post-bypass ischemia and post-operative deterioration. However, the other

intraoperative difficulties compounded that problem. It will never be clear why the ACT

management for this patient was so difficult. It may have been the batch of heparin, a

fault with the older ACT machine, or simply that the patient was resistant to heparin. The

difficulty in bringing the ACT down toward the end of the operation suggests that too

much heparin was given initially, and the eventual ACT of 239 indicates that there may

have been an unusual delay either in the response of the patient to the protamine, or the



measurement of the ACT. Whether the intramural artery pattern and the ischemic events

were related to these difficulties, or purely co-incidental, is uncertain.



6.9 Multiple uncertainty leads to teamwork and task breakdown.

It is in the interests of every surgical team to control for as much uncertainty as possible

prior to the operation. Primary knee replacement operations usually follow a predictable

and well rehearsed course. However, revision operations are far less certain, and require

more equipment. This reduces the ability of the operating team to think ahead and to be

able to respond to one another in the best way. The introduction of new equipment into

the operating theatre also adds to the uncertainty. It rarely maps directly to existing

processes, and requires new skills and knowledge to operate. Uncertainty associated

with new equipment or an uncertain plan makes an operation more difficult.

A 66 year old male was anaesthetised and prepared for a primary total knee replacement operation. The surgeon was already aware that the knee was in an advanced state of wear, and might require additional prosthetic components usually required only for second stage TKR revisions, so the corresponding instrument trays and prostheses were kept on standby. The sales representative for the prosthetic manufacturer was present for the operation, as was normal for TKR revisions, to assist in the selection and configuration of the implants. On this occasion, the surgeon also chose to use a new computer navigation device that was on trial from the prosthetic manufacturer. This was normally used for standard primary TKRoperations, and had not previously been used by the team for this type of operation. Though the surgical team were all experienced and familiar with one another, early in the operation the scrub nurse showed signs of uncertainty both about the use of the computer equipment, and the surgical direction being taken. The sales representative offered his support to the scrub nurse, but his advice conflicted with instructions from the surgeon, and the scrub nurse became confused about what was required next. At the same time, the surgeon was struggling to operate the computer navigation system, becoming increasingly frustrated with it. The sales rep also tried to support the surgeon, but was unable to support both scrub nurse and surgeon at once. He was also trying to coach a second sales representative at the same time. There was a high level of conversational noise in the operating theatre. As the operation progressed, the scrub nurse began to unpack several extra trays of instruments required for the additional implants, with help from the sales rep. While the rep was outside the theatre, the surgeon removed one of the navigation sensors, and on his return the rep expressed concern that this may have upset the calibration of the navigation system. A few minutes later, the assistant surgeon became concerned about the tibial cut that had been made. The surgeon and sales rep dismissed his concerns and instead teased him for being overly anxious. In the following minutes there were a number of other confused exchanges between the sales rep, the scrub nurse and the surgeon. Sixty three minutes after the start operation the surgeon noticed that the tibial plateau had not been cut properly and was on a slope. The operating theatre went silent. The assistant surgeon then confirmed that this was his earlier concern, and appropriate cuts are made to rectify the problem. Following a number of further co-ordination difficulties, the operation was eventually completed after 126 minutes.



In this operation, the uncertainty about the patient and course of treatment that would be

required, combined with the unfamiliarity with the computer equipment and associated

procedural requirements raised the difficulty of the operation. These difficulties increased

the workload on all the team members. The reduced ability of the team to think ahead

meant that they were only able to see or respond to present events, rather than

anticipating and planning for future events. This lack of awareness ultimately manifested

itself in a poor tibial cut. When workload increases, spare mental capacity reduces. This

may have been a contributing factor in the failure of the surgeon and sales

representative to listen, to hear and to act upon the concerns of the assistant surgeon.

This led to the late identification of the error.

6.10 From minor failures to major failures

These major failures form a cross-section of many issues apparent in theatre. Table 6.1

shows the major failure descriptions, and the associated minor failures in approximately

chronological order. The reader will note the association between the vignettes and

many of the minor failures. This provides a guide as to how the assessment

methodology was applied. However, detailed assessment will show two features; that

not all events described in the vignettes are described in terms of minor failures in the

table; and that some failures in the table are not apparent in the vignettes. This is a

reflection of both the simplification required for convenient narration of detailed

intraoperative events, and the difficulty faced by researchers in finding a stopping rule for

the aetiology of a particular failure1. In the present definition of major failures, we opted

only to select the events that formed the window for the major failure and the

subsequent recovery, even though prior contextual issues may also have had indirect

effects. Regardless of where precisely the stopping rule is applied, it is clear that critical

errors, even rare technical errors, occur in close proximity to usually innocuous failures,

and it is this accumulation that can bring about an event with adverse impact.

It seems that either a sequence of small failures can build to create something more

dangerous, or a single, more serious mistake is exacerbated by a sequence of smaller

failures. Since both can work together, the propagation of error from minor isolated

failures to major failure sequences is likely to be a combination of cumulative effect of

minor failures, coincident with some critical stage or process in the operation. This effect

is more pronounced in higher risk operations both because they are more difficult and so



naturally predispose more minor failures, and because these operations have more

critical stages. Though half of these major failures could have occurred in any single

operation, presumably it was this additional workload in the higher risk operations that

provided the window of opportunity for error.

Major Failure Associated minor failures 1 Mastoid strip causes

compression of right coronary artery

• Equipment / Workspace management • Absence • Unplanned procedural affects on patient • Equipment / Workspace management • Vigilance / Awareness failure • Expertise / Skill failure

2 Ex-sanguination during post-bypass heamofiltering.

• Planning failure • Team conflict • Co-ordination / communication failure x2

3 Omission of key surgical step • Sub-optimal diagnosis • Co-ordination / communication failure • Distraction • Absence • Procedure-related surgical error • Temperature control difficulties • Co-ordination / communication failure

4 Breakdown in team co-ordination.

• Procedure-related surgical error • Co-ordination / communication failure x2 • Procedure-related perfusion error • Team conflict x2 • External resource failure • Vigilance / Awareness failure • Temperature control difficulties

5 Aortic homograft ruptured during redo midline sternotomy.

• External resource failure • Planning failure • Patient-sourced procedural difficulties • Psychomotor-related surgical error • Absence • Decision-related Surgical Error • Team conflict • Unplanned procedural affects on patient • Cannulation difficulties • Unplanned procedural affects on patient • Team conflict • Resource management failure • Equipment / workspace management failure • Unplanned procedural affects on patient • Co-ordination / communication failure • Equipment failure x2 • Equipment configuration failure • Psychomotor-related surgical error • Equipment / workspace management failure

6 Incorrectly labelled donor tissue.

• External Resource failure • Known Problem



7 Difficult management of activated clotting time.

• Equipment failure • Vigilance / awareness failure • Patient-sourced procedural difficulties • Absence • Co-ordination / communication failure • Patient-sourced procedural difficulties • Patient-sourced procedural difficulties • Unplanned procedural affects on patient

8 Multiple uncertainty leads to teamwork and task breakdown.

• Planning failure • Equipment configuration failure x2 • Expertise / skill failure • Equipment configuration failure • Team conflict • Decision-related surgical error

Table 6.1: Minor fai lures, in approximate chronology, associated with major fai lures. Teamwork, absence, communication problems, co-ordination problems and equipment

failures can all exacerbate an already difficult and uncertain case. These are recognised

features, but apparently are rarely considered to be significantly related to performance

or outcome. They involve skills which are, at present, not formally included in training

curricula. It is apparent that the surgical teams were almost entirely unprepared for the

events that these major failures demonstrate. Two of these failures (sections 6.2 and

6.6) were briefly described in the post-operative surigcal report appended to the patients

notes, but no other record of these events was kept. There was no further attention paid

to these events, no post-operative debriefing, no assessment of why these events had

happened, and consequently nothing was learned about how these clear deficiencies in

surgical quality might be improved. Though each case is unique, the underlying

problems are common, and the major failures which result are frequent. Though the size

of the present sample limits confidence, estimates based on this sample would suggest

that 21% of paediatric cardiac surgery cases and 5% of orthopaedic cases exhibit these

types of difficulties. This would approximate to previous estimates of serious errors

associated with healthcare treatment4. Furthermore, it may suggest why the Norwood

procedures (level 6 risk) – all of which had a major failure associated with them in the

present study – is associated with a mortality of 41%48. Clearly, attention to these minor

failures could have a considerable impact on mortality and morbidity, particularly in high

risk operations. Though outcome in all 8 cases was not measurably affected, the

vignettes form a cross-section of the problems encountered by surgical teams, and

consequently the ways in which patient outcome can be affected by system threats and

errors.



6.11 Summary

Assessment of the sequence of events that lead to the 8 major failures, and the

identification of minor failures that contributed to these events, can be used to avoid

these types of failure in the future.

• Major failures were all associated with a sequence of minor failures

• The transition from minor failure to major failure appears to be from both cumulative

effect and co-incidence with a critical phase in the operation

• This effect is amplified in high risk cases because they are longer and more difficult,

and so feature more minor failures, and because they generally have more critical

phases.



7. Non-Technical Skills and Errors In Surgery

7.1 Introduction

Successful surgery is dependent upon the management of team resources, yet there is

little in the way of formal training for these skills in healthcare. Non-technical skills

describe the cognitive or mental skills (decision making; planning; situation awareness),

and social or interpersonal abilities (team-working, communication, leadership)78

possessed by individuals and displayed by a team during simulated or real events. The

value of non-technical skills as a method for the prevention of catastrophic failure has

been confirmed following the study of a wide number of systemic failures in high

technology domains, most notably aviation79. A substantial body of evidence has been

gathered in that field that has successfully distinguished between technical and non-

technical skills80, identified the types of non-technical skills that promote success or

failure, and provided behavioural marker systems to assess those skills, either during

simulation, or during normal flight81. Given that modern aviation systems can create

safety through technology, the benefits to the healthcare industry, which is still heavily

reliant on individual performance, might be even greater. It was therefore useful to

examine the application of aviation-derived training and assessment techniques in

surgery. Indeed, the ability of operating theatre teams to avoid failures, or capture them

before they can accumulate to influence outcome, may be critical7, and it was the ability

of the team to cope with the intraoperative failures described in previous chapters that

resulted in successful completion of the more problematic operations. Moreover, the

cascade of minor human and systemic failures to a major failure may also be prevented

by effective teamwork82-85. In this section we describe a re-examination of the same 20

orthopaedic and 24 paediatric cardiac cases utilising a non-technical skills assessment

technique previously developed in aviation.

The non-technical performance in a distributed process will relate both to behaviours of

individuals, and to team functions. Recent interest in matters of patient safety have led to

the development of useful behavioural marker tools for individual assessment. A

technique for non-technical skills training in anaesthetics46, was carefully developed after

the NOTECHS work in aviation49 and follows good scale design principles86, but has yet

to be used in a live operation, or validated against objective performance criteria. A



second scale for the assessment of surgeons87 was based upon data relating to errors in

surgery, but is difficult to adopt as a behavioural marker system as it does not easily lend

itself to observation. As both methods focus on only one specialty in the team, neither

considers the operating team as a whole. Consequently, both fail to address the non-

technical skills required for inter-disciplinary teamwork across individuals of different

specialties, levels of skill, experience and training88. This mix of skills and specialties,

and the requirement that all team members closely co-ordinate, is particularly noteworthy

in cardiac surgery. Early attempts to examine the team as a whole45 should be seen in

the light of more recent advances in non-technical skills evaluation49, and more recent

evidence of human factors failures in cardiac surgery7;16. The assessment of teams, not

individuals, is also consistent with the adoption of a just and blame-free approach to

error that is vital if medical errors are to be reported and understood89.

It is likely that the ability of a team to work well together has the greatest impact on

operative performance. Consequently, two methods for the analysis of non-technical

skills in operating theatre teams were examined. A range of non-technical errors had

been identified in both orthopaedic and cardiac operations through examination of minor

failures. This formed a specific, failure-only measure of non-technical skills. The second

measure, utilising the surgical NOTECHS behavioural marker system, combined both

effective and ineffective non-technical skills in a single global intraoperative measure.

The purpose of this work was threefold. Firstly, the relationship was measured between

non-technical performance and other indicators of performance. It was hypothesised that

since cardiac surgery is more reliant on shared tasks, there would be a closer

relationship between non-technical performance and technical errors, minor failures and

operative duration. Secondly, the use of non-technical error data for the diagnosis of

non-technical deficiencies was examined. Thirdly, the agreement between the two

measurement techniques, and between observers was examined. The overall aim was

to how non-technical skills contribute to patient safety, and how they might be assessed

and diagnosed in the future.



7.2 Method

The general methodology adopted has been described in chapter 3. The non-technical

error data were derived from the observations of minor failures. Non-technical errors

were identified through application of the source failure model, and then assigned to a

dimension on the NOTEHCS scale. The global non-technical skills rating was based on

scores between 1 (below standard) and 4 (exceed) carried out on four dimensions in

each of three intraoperative periods. Each operation was then ranked from best to worst

according to the number of below standard, basic standard, standard, and exceed

scores. In paediatric cardiac surgery a total of six first surgeons, nine assistant surgeons,

eight scrub nurses, eight consultant anaesthetists, nine anaesthetic registrars and four

perfusionists were studied across the 24 operations. In orthopaedic surgery, a total of

two first surgeons, five assistant surgeons, four scrub nurses, three circulating nurses,

and three circulating nurses were studied across 20 operations. Non-technical error and

skills measures were available for all operations with observer one. Non-technical error

measures were available for all 15 operations with observer two, while non-technical

skills ratings were available for 11 of those operations. The data from observer one

alone were used in the majority of the analysis, with dual data combined to examine

inter-rater reliability.

7.3 Results

Non-technical errors

Table 7.1 shows the relationship between non-technical errors and other intraoperative

parameters using spearman’s rho correlations. As would be expected given that threats,

failures and errors are all derived from the same data, via the source failure model, there

is a strong relationship between non-technical errors and both minor failures and threats

in both surgical domains. The relationship with duration is less clear, being significant in

orthopaedics, but not in paediatrics. The relationship with technical errors is far stronger

in paediatrics than in orthopaedics, despite also being derived from the failure source

model.



Rho n p

Minor Failures 0.825 <0.001

Threats 0.642 <0.005

Technical Errors 0.438 <0.05

Paediatric Cardiac [observer 1 only]

Operative Duration 0.345

24

0.099


Threats 0.724 <0.001

Technical Errors 0.263 0.263

Orthopaedic [observer 1 only]


20

<0.01

Table 7.1: Correlations between non-technical errors and other intraoperative performance parameters.

Figure 7.1: Distribution of non-technical errors along NOTECHS dimensions. The left panel shows paediatric cardiac surgery, the right panel shows orthopaedic surgery. Figure 7.1 shows the distribution of each non-technical error in the four NOTECHS

dimensions across the 24 operations studied. The most numerous deficiencies were

those in leadership and management, followed by situation awareness with problem

solving and decision making errors the least numerous. The major difference between

the two types of surgery is in problem solving and decision making errors, which are far

Leadership & Management

47, 37%

Teamw ork & Co-ordination

24, 19%

Situation Aw areness 43,

34%


12, 10%

Problem solving & Decision Making,

1, 1%

Teamwork & Co-ordination, 30,

20%

Leadership & Management, 56,

37%Situation

Awareness, 63, 42%


Orthopaedic Surgery



smaller in orthopaedics, reflecting the considerably lower group decision making

demands required during this type of surgery. Figure 7.2 shows those dimensions

divided into individual NOTECHS elements for both types of surgery, again showing

considerable similarity. The most numerous errors were associated with planning and

preparation, followed by failures to notice important elements in the environment. Only

three elements – leadership, option generation, and outcome review – did not have any

observable non-technical error associated with them. Planning and preparation,

workload management and understanding team needs were also frequently observed in

both operating theatres.

0 10 20 30 40 50

Leadership

Maintenance of Standards

Planning and Preparation

Workload Management

Authority & Assertiveness

Team building & maintaining

Support of others

Understanding team needs

Conflict solving

Problem definition & diagnosis

Option Generation

Risk assessment & option selection

Outcome Review

Noticing

Understanding

Thinking Ahead

Total Number of Observations


Orthopeadic Surgery

Figure 7.2: Non-technical errors attributed to NOTECHS elements.

NOTECHS Scores

As with non-technical errors, the relationship between the ranked NOTECHS scores and

other intraoperative parameters were investigated utilising Spearman’s rho correlations.


Teamwork & Co-ordination


Situation Awareness



These data are shown in table 7.2. There were strong or moderate relationships in

cardiac surgery between NOTECHS rank and minor failures, threats and operative

duration. No such close relationships were observed in orthopaedics, though there

appears to be a weak relationship with minor failures. Since these data are derived

independently, they reinforce the view that non-technical skills can affect operative

performance, at least in paediatric cardiac surgery.

Rho n p


Threats 0.584 <0.005


Paediatric Cardiac

[observer 1 only]


24

<0.005

Minor Failures 0.400 0.081

Threats 0.249 0.290


Orthopaedic

[observer 1 only]


20

0.250

Table 7.2: Correlat ions between NOTECHS ranks and other intraoperative performance parameters.

Assessment of NOTECHS Scoring Method

Though all 24 cardiac operations featured the observations from observer one only,

fifteen of the twenty orthopaedic operations utilised dual observers, with eleven yielding

a complete set of NoTechs ratings from both observers. This provided a means by which

scale agreement and inter-rater reliability could be examined.

To examine the relationship between non-technical errors and NOTECHS scores, and

so partially validate the NOTECHS scoring method, the relationship between the two

measurement methods was examined. These results are shown in table 7.3. There were

agreements between the measures scored by observer one in both paediatric cardiac

operations, and across all the 20 orthopaedic operations. When considering only the

eleven operations with full dual observed data, the strength of association between

observer scores is slightly less strong, but remains high. However, the agreement with

scores from observer two for these eleven operations is very low. Thus, there is



reasonable agreement between the two methods when employed by observer one, but

not when employed by observer two.

Rho n p

Paediatric Cardiac Observer 1 all 0.740 24 <0.001

Observer 1 all 0.603 20 <0.01

Observer 1 dual only 0.576 0.064 Orthopaedic

Observer 2 dual only 0.108 11

0.751

Table 7.3: Agreement between non-technical errors and ranked NOTECH scores

Examining the agreement between observers on the results from their non-technical

error observations, there was a moderate correlation between the numbers of non-

technical errors seen by each observer (rho = 0.559, n = 15, p<0.05). The Bland-Altman

plot can be seen in figure 7.9, and the data would appear again to be heavily influenced

by the one computer navigated procedure.

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12 14 16

Mean of Non-Technical Errors [(Observer 1 + Observer 2)/2]

Rat

io o

f N

on

-Tec

hn

ical

Err

ors

[Obs

erve

r 1/

Obs

erve

r 2]

Figure 7.3: Bland-Altman plot for observed non-technical errors between observers The NOTECHS scale achieved little inter-rater reliability. There is no agreement between

ranked NoTechs scores from each observer, and Cohen’s Kappa shows a very low level




of agreement (Kappa = 0.158, n=132) between observer ratings of NOTECHS on each

scale item. Table 7.4 shows the crosstabulation of this data. The low kappa is brought

about by the high prevalence of “Standard” ratings across all operations and both

observers, a lack of agreement between observers for the small number of scale

extremes, and inconsistencies between what observer two scored on the NOTECHS

scale, and his non-technical error observations. Perhaps most tellingly, while the ranked

NOTECHS scores from observer one identified the computer navigated operation as the

worst, observer 2 scored it as the fourth best. Given the clear failure in teamwork that

this operation represented (see section 6.9), it would appear that observer two, who was

less experienced in the method, found the NOTECHS scale difficult to use.

Observer 2

Below Basic Standard Exceed Total Below 1 1 Basic 1 4 19 2 26 Standard 10 73 2 85 Exceed 5 11 4 20

Observer

1

Total 1 19 103 9 132 Table 7.4: Crosstabulat ion of NOTECHS scores from both observers.

7.4 Discussion

It has been argued that training in non-technical skills might help to reduce the impact

and frequency of failures in both orthopaedic and paediatric cardiac surgery. This study

offers evidence that non-technical performance can influence the success of the surgical

process, particularly in cardiac surgery30;90 and should be assessed alongside technical

performance. The measurement method would need to be improved and the skills of

observers would need to be better defined to obtain a more reliable measure. However,

it is encouraging that both non-technical error assessments and global non-technical

skills assessments are in agreement when administered by the more experienced

observer. Clearly non-technical errors were much more common than technical errors,

and the major vignettes in section 6 demonstrate how failures in non-technical skills can

create highly undesirable situations. The detailed attention given to non-technical skills

in this section shows where those failings may lie. To this end, the results from these

studies are worthy of further consideration as they can inform future work both in the

development of assessment scales, and in the understanding of the relationship

between technical skills, non-technical skills, error and outcome.



Technical and non-technical errors were independent in orthopaedics; there was also

little relationship between non-technical skills and other intraoperative performance

measures. This suggests that team working has less impact on the surgical process in

orthopaedics than in cardiac surgery, where several teams within the theatre must

interact. However, close coupling between equipment, procedure, and the team is just as

much a feature of orthopaedic surgery as it is of cardiac surgery. Whereas the

relationship between anaesthetist, perfusionist and surgeon is central to the success of a

cardiac operation, and may require detailed diagnostic discussions, it was the largely

tacit relationship between the scrub nurse and surgeon that was central to the success

of a TKR procedure (though we should note that in this respect our results may have

been skewed by how familiar the team were with one another). While the surgeon and

the anaesthetist rarely shared any discussion of the patient during the orthopaedic

operation, the task sharing and diagnostic discussions in cardiac surgery were frequent,

explicit and critical to the operation. Though not directly a safety compromise, common

critical tasks, such as the application of bone cement, would have benefited from better

co-ordination, especially given the acute effects on the patient that a poor reaction to the

cement would produce. Any such infrequent but well recognised emergency situations

would be exacerbated by the apparent dissociation between the anaesthetist and the

surgeon. Moreover, this reveals a poor use of team resources.

Comparison between the analysis of failures on the NOTECHS dimensions and the

distribution of those failures across the elements of each dimension shows additional

detail. Failures in the leadership and management dimension were frequent, but

assessment of the elements suggested that it was in the planning and preparation,

rather than procedural leadership, that failed. Indeed, it may be that overly strong

leadership creates an atmosphere where the rest of the team follows the leader without

making sufficient plans or preparations of their own. There was little difference between

the types of non-technical errors in the different treatment centres, which suggests these

are healthcare-wide behaviours. In both operating theatres there was a tendency to start

before the whole team was ready, and there were clear divisions between different

specialties, which sometimes gave rise to conflict, and sometimes simply to independent

actions that would have been better co-ordinated through the team. These are

reflections of cultural attitudes. A better team culture would have a recognised leader



who encouraged participation from all members, and provided an environment where

understanding, planning and action are shared, reducing the chance of individual errors.

Though the two techniques for non-technical assessment were usually in agreement,

they were markedly different in the way in which they are applied. Each method has both

advantages and disadvantages. The error analysis, which was based on assessment of

minor failure types (sections 4 and 5) provides the absolute number of non-technical

failures in an operation, and allows detailed diagnostic assessment of each failure. It is,

however, time-consuming to employ, without providing any measure of positive non-

technical performance. The global skill rating arguably provides a better indication of

overall team performance and is less demanding to employ, but is far more reliant on the

judgement of the observer. This suggests a trade-off between the accuracy of the

measure and the effort that needs to be expended to achieve it. Wider use of this

technique would require refinement to improve accuracy and reliability.

The disappointing reliability of the global NOTECHS scores arose for a number of

reasons. The clear divide in orthopaedics between the anaesthetist and the surgeon

made scoring performance of the team difficult. Further development of the scale should

allow for the rating of differences between individual team members or groups. Indeed,

since differences in shared understanding suggest poor teamwork, this might make a

sensitive and useful evaluation criterion. As well as requiring the integration of

observations between individuals, the global NOTECHS scoring method also required

integration of observations over time. This meant that the scale was not well suited to

subtle or transient changes in non-technical performance, and may be because it was

originally designed as a more simplistic training aid with a binary pass/fail outcome

rather than rankings or scores. A further difficulty was that while the team-related

dimensions (leadership and management and teamwork and co-ordination) directly

related to observable behaviours, the cognitive dimensions (problem solving and

decision making and situation awareness) do not, and needed to be inferred. This

proved particularly difficult for observer two, who did not have experience in the

observation of human performance and non-technical skills, and clearly found

NOTECHS challenging to use. Four operations that involved this observer had

incomplete data sets and had to be discarded, simply because he was unable to provide

a score. There was also a difference in the focus of attention between observers.



Observer two concentrated on the work of the anaesthetist because he had prior

experience of this type of work. In particular, this observer felt that the situation

awareness of the anaesthetist was often far lower than the rest of team. Observer one

concentrated on the surgical team as a whole, while not assessing any individual with as

much detail. This difference in attention is probably why observer two gave a high score

to a clear failure in teamwork and situation awareness. Though non-technical skills

evaluation in the team-critical tasks of cardiac surgery may have been easier, it was

difficult to achieve a sensitive measure of non-technical performance in orthopaedics

using this global rating method. The method derived from failure observations, focusing

only on the non-technical errors, was more sensitive.

Given an analysis framework developed in aviation, it is possible objectively to evaluate

non-technical skills against observations of failures in operating theatres. Modifications

will be required to provide better reliability and allow for variation in team composition

and different procedures. There remains a strong case for developing training tools that

are developed specifically for healthcare91. While the development of these tools cannot

always be expected to be quantitatively derived92, we would suggest that quantitative

methodologies should be adopted as frequently and as openly as possible. However,

given that humans are the key component of all high technology systems, and that

effective non-technical skills are likely to be similar for people functioning in many

different environments, substantial benefit can be gained in healthcare by utilising

existing knowledge from aviation. It is also critical to measure the non-technical skills

across specialties as well as within them. This study has demonstrated how that might

be done, and has identified the trade-offs that should be considered in the non-technical

assessment of surgical teams.

7.5 Summary

In this section, the relationship between failures, errors, and non-technical skills was

examined.

• Intraoperative performance was clearly affected by non-technical behaviours in both

types of surgery.



• Since surgical demands are different, they have different non-technical skills

demands. In particular:

o Teamwork between surgeon, anaesthetist and perfusionist is critical in

cardiac surgery, whereas the key relationship in orthopaedics is between

the surgeon and the scrub nurse.

o Communications in cardiac surgery often involve complex diagnostic

discussions within the team, whereas in orthopaedics the communication

are often non-verbal, or limited to discussion of the process.

o The common requirement for effective teamworking, particularly when

responding to emergencies, may require specific processes for these

environments. Maintaining vigilance and awareness will always be

challenging in situations of reduced demand or arousal, but is

nonetheless required.

• There will be a trade-off between accuracy and ease-of-use in applying non-technical

assessment scales. The attempts to apply the more simple NOTECHS rating system

was partly successful, but was not sufficiently reliable to provide a suitable measure.

Since both scale design and the skill of the observer are important, future work should

examine the requirements for both.



8. General Findings from the Observational Studies

8.1 Overview of the Studies

Behaviour in operating theatres has rarely been studied beyond the assessment of the

specialist technical skills of individuals. In the past there has been a tendency to blame

either the condition of the patient, or the skill of individual practitioners, following poor

patient outcome. The evidence presented here reinforces the view that study of the

aetiology of failure and the failure-inducing elements of the surgical system are essential

if a reduction in errors and their consequences is to be achieved.

The data presented here form one of the first attempts to understand how the context in

which operating theatre teams work can influence their behaviour and the course of the

operation. The patterns of the minor failures that were observed, minor failures leading

to major failures, and the influence of risk on both minor and major failures, was similar

in both orthopaedic and paediatric cardiac surgery. Though teamwork was more

obviously critical for success in paediatric surgery, there was similarity in the non-

technical skills deficiencies identified in both operating theatres. Since these elements of

safety have rarely been examined before, and non-technical skills are not formally taught

in healthcare, it is likely that the findings of this study are broadly applicable to other

surgical centres. Common minor failures are allowed to recur because they are

considered, in isolation, to be innocuous. The evidence presented here suggests that

when co-incident with other types of failure, especially in higher-risk operations, they can

be highly potent. This also demonstrates the misleading effect that can be created by

weighting failure types for severity. A failure that is innocuous in one situation can be

critical in another.

In order to conduct this research, the co-operation of the teams at both centres was vital,

but not easily achieved. During the early studies we became acutely aware that events in

theatre are not commonly or objectively discussed outside the theatre doors. Clearly, the

need to understand better how events in surgery influence patient outcome was not

universally acknowledged by healthcare practitioners, who either did not believe this

type of work could help to improve surgical quality, or preferred to tolerate surgical error.

Our research also provoked negative reactions from some at both centres which limited



our ability to collect data, and almost certainly skewed our results toward the more

effective operating theatre teams. It was encouraging that the theatre teams were

generally supportive to the aims of the program, but less encouraging that only one

orthopaedic surgeon was supportive of our work. Without the participation of this

individual, the orthopaedic studies would have been impossible. Researchers or

solution-providers in this field are likely to face similar defensive cultural problems even

though this inappropriate professional sensitivity is ultimately neither in the interest of the

public, nor of the profession.

Once our observations were underway, the theatre operating teams were almost

universally polite, helpful, and positive in their attitude. Most individuals that participated

in our studies recognised that improvements to surgical safety must involve post-

operative learning. Though frequent recurrent failures were observed, the study

participants were clearly more concerned with improving surgical quality than those that

chose not to co-operate. Since the surgical teams studied varied, and technical

competence of the individual team members was rightly assumed, the general patterns

of failure identified here cannot be related to the sub-standard performance of

individuals. Rather, since one treatment centre was of world-class standard, and the

operating theatre team in the other treatment centre was known to be particularly

effective, we consider the data presented here to be a conservative estimate of the

failures that regularly occur in operating theatres across the UK. Though we made no

direct link between our observations and patient outcome, this level of recurrent failure,

coupled with the inability of some practitioners even to acknowledge the importance of

learning from intraoperative failures, demonstrates why the improvement in safety in

healthcare is so problematic.

8.2 Threat and Error Profiles in Orthopaedic and Paediatric Cardiac Surgery

The propagation of minor failures to major failure, and the combination of accumulated

minor failures with critical opportunities was not as potent in shorter, less complex

orthopaedic operations as it was in the longer, more complex paediatric cardiac

operations. Though the same major failure definition was used in both types of surgery,

the single major failure in orthopaedics was very close to the threshold of the definition,

but was not as complex or threatening as the major failures identified in paediatrics.



It seems reasonable to suggest that serious compromises to patient safety in

orthopaedic surgery were less likely because there are fewer critical phases in the

operation, which meant that there were fewer opportunities for minor failures to

accumulate during critical moments. It should also be noted that since it may take days,

weeks, or years for differences in surgical quality to be noticeable in orthopaedics, the

coupling between intraoperative events and outcome will be less clear than in cardiac

surgery, where the effects of failures can be noticeable immediately after separation

from bypass, or within hours of leaving the operating theatre. The standardised and

high-volume nature of orthopaedic surgery also means that the surgical processes are

far more clearly defined before the operation, re-locating the source of many potential

intraoperative failures to other pre- and post-operative system processes. The

characteristics of paediatric cardiac surgery – higher risk, greater technical difficulty,

greater reliance on teamwork, greater uncertainty over the patient and the details of the

surgery – provides the greater opportunity for errors to act in critical circumstances to

create major failures in theatre. However, it is also important to note that the volume of

TKR procedures in the UK is much greater than the volume of open-heart paediatric

cases. Consequently, a simple assessment of risk (probability x frequency) would

suggest that the large number of failures identified in orthopaedics in the current study

warrant further consideration, even if the link with outcome has yet to be established.

The long period between events in orthopaedic surgery, and reliable outcome

measures93 makes prospective study particularly important.

The results of threat and error analysis for the different surgical types are particularly

useful, as they condense these observations into profiles, displayed in figure 8.1, that

make the differences far more apparent. Patient and task threats accounted for fewer of

the failures in orthopaedics than in paediatric cardiac surgery, but cultural / organisation

threats were far greater. This is because the patients were more uniform, the procedures

were less variable, and the distractions greater, in orthopaedics. The threats from

equipment and errors that were not associated with any threat were similar, while clearly

cultural issues had considerably more salience in the orthopaedic theatre. The level of

threats that cause no error was approximately the same, but the difference in relative

frequency between technical and non-technical errors was greater in orthopaedics,

again demonstrating the challenges faced by the paediatric cardiac teams. It also

reflects the closer relationship between anaesthetic, nursing and surgical tasks in this



0%

10%

20%

30%

40%

50%

Patien

tTa

sk

Enviro

nmen

t

Culture

/ Orga

nisatio

n

No Thre

at

Tech

nical

Error

Non-Te

chnic

al Erro

r

No Erro

r

0%

10%

20%

30%

40%

50%

Orthopaedic Surgery

THREATS ERRORS


type of surgery, where a non-technical error can often lead directly to a technical error. In

orthopaedics, non-technical errors were far more likely to occur independently of

technical errors. Thus, the threat and error profile generated using the failure source

model can help illustrate the different demands of different types of surgery in terms of

the skills required of the individuals, and the requirements of the system to support those

individuals. Since they arise from observations of breakdown in operations or function47,

these profiles illustrate the weaknesses within the systems of surgery which require the

operating team to adapt and compensate. Improving the system by reducing these

deficiencies would lead to increases in patient safety.

Figure 8.1: Threat and error profiles for orthopaedic and paediatric cardiac surgery The percentage scores show the proportion of the total number of threats, and the total number of errors, accounted for by each threat or error dimension.

8.3 Reducing Errors in Surgery

As weaknesses apparent in the operating theatre may be associated with deficiencies at

national, centre-specific and surgery-specific levels, the consideration of solutions needs

to be both local and global. Local (surgery- and centre- specific) solutions have been

suggested within the discussions of the relevant results (sections 4.4 and 5.4). In this



section, we examine global solutions to reducing errors in surgery, for which there are

three approaches:

• Avoidance of failure by the resolution of systemic properties that predispose human

error1

• Capture of failures before they can accumulate to generate a more dangerous

situation10;18

• Mitigation or compensation for failures before they can impact on patient mortality and

morbidity7.

Adverse outcomes are best avoided by removing the latent conditions within the system

that predispose undesirable events. Technology-oriented industries utilise Standard

Operating Procedures (SOPs) and checklists to reduce variability in human behaviour,

and so improve safety. This approach may sometimes be too rigid to suit healthcare, as

it cannot rely on technology and must allow individual patient requirements.

Intraoperative checklists and SOPs, however, would certainly help reduce many of the

task- and equipment-related threats encountered by surgical teams. Aside from

improved surgical techniques and diagnostic processes, which continuously evolve as

part of ongoing medical practice59 there are few opportunities to avoid task and patient

related threats. The development of appropriate technologies that support the surgical

process63;64;94;95 may improve existing standards or create better ways of managing

difficult cases. However, with the large number of existing equipment-related problems

encountered in this and other studies conducted over many years60-62, caution must be

urged in the introduction of more complex or safety-critical technologies to the operating

theatre. Attention to the design and maintenance of existing equipment provides

attainable solutions that the introduction of new surgical technologies cannot yet

demonstrate. Moreover, our own observations suggest that while equipment

management issues are being partially addressed in new technologies, appropriate

human factors96-98 and safety considerations99 are not being taken into account.

Increasing technical complexity may merely re-locate failures, making instrumentation

more difficult to use, and consequently less reliable, in the future. It can also lead to de-

skilled personnel, boredom, fatigue, and workload polarisation, where a situation can

suddenly change from being very safe, to safety-critical, in a short period of time. These

mistakes have been documented frequently before in other industries100-102 and systems



designers should be directed to a considerable body of scientific work demonstrating the

vital role that human-centred task and equipment design play in systems safety23;60;61;96-

98;103. We would recommend the development of appropriate user-centred design and

procurement protocols104, combined with auditable maintenance and failure-monitoring

programmes.

Cultural threats were observed in both treatment centres. In particular, theatre discipline

– adherence to safety protocols and the management of absences, distractions – was

deficient in both operating theatres. Though the problem was greater in orthopaedics,

responsibility for patient safety was confused in both theatres, with a great deal of

variation in levels of discipline, and few opportunities for individuals to suggest or

facilitate improvements. A universal approach to, for example, mask discipline, was

unlikely because anaesthetic consultants often felt that they did not need to adhere to

protocol, even when explicitly asked to do so by consultant surgeons. This also set an

example for others in the operating theatre, and it sometimes seemed that the more

senior the practitioner, the more safety protocols they would break. The use of mobile

phones in theatre by senior staff was particularly apparent in this respect. This culture

may have arisen because of the lack of evidence that discipline improves safety, rather

than any suggestion that adherence to safety protocols was detrimental to the patient.

This reveals a conflict between the putative evidence-based culture and the real safety

culture. The effect that safety procedures have will not always be scientifically

demonstrable, and a culture that concerns itself only with positive evidence is one in

which basic safety practices become eroded. Since healthcare lends itself far better to

scientific evaluation of safety-related interventions than other industries, demonstrating

the value of safety procedures is vital. In this respect, the results of the present studies

are of particular value. Though it may be difficult to prove that the management of

distractions, for example, can directly improve patient outcome because they only rarely

interfere with a critical task, it provides an obvious and easy opportunity for

improvement. However, unless adopted and adhered to by the all team members,

particularly the most senior, these threats will continue. The assessment of safety

attitudes, and techniques for examining other aspects of work organisation and patient

management, are examined in Part II and Part III of this report.



It is the failure to learn from the negative events in each operation that allows the

recurrence of minor failures. This is partly because theatre teams believe they should be

“able to cope with anything”13, partly because they may prefer to develop a work-around

than force an examination of the source of the problem105;106, and partly because there

has been little clear evidence that minor failures can affect patient outcome. The first

important step to improving safety within healthcare is to recognise and learn from

previous mistakes, and understand how and why errors occur. Non-technical

deficiencies were identified in planning and preparation, understanding team needs,

noticing important elements in the environment, and workload management. Operations

were often started before all the team members were ready, there were clear

delineations between different specialists in the operating theatre, a great number of

distractions from important information, and there were clear deficiencies in the ability of

the teams to think ahead and anticipate problems during an operation. There are simple

interventions that could be developed to address these issues and provide immediate

benefits. An awareness of the relationship between errors, and the properties of the

system that predispose to them provides practitioners with a clearer view of the

importance of identifying and reducing systemic threats, rather than simply learning to

cope with them. Furthermore, an appreciation of how errors in surgery and in other

domains can occur will help teams compensate for the problems that they face. The

ability to ‘flag’ certain events or impending situations as having the potential for error

would be a more effective safety-management technique than the false optimism that is

often observed. At least five of the seven major failures (sections 6.2, 6.3, 6.4, 6.5, 6.7

and 6.8) arose because of an inability to recognise and reduce the potential for error in

uncertain, patient- and task- induced high workload situations. A postoperative

debriefing after each operation to identify both the successes and failures in the

operation, complemented with an audit trail or database system to record the salient

information would allow the tracking of these undesirable events over time and between

teams. Providing an enhanced ability to identify sources of recurrent failure, and develop

strategies for their management and future reduction107 may result in the reduction of

many recurrent failures.

Briefing with the entire team prior to commencing surgery, which is successfully used in

aviation to explore areas of concern and potential difficulty, would also be useful.

Planning of the required sequence of events, identifying potential or expected threats,



and agreeing avoidance or mitigation contingencies serves as a method to predict and

manage potential errors. It also serves as a method to build the team preoperatively –

especially important where new or less experienced team members are present – and to

set the required standards of behaviour. Briefing provides a shared model for best

practice, and encourages the participation of all members in the avoidance, capture, and

mitigation of errors and the identification of best practice. It was concerning that during

the major event that involved the surgical omission (section 6.3), no team member

reminded the surgeon of his omission. This reflects either poor knowledge in the team

about the procedural details, or the failure of all team members to be involved in the

safety of the patient. A range of threats, errors and failures could be anticipated and

managed better by a relatively simple briefing process.

In this study we chose not to examine the technical skills of the individuals involved.

Though we measured a considerable number of technical errors, we do not believe any

individual was seriously deficient in their technical skills. Indeed, we have already

suggested that the individuals studied were highly competent. However, given the

enormous variation in the composition of the operating teams, and the lack of any kind of

formal training in non-technical skills, it was unsurprising that non-technical errors proved

frequent in both types of surgery. Non-technical skills provide a trainable and

measurable basis on which safety-related interactions occur within the team. Having an

agreed and effective way to interact is especially important where large variations in

team composition, experience, and responsibility are present. Non-technical skills

training tends also to level the authority gradient between senior and junior staff, which

otherwise discourages effective teamwork and safe practice. In this respect, these skills

will be closely associated with the success of briefing and debriefing, and in helping to

avoid systemic threats or capture minor failures before they can accumulate7;18;42;84;90.

The experience of aviation in the training and evaluation of non-technical skills should be

utilised36;52;86;108. However, since there are fundamental differences in team composition

and task demands, scales or training regimes derived from aviation form only the first

step towards understanding non-technical skills in healthcare. Training programmes

should be developed with anaesthetists, surgeons, and other theatre staff specifically to

address the requirements of surgery, and should be evaluated objectively against

measures of intraoperative performance. Since safety-critical errors are more frequent in

healthcare than aviation, the operational importance of non-technical skills could be



more directly measurable and have a greater positive benefit than has been found in

aviation.

8.4 Methodological Considerations

These studies demonstrate the potential benefit for comprehensive data regarding

intraoperative failures. They are, however, subject to fundamental difficulties with the

measurement and classification of errors obtained from observations of events40;109. The

quantification of failures will always be inexact, given that certain failures (such as pre-

operative diagnostic failures) can have longer lasting effects than more transient failures

(such as a dropped scalpel). Weighting for severity, as has already been argued, may

also be misleading. Similarly, the measurement of failures over time, interoperative rates

or absolute values depends on the purpose of study. While some technical errors will not

have been captured by the expert observer, this method enabled the identification of

failures that would have otherwise been accepted as normal by medically-trained

participants87. Bias away from individual technical skills was appropriate for a study that

sought to avoid blame and examine systemic properties. In the measurement of

problems between specialties this approach is preferable when compared with

observations biased toward any single specialty. Our studies suggest that for this type of

prospective observation, experience in behavioural analysis methodologies is preferable

to experience of working in healthcare.

The use of video recordings required considerable sensitivity toward the staff, and

resulted in problems recruiting participants, and considerable attrition in the number of

consenting patients available for study. Access to the video data was strictly controlled,

which placed further limitations on using these data. Despite evidence of the benefits

that it can bring, the difficulties of video recording behaviour in healthcare contrasts with

the acceptance that it has found in other safety-critical industries31;110-113. Ethics

protocols, which are configured for the study of patients rather than healthcare

practitioners, may be a particular barrier to the progress of research in this area.

Nevertheless, the combination of semi-explicit data collection, the use of video

techniques, and the careful definitions used to analyse the data complements the

continuing development of video methodologies in medicine31;32;37;111;114-116. Further

attention to acceptance by operating teams, medico-legal issues, and to video and audio



recording fidelity, flexibility and intrusiveness are recommended to maximise the utility of

video for the assessment of surgical error.

Operating teams are often reliant on implicit procedures or concepts of good practice

which change from team to team and are consequently highly variable. The largely

unfruitful attempts to define processes and error through task analysis reflect this, and

suggests that observation methods used in healthcare cannot be directly transferred

from aviation, or other highly procedural industries. Nonetheless, dual observers in

orthopaedic surgery were able reliably to determine more and less failure-prone

operations. Both observers were able to identify a considerable number of failures, with

the small overlap between observations suggesting differences in experience and focus

of attention. Intra-observer reliability might be improved by the explicit definition of salient

events. Future work should develop the minor failure definitions into an explicit checklist,

and should examine the reliability of minor failure classification in order to improve the

measurement value of this tool. However, given the unpredictable nature of errors and

the importance of understanding the context in which they happen, we would advocate

the value of implicit error capture conducted by experienced observers utilising a

structured post-observation assessment protocol. This retains context, allows scrutiny of

the method, and alternative assessments of the data. Understanding the threats and

errors encountered by surgical teams in this way makes possible the diagnosis of these

important but often neglected events. The techniques developed here therefore

contribute to the development of methodologies for the direct observation of human error

in medicine7;31;32;37;108;111;117.

The present research has demonstrated the sensitivity of this method to the demands of

different types of surgery, and the measurement of systemic threats, human errors, non-

technical skills, and safety-critical events. It emphasises the effect of system

components and human factors in patient safety; and thus why adverse outcomes in

surgery must, and can be, examined beyond the traditional considerations of patient and

practitioner alone.



8.5 Summary

• The methods developed through these studies, utilising a systems approach to

the assessment of surgical error, allow the prospective identification of

deficiencies in the system that create those errors.

• Adverse events in surgery are likely to be associated with a number of frequently

recurrent, co-incident and cumulative human errors predisposed by threats

residing in the system, rather than individual incompetence or negligence.

• More evidence is needed to inform healthcare practitioners of the prevalence and

nature of error in surgery, and the benefits of adhering to safety practices.

• User-centred equipment design and procurement, coupled with improved

maintenance standards, would reduce equipment-related problems.

• Understanding how errors in surgery occur can help the anticipation and

avoidance of error-inducing situations.

• Briefing and debriefing can help avoid, capture, and mitigate present and future

failures.

• Non-technical skills training would provide the mechanism by which all these

safety practices can be managed and communicated effectively within varying

teams.

• Development and deployment of observational and non-technical skills

evaluation techniques utilised here would increase the body of safety-related

knowledge in operating theatres, and allow objective measurement of future

interventions.



PART II: ASSESSMENT OF CULTURE AND INSTITUTIONAL RESILIENCE

9. Introduction

In the previous chapter, data relating to errors in surgery, teamwork, system

configuration, and culture were presented. In particular, cultural, teamwork and discipline

issues associated with safety in surgery were identified. It is clear that the organisation

can influence patient safety – or conversely, create the risk of an organisational

accident118. Organisational safety health is a term used to describe the ability of an

organisation to prevent the occurrence of adverse incidents, to recover from and

minimise the damage caused by adverse incidents, and to promote an environment

whereby the organisation can learn from such incidents and educate staff in a blame-

free culture. An Organisation with a Memory2 summarised how the concepts of good

organisational safety health should be applied to NHS hospitals. It is clear that in

addition to implementing good organisational safety health methods, there need to be

techniques to assess these methods. Such techniques might be used to assess

deficiencies in trusts or departments in a clear manner, to guide improvements to

organisational health in individual hospitals.

One of the aims of this research project was to pilot the use of a method for auditing the

organisational safety health of healthcare institutions. The method consisted of two

questionnaire surveys that were distributed amongst staff and a checklist to be

completed by whomsoever is conducting the audit. The Checklist for Institutional

Resilience (CAIR) examines the structure and procedures that are in place for

monitoring, avoiding and managing errors at the organisational level29. The Incident

Reporting and Attitude Survey (IRAS) collects data from healthcare practitioners on the

management of and attitudes to the collection and dissemination of safety information

and critical incidents. Finally, the Operating Theatre Team Management Attitudes

Questionnaire (OTTMAQ) examines the practitioners perception of team and individual

performance in relation to safety in surgery88;119. Each of these three components is

designed to assess a different facet of the overall organisational safety health associated

with a particular clinical department. In Part II of this report we describe each component

of the audit method and report on our experience in piloting them.



10. Checklist for Assessing Institutional Resilience

10.1 Introduction

In this section, a pilot study of the CAIR (Checklist for Assessing Institutional Resilience)

in the paediatric cardiac unit is described. The CAIR was designed by Carthey, de Leval

and Reason2 and is intended to be a convenient means to estimate the organisational

safety health of a hospital department. It examines three dimensions of safety culture;

• Commitment. The motivation and resources applied to the pursuit of patient safety

within a trust.

• Competence. The ability of an organisation to enhance safety, defined by, for

example, the hazard identification processes and organisational flexibility to manage

hazards

• Cognisance. The capacity of the organisation to understand safety and be continually

aware of risks.

The CAIR examines these three dimensions from 20 different markers of good

organisational safety health which assess the principles, policies, procedures and

practices of the healthcare organisation. Together, the presence of these factors is

considered to be indicative of a system with a good defence against threats to patient

safety3. The markers concern diverse issues, such as the effectiveness and presence of

the risk management team, how the senior management of the Trust disseminate

lessons learned from the reporting of adverse incidents and how proactive the high-level

management of the Trust is when it comes to patient safety. Full details of the CAIR are

given in Appendix 5.

The CAIR therefore potentially provides a method to measure and assess the ability of a

healthcare organisation to manage and respond to organisational risks or accidents.

10.2 Methods

In order to gather the data necessary to derive a score using the CAIR two approaches

were adopted. The first consisted of a review of the policies, meetings and procedures

related to patient safety that were in place at the trust. Both trust-wide and departmental



policies concerning patient safety were reviewed. A meeting was held with the Trust risk

management team and details of their role and function were obtained. Adverse

incidents relating to the ICU were reviewed and minutes of the weekly departmental

clinical governance meeting, multi-disciplinary conferences and senior management

meetings were also reviewed. Equipment purchasing policies were discussed with senior

management. In conducting these reviews, the issue of patient safety was examined

from both organisational and patient-specific perspectives.

The second approach was to conduct semi-structured interviews with a wide range of

both clinical and non-clinical members of the cardiothoracic unit. The aim of the

interviews was to collect evidence relating to adverse incident reporting, information flow,

frequently occurring problems in the cardiac patient pathway, suggestions for

improvements that would increase patient safety, staff interaction with the risk

management team and also to elicit ‘vignettes’ concerning adverse incidents that the

members of staff had experienced. Scores for each item on the checklist were assigned

by a member of the research team based on the information collated during the two

complimentary evidence gathering activities. The scoring convention used for each item

of the CAIR is as follows:

Ø “YES = This is definitely the case in this institution" 1 point

Ø "? = Don’t know, maybe, or could be partially true" ½ a point

Ø "NO = This is definitely not the case in this institution" 0 points.

The total score, obtained from a sum of the scores for the individual items, indicates the

overall level of the institution’s defence against threats to its safety. The following

categories are used to guide the interpretation of the overall score.

16-20 an institution so healthy as to be barely credible;

11–15 a moderate to high level of intrinsic resistance

6–10 considerable improvements are needed to achieve institutional resilience

1–5 moderate to high institutional vulnerability

0 a complete rethink of organisational culture and processes is needed.



10.3 Results

The completed CAIR is shown in Table 1.The total score from the Checklist was 16 out

of a possible 20, which gives the paediatric cardiac unit an overall rating of "So healthy

as to be barely credible". Note that each of the items in the checklist was judged to be

potentially present to some degree in the paediatric cardiac unit, hence the overall high

score. Details of the evidence collected can be found in Appendix 6 and a commentary

explaining the scores given for each item of the CAIR is included in Appendix 7.

INDICATORS OF RESILIENCE YES ? NO 1) Patient safety is recognised as being everyone’s responsibility, not just that of the risk management team.

2) Top management accepts occasional setbacks and nasty surprises as inevitable. It anticipates that staff will make errors and trains them to detect and recover them.

3) Top managers, both clinical and non-clinical, are genuinely committed to the furtherance of patient safety and provide adequate resources to serve this end.

4) Safety related issues are considered at high level meetings on a regular basis, not just after some bad event.

5) Past events are thoroughly reviewed at high level meetings and the lessons learnt are implemented as global reforms rather than local repairs.

6) After some mishap, the primary aim of top management is to identify the failed system defences and improve them, rather than seeking to pin blame on specific individuals.

7) Top management adapts a proactive stance towards patient safety. It does some or all of the following: Takes steps to identify recurrent error traps and remove them; strives to eliminate the workplace and organisation factors likely to provoke errors; brainstorms new scenarios of failure; conducts regular ‘health checks’ on the organisational processes known to contribute to mishaps.

8) Top management recognises that error-provoking institutional factors (e.g. under-manning, inadequate equipment, inexperience, patchy training, bad human-machine interfaces, etc) are easier to manage and correct than fleeting psychological states such as distraction, inattention and forgetfulness.

9) It is understood that effective management of patient safety, like any other management process, depends critically on the collection, analysis and dissemination of relevant information.

10) Management recognises the necessity of combining reactive outcome data (i.e. from the near miss and incident reporting system) with proactive process information. The latter entails far more than occasional audits. It involves regular sampling of a variety of institutional parameters (e.g. scheduling, roistering, protocols, defences, training).

11) Meetings relating to patient safety are attended by staff from a wide variety of departments and levels within the institution.

12) Assignment to a safety-related function (quality or risk management) is seen as a fast track appointment, not a dead end. Such functions are accorded appropriate status and salary.

13) It is appreciated that commercial goals, financial constraints and patient safety issues can come into conflict and that mechanisms exist to identify and resolve such conflicts in an effective and transparent manner.

14) Policies are in place that encourage everyone to raise patient safety issues.



INDICATORS OF RESILIENCE YES ? NO 15) The institution recognises the critical dependence of a safety management system on the trust of the workforce, particularly in regard to reporting systems. (A safe culture, i.e. an informed culture, is the product of a reporting culture that, in turn, can only arise from a just culture).

16) There is a consistent policy for reporting and responding to incidents across all of the professional groups within the institution.

17) Disciplinary procedures are predicated on an agreed distinction between acceptable and unacceptable behaviour. It is recognised by all staff that a small proportion of unsafe acts are indeed reckless and warrant sanctions, but that the large majority of such acts should not lead to punishment. (The key determinant of blameworthiness is not so much the act itself - error or violation - as the nature of the behaviour in which it was embedded. Did this behaviour involve deliberate unwarranted risk-taking, or a course of action likely to produce avoidable errors? If so, then the act would be culpable regardless of whether it was an error or violation).

18) Clinical supervisors train junior staff to practise the mental as well as the technical skills necessary to achieve safe performance. Mental skills include anticipating possible errors and rehearsing the appropriate recoveries.

19) The institution has in place rapid, useful and intelligible feedback channels to communicate the lessons learnt from both the reactive and proactive safety information systems. Throughout the institution the emphasis is upon generalising these lessons to the system at large rather than merely localising failures and weaknesses.

20) The institution has the will and the resources to acknowledge its errors, to apologise for them, and to reassure patients (or their relatives) that the lessons learnt from such mishaps will prevent their recurrence.

Table 10.1: The completed Checkl ist for Assessing Insti tut ional Resi l ience.

10.4 Discussion

The achievement of a score of 16 out of 20 suggests results that were unrealistically

high. However, before a discussion of the scoring methods, it is worth examining the

utility of the scale for identifying weaknesses in the organisation. There are two principal

areas which would contribute to a greater score on the checklist.

At present, the multi-disciplinary mortality and morbidity meeting is held in wide regard

as being an open and informative forum. However, due to the fact that its agenda

focuses on incidents that have occurred in ICU, there is potential for a number of vital

issues to be missed, such as those occurring in theatre or outpatients. It is also difficult

given the extremely busy nature of NHS staff members to schedule extra meetings and

expect them to be well-attended. Otherwise, dissemination of findings from

investigations of incident reports is generally via quarterly reports and the Trust intranet,

both of which are passive means of dissemination. Conducting the CAIR process thus



helped identify the need to develop a method to inform and disseminate the outcomes of

investigations into adverse incidents actively to a wider number of staff.

The CAIR also helped to identify a gap in the chain of discussion and dissemination of

reports of adverse incidents. Discussions involving top-level management and the risk

management team are held at regular intervals. It is clear that at divisional level and

above, there is a considerable amount of discussion leading to changes in policy, for

example. However, while the risk management team is pro-active in terms of liaising with

clinical staff involved in the incidents themselves and also at producing the quarterly

Trust-wide summaries, neither the interviews conducted for this research nor the reviews

of the policies appeared to show a similar mechanism for clinical staff working at the

divisional level or below (for example, registrars). This was one of the principal reasons

in which the scores for these items on the Checklist were given as 0.5 rather than 1.

Consequently, though the policy was to discuss the outcome of Root Cause Analyses of

incidents at the weekly mortality and morbidity meetings, to the knowledge of the

research team, who attended the vast majority of these meetings, only one was

presented in two years of study. Furthermore, this was not presented at the well-

attended mortality and morbidity meeting, but during a less well attended teaching

session, and was generally badly received by the attendees. Thus, the CAIR helped

identify problems within the organisation of presenting this type of information, and of

obtaining the appropriate support of practitioners to implement changes. To effectively

manage this aspect of safety these incident feedback mechanisms therefore need to be

presented more frequently, and at a more appropriate time, and they need to involve

practitioners more closely. This also requires that resources are allocated specifically for

these purposes.

Though the completion of the CAIR process by a safety practitioner not directly

associated with the trust provides a less biased assessment, it can be difficult to

establish whether a certain piece of information does not exist or whether it has simply

not been accessed. This difficulty in demonstrating a null led to a score of 0.5 for several

items that might have been scored zero if it had been possible to establish a complete

information set. As has already been indicated, it was also clear that there were

differences between policy and practice, again leading to a greater score than should



realistically have been obtained by this organisation. Furthermore, there are no

interpretations for the scores 15 ½, 10½, 5½ and ½ which are feasible, but which are not

covered by the current categorisation. However, clearly the process of the CAIR

assessment provided a structured way of assessing the potential safety gaps within the

organisation for a non-safety specialist. Thus, while this pilot work suggests that further

assessments of the validity and reliability of the CAIR scoring mechanism would be

required before the tool offered reasonable differentiation between Trusts, it was found

that the tool was useful for identifying organisational weaknesses. Further development

of the scoring methodology should be conducted before renewed attempts are made to

validate the scale.

10.5 Summary

The pilot work reported here examined the CAIR as a tool for assessing and scoring the

safety health of an organisation. The key findings of applying the tool were that:

• The score of 16 out of 20 that was obtained may have been unrealistically high

because of difficulties in applying the scoring criteria.

• Difficulties of disproving the existence of key practices, and the failure of the scale to

accommodate a difference between policy and practice, led to the surprisingly high

score.

• The CAIR assessment process was useful in helping to identify organisational

weaknesses, most notably:

§ Dissemination of the findings to clinical staff of incident reports and

investigations

§ Involvement of clinical staff in patient safety discussion at

management level

• In the present form, therefore, the CAIR serves as a useful assessment tool, but is not

a sufficiently robust scoring method to allow comparison between Trusts. Further

scale development could produce an improved scoring method.



11. Incident Reporting Attitude Survey

11.1 Introduction

In this section, the implementation of a tool for assessing the incident reporting culture in

an organisation is described. The Incident Reporting Attitude Survey (IRAS) IRAS is an

anonymously completed questionnaire which was used within the cardiac and

orthopaedic units. It is composed of a series statements that respondents to rate their

level of agreement on a five point Likert scale (1=strongly agree to 5=strongly disagree),

each of which relates to a key aspect of adverse event reporting. This allows the

measurement of the attitudes to safety-related items such as “I am likely to be

reprimanded when I admit to making a mistake”, “Directorate managers are not

interested in near misses”, and “Good clinicians will always report their mistakes”. The

full questionnaire is given in Appendix 8. IRAS includes both positive and negative

statements to reduce structurally-induced response bias, contains nine items which

concern potential reasons why health care staff may not report adverse incidents and

also contains questions concerning job type, department, length of experience, length of

time in the current department, date of qualification and gender.

Since this assessed the attitudes of practitioners, whereas the CAIR assessed the

management policies and processes, some overlap between the two was expected.

11.2 Method

IRAS was distributed in the paediatric cardiac unit following a short initial presentation at

one of the regular clinical governance meetings, and in the orthopaedic unit after

presentations were made at operating theatre staff and anaesthetic department

meetings. These meetings were known to be well supported and well attended. The

IRAS questionnaires were targeted at all staff working in the each surgical unit, both

clinical and non-clinical. Collection boxes were placed in the cardiac reception unit and in

the operating theatre coffee rooms at both sites. As an incentive for encouraging people

to complete the questionnaires a prize draw was arranged by the use of raffle tickets

attached to the front of the questionnaires. To ensure anonymity, the prize was held by a

party independent from the research project.



Items in the questionnaire were first grouped into 5 different themes, with some items

contributing to more than one theme:

• the value of reporting incidents (Items 12, 17,18)

• the personal consequences for those reporting incidents (Items 1,3, 10,13,14,15, 16,

19, 22, 23)

• the value of reporting minor incidents and near misses (Items 4, 5)

• the risk management team and incident investigations (Items 3,10,11)

• the performance of the Directorate/management concerning incidents (Items 2, 6, 7,

8, 9, 14, 20, 21, 24)

For each questionnaire item, the number of respondents that chose each potential

response were counted. Since some questions were negatively positioned, some

statements were reworded in the analysis, and the responses reversed, to allow

appropriate comparisons. Details are provided in Appendix 9.

Responses for each topic theme were summarised by calculating the proportion of

respondents agreeing with each reworded statement and the exact 95% confidence

intervals for these proportions120. All free text comments were reported verbatim.

11.3 Results from Paediatric Cardiac Surgical Centre

Out of approximately 100 staff working in the department, 27 questionnaires were

returned. The section of the questionnaire giving details of the staff member’s job

description, length of service in the Trust and their overall length of career was filled in

satisfactorily by 76% of those who responded. It was apparent that a reasonable cross-

section of clinical and non-clinical staff had participated in the survey. Respondents

included: surgeons (consultants and registrars), anaesthetists (consultants and

registrars), scrub nurses, clinical nurse specialists, perfusionists, operating department

practitioners, ward nurses, cardiologists (consultants and registrars), administration staff

and both clinical and non-clinical management staff.



A detailed breakdown of the responses received is given in Appendix 10. Table 11.1

indicates considerable support amongst the respondents for the concept of incident

reporting, though approximately 40% of respondents did not agree with the statement

"When I report an incident it is never a waste of time." Table 11.2 suggests that it is

recognised within the trust that a culture where staff feel able to admit mistakes without

fear of personal consequences is preferable (items 3, 10, 13, 23). However, the

responses to items 14, 15, 19 indicate that the Trust may not yet have fully developed

such a culture. Few of the respondents agree with the notion that a "no blame" culture is

a utopia (item 22). Table 11.3 shows that there is no consensus amongst the

respondents concerning the management's interest in near misses (item 4) nor

concerning the value of reporting minor incidents (item 5). Table 11.4 indicates that the

respondents view the risk management and incident investigation processes within the

Trust to be focussed on understanding how incidents happen rather than judging the

actions of individuals concerned (items 3, 10 & 11). Table 11.5 indicates that there is

broad agreement that the directorate attempts to foster learning from mistakes and that

the directorate is proactive regarding incident prevention (items 24, 8 & 2). Few

respondents think that communication between staff and directorate managers is free

and easy (items 21 & 7). There seems to be little consensus concerning other areas of

the directorates management of incidents and incident reporting (items 14, 20, 6 & 9).

Table 11.6 concerns potential barriers to incident reporting. These indicate concerns with

the lack of incentive for incident reporting, the complexity of the reporting process, a lack

of action following reports, a lack of feedback and the prospect of getting others into

trouble (items 25 c, d, f, g & i). Also, a majority of respondents agreed that "Some things

are too minor to warrant the extra work" of completing an incident report. Few

respondents indicated a concern over adverse personal consequences of completing an

incident report. Finally, table 11.7 shows the verbatim comments gathered from the

IRAS questionnaire, each of which addresses a slightly different theme.



1009080706050403020100

UPPER

LOWER

MEAN

12 Good clinicians/nurses will always report their mistakes.

17 When I report an incident it is never a waste of my time.

18 Incident reporting is a valuable tool in the drive to continuously improve the safety of the patient.

Reworded questionnaire item

1009080706050403020100

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN





Table 11.1: Att i tudes to the value of report ing incidents (cardiac surgery) The proportion of respondents agreeing with the statement (with exact 95% confidence intervals) for the topics concerning attitudes to the positive value of reporting incidents in the paediatric cardiac unit.

100806040200

UPPER

LOWER

MEAN

1 In the event of an incident, the Directorate does not seek to findthe individual responsible and discipline them.

3 Incident investigations attempt to find out the real causes of an adverse incident, rather than just blame those involved.

10 The risk management team is not there to enforce rules andregulations and does not exist to punish staff that have put patients at risk.

13 Admitting to failure is not like telling the medical world you area bad clinician/nurse.

14 I feel supported when I admit a mistake.

16 Good clinicians/nurses understand they are not perfectand sometimes they make mistakes.

19 Clinicians/nurses are only blamed and disciplined whenthey deserve it.

22 A ‘no blame’ culture is a utopia.

23 Any policy of naming, blaming and shaming is unhelpfuland out of date.

15 I am unlikely to be reprimanded when I admit to making a mistake.


100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN

1 In the event of an incident, the Directorate does not seek to findthe individual responsible and discipline them.











Table 11.2: Adverse personal consequences for report ing incidents (cardiac surgery) The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 2 concerning perceptions that there are adverse personal consequences for reporting incidents in the paediatric cardiac unit.



1009080706050403020100

UPPER

LOWER

MEAN

4 Directorate managers are interested in near misses.

5 Reporting minor incidents (and nearmisses) is not generally perceived asextra work and is done where possible.


1009080706050403020100

UPPER

LOWER

MEAN

1009080706050403020100

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN


5 Reporting minor incidents (and nearmisses) is not generally perceived asextra work and is done where possible.


Table 11.3: The perceived value of report ing minor incidents and near misses (cardiac surgery). The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 3 dealing with perceptions concerning value of reporting minor incidents and near misses in the paediatric cardiac unit.

1009080706050403020100

UPPER

LOWER

MEAN

3 Incident investigations attempt to find out the real causes of an adverse event,rather than just blame those involved.

10 The risk management team is not there to enforce rules and regulations anddoes not exist to punish staff that have put patients at risk.

11 The risk management team are interested in the chain of events that led up to the incident, rather than how any one individual acted.


1009080706050403020100

UPPER

LOWER

MEAN





Table 11.4: Att i tudes to the r isk management team and incident investigations (cardiac surgery) The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 4 concerning attitudes to the risk management team and incident investigation in the paediatric cardiac unit.



100806040200

UPPER

LOWER

MEAN


2 The Directorate is proactive regarding incident prevention.

20 The attitude in this department is that to err is human,to report divine.

21 Communication between clinicians and managers israrely difficult.

24 In this department we are encouraged to disclose ourerrors in order that everyone may profit from any lessons to be learnt.

6 Directorate managers can be trusted to follow up any incidentthat is reported, however minor.

7 Communication between staff and Directorate managersis open and easy.

8 In this department a forum exists to share experiences andlearn from others’ errors.

9 When I report an incident I am confident that action will be taken.


100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN





24 In this department we are encouraged to disclose ourerrors in order that everyone may profit from any lessons to be learnt.

6 Directorate managers can be trusted to follow up any incidentthat is reported, however minor.


8 In this department a forum exists to share experiences andlearn from others’ errors.



Table 11.5: Att i tudes to the performance of the directorate or management concerning incidents (cardiac surgery) . Proportion of respondents agreeing with each statement (with 95% confidence intervals) for the perception the performance of the directorate or management concerning incidents in the paediatric cardiac unit.

100806040200

UPPER

LOWER

MEAN

25a) I will get the blame.

25b) Some things are too minor to warrant the extra work.25c) I cannot be sure that any action will be taken anyway.

25e) I don’t want to blot my copybook.

25f) The reporting method is too complicated and/or time consuming.

25g) I get no feedback.

25h) If the consequences are not serious, no-one need know.

25i) I might get someone else intotrouble.

25d) There is no reward for doing so.

In this department I might hesitate tomake an incident report because:

100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN

LOWER

MEAN










100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN










100806040200

UPPER

LOWER

MEAN

LOWER

MEAN

UPPER

LOWER

MEAN

LOWER

MEAN










Table 11.6: Barriers to incident report ing (cardiac surgery) The proportion of respondents agreeing with each statement regarding the barriers to incident reporting (with 95% confidence intervals) in the paediatric cardiac unit.



“Hospital incident forms not well designed/used, and often used to criticise another department for an incident” “There are not enough multidisciplinary M&M meetings in the hospital” “My responses to a certain extent reflect my ‘feelings’ rather than my felt experiences because I rarely am in the position of incident reporting” “I still have to admit to a feeling that whilst ‘to err is human’ it nevertheless does to a certain extent reflect badly on the individual unintentionally – despite all the attempts to change the ‘blame culture’ this is because patients/families are not so generous with their acceptance of human fallibility” “Its easier to ‘own up’ to ones own errors than identify those of a colleague” “In assessing staff attitudes to incident reporting I feel the dimension used to measure it, i.e. the institution’s position is inadequate as this only addresses part of the problem” Table 11.7: Free text comments (cardiac surgery) .

11.4 Results from the Orthopaedic Surgical Centre

Out of approximately 70 staff working in the Department of Orthopaedic Surgery, 18

questionnaires were returned. The section of the questionnaire giving details of the staff

member’s job description, length of service in the Trust and their overall length of career

was filled in satisfactorily by 84% of those who responded. It was apparent that a

reasonable cross-section of clinical and non-clinical staff had participated in the survey.

Respondents included: surgeons (consultants and registrars), anaesthetists (consultants

and registrars), scrub nurses, operating department practitioners, administration staff and

both clinical and non-clinical management staff. A detailed breakdown of the responses

received is given in Appendix 10.

Table 11.8 suggests support amongst the respondents for the concept of incident

reporting (item 18) although less than 60% of respondents agreed that good

clinicians/nurses will always report their mistakes (item 12) and less than 40% of

respondents agreed with the statement "When I report an incident it is never a waste of

time." (item 17). Table 11.9 shows that it is recognised within the Trust that a culture

where staff feel able to admit mistakes without fear of personal consequences is

preferable (items 3, 10, 13, 23). However, the responses to items 14, 15, 19 indicate that

the Trust may not yet have fully developed such a culture. Less than 40% of the

respondents agree with the notion that a "no blame" culture is a utopia (item 22). Table

11.10 indicates that the majority of respondents perceive that little value is attached to

the reporting of minor incidents and near misses (item 5). There is no consensus

amongst the respondents concerning the Directorate management's interest in near



misses (item 4). Table 11.11 indicates that the respondents view the risk management

and incident investigation processes within the Trust to be focussed on understanding

how incidents happen rather than judging the actions of individuals concerned (items 3,

10 & 11).

As shown in Table 11.12, there seems to be broad consensus amongst the respondents

that the directorate's management of incidents and incident reporting is poor (items 14, 6

& 9). The majority of respondents do not think that the directorate attempts to foster

learning from mistakes nor that the directorate is proactive regarding incident prevention

(items 24, 8 & 2) and few respondents think that communication between staff and

directorate managers is free and easy (items 21 & 7).

Table 11.13 concerns potential barriers to incident reporting. These indicate some

concerns with there being no reward for incident reporting, the complexity of the

reporting process, a lack of action following reports and a lack of feedback (items 25 c, d,

f, g). Few respondents indicated a concern over potential adverse consequences for

themselves or others associated with completing an incident report (items 25 a, e & i).

Although a majority of respondents agreed that "Some things are too minor to warrant

the extra work" of completing an incident report (item 25b) few respondents agreed that

they might hesitate to report an incident because it did not have serious consequences

(item 25 h). Table 2.14 shows the verbatim comments gathered from the IRAS

questionnaire, with each providing a slightly different perspective, and no clear theme.



100806040200

UPPER

LOWER

MEAN





100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN





Table 11.8: Att i tudes to the value of report ing incidents (orthopaedic surgery) The proportion of respondents agreeing with the statements (with exact 95% confidence intervals) concerning attitudes to the value of reporting incidents in the orthopaedic unit.

100806040200

UPPER

LOWER

MEAN

1 In the event of an incident, the Directorate seeks to findthe individual responsible and discipline them.











100806040200

UPPER

LOWER

MEAN

UPPER

LOWER

MEAN

1 In the event of an incident, the Directorate seeks to findthe individual responsible and discipline them.











Table 11.9: Adverse personal consequences for report ing incidents (orthopaedic surgery) The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 2 concerning perceptions that there are adverse personal consequences for reporting incidents in the orthopaedic unit.



100806040200

UPPER LOWERMEAN


5 Reporting minor incidents (and near misses) is not generally perceived as extra work and is done where possible.


100806040200

UPPER LOWERMEAN

UPPER LOWERMEAN


5 Reporting minor incidents (and near misses) is not generally perceived as extra work and is done where possible.


Table 11.10: The perceived value of report ing minor incidents and near misses (orthopaedic surgery). The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 3 dealing with perceptions concerning value of reporting minor incidents and near misses in the orthopaedic unit.

100806040200

UPPER

LOWERMEAN





100806040200

UPPER

LOWERMEAN





Table 11.11: Att i tudes to the r isk management team and incident investigations (orthopaedic surgery) The proportion of respondents agreeing with the statement (with 95% confidence intervals) for the topics in theme 4 concerning attitudes to the risk management team and incident investigation in the orthopaedic unit.



100806040200

UPPER

LOWERMEAN

24 In this department we are encouraged to disclose our errors in order that everyone may profit from any lessons to be learnt.


8 In this department a forum exists to share experiencesand learn from others’ errors.



6 Directorate managers can be trusted to follow up anyincident that is reported, however minor.


2 The Directorate is proactive regarding incidentprevention.



100806040200

UPPER

LOWERMEAN

UPPER

LOWERMEAN



8 In this department a forum exists to share experiencesand learn from others’ errors.



6 Directorate managers can be trusted to follow up anyincident that is reported, however minor.


2 The Directorate is proactive regarding incidentprevention.

14 I feel supported when I admit a mistake.14 I feel supported when I admit a mistake.

21 Communication between clinicians and managers israrely difficult.21 Communication between clinicians and managers israrely difficult.

Table 11 .12: Att i tudes to the performance of the directorate or management concerning incidents (orthopaedic surgery) . Proportion of respondents agreeing with each statement (with 95% confidence intervals) for the perception the performance of the directorate or management concerning incidents in the orthopaedic unit.

100806040200

UPPER

LOWER

MEAN










100806040200

UPPER

LOWER

MEAN

LOWER

MEAN










Table 11.13: Barriers to incident report ing (orthopaedic surgery) The proportion of respondents agreeing with each statement regarding the barriers to incident reporting (with 95% confidence intervals) in the orthopaedic unit.



“NO FEEDBACK. It feels like you might as well put the reporting form straight in the recycling bin.” “I am on the risk management committee and we discuss all forms. However, strategy and action plans are lacking.” “Often when mistakes and mishaps occur it is because we are short staffed, under pressure and stressed. There is not time to stop and fill out forms. At the end of such a shift all you want to do is go home not stay back filling forms and searching for the other staff involved. “It was difficult to answer totally honestly due to the brevity of my time in this current post. Some answers are based over the last 6 years in hospital medicine.” “Lack of equipments – several forms filled, nothing happens e.g. no Tiva pumps, Diprifusors – therefore no Tiva training. Not enough fibroscopes – only one in main theatres.” Table 11.14: Free text comments (orthopaedic surgery).

11.5 Discussion

The response rate to the questionnaire survey of 26-27% was disappointingly low

especially given the attempts that were made to encourage response. Though there was

representation from all the main professional groupings within the department, this low

response rate suggests that this may be an unrepresentative sample. Consequently, we

limit the remaining discussion to general thematic impressions from both surgical units.

In both units there was clearly a recognition of the importance of reporting and learning

from incidents. However, there appeared to be some uncertainty about how the data

would be used by the trust. Respondents in both units thought that reporting an incident

might be a waste of time, and that the information was not used effectively by the

directorate or management. In the paediatric unit there were also concerns that reporting

incidents would either have a negative personal effect, or a negative effect on others.

Fortunately, this was not the case in the orthopaedic unit, though there was some

perception in this unit that incident reporting is not the sign of a good practitioner. It

seems therefore that the barrier to effective incident reporting in both units it is not the

awareness of the value of the data, but rather a perceived lack of trust that the

management will utilise that data appropriately. This reflects our findings from the CAIR,

which also suggested that major gaps existed not only in the dissemination of safety

information, but also in the appropriate involvement of practitioners in the use of that

information.

Given the low response rate, care must be taken with conclusions. However, this

evidence seems to indicate that the perceived failure of the management to use incident



data, or to use it inappropriately, reduces the perceived value of reporting incidents.

Presumably, therefore, better feedback from management to practitioner of incident data

would result in better incident reporting, and so an improvement in safety in both surgical

centres. Assuming appropriate use of the data, this would also assuage the deterring

perception that reporting an incident might get someone else in trouble.

11.6 Summary

Response rates to the IRAS were low (26-27%) in both centres, which makes

conclusions difficult. However, the data suggest a general recognition of the need to

learn from incidents, but a mistrust of the management appropriately to utilise the

information. Improved feedback from management to practitioner is needed to improve

the error reporting culture in both of the organisations studied.



12. Operating Theatre Team Management Attitude Questionnaire

12.1 Introduction

In this section, the piloting of the Operating Theatre Team Management Attitude

Questionnaire (OTTMAQ) in the paediatric cardiac unit and the orthopaedic unit is

described.

The OTTMAQ is an anonymously completed questionnaire which is designed for use in

evaluating the cultural attitudes to safety and error reporting in a surgical population. It is

also sometimes found in a slightly different form as the ORMAQ (Operating Room

Management Attitude Questionnaire). Both derive from a similar tool developed and

utilised in aviation for over a decade, known as the FMAQ (Flight Management Attitude

Questionnaire). The FMAQ was used to assess safety culture between airlines in

different countries, and to assess changes within an airline or culture over time. The

data, which has been used by both safety managers seeking to identify cultural or

organisational problems, and researchers seeking to examine the relationship between

culture and error, has produced increasingly convincing evidence of the value of

encouraging a culture of safety on the flight deck88. A similarly comprehensive data set

has yet to be established in medicine.

The value of the OTTMAQ is in providing insight into the gaps in attitudes to safety that

might exist within a pool of operating theatre teams. It is worth briefly summarising the

results of ORMAQ data collected from 1033 operating theatre and intensive care staff in

the United States, Israel, Germany, Switzerland, and Italy119. The majority of

respondents thought that neither fatigue, personal problems, nor emergencies affected

their performance, and the handling of errors was felt to be inappropriate by two thirds of

the respondents. On the other hand, most believed that the reporting of errors is

important for patient safety. In this study, the OTTMAQ complemented the IRAS and

CAIR by assessing these attitudes towards teamwork, safety and error reporting in the

surgical departments in which our observational studies were conducted.

Firstly, we sought to apply the OTTMAQ to examine whether similar results would be

obtained in the UK to those obtained elsewhere. Secondly, we were interested in



whether it would be possible to measure differences between the two very different

surgical units involved in our studies. Finally, since the data gathered here related both

to the IRAS and CAIR data, and to the observations of team performance described in

Part I, themes identified across the different measures presumably might suggest

appropriate safety solutions.

12.2 Methods

The OTTMAQ was distributed in the same orthopaedic and paediatric cardiac units

which took part in all other studies documented here. The same strategy was adopted as

for the IRAS questionnaire, with both IRAS and OTTMAQ questionnaires being deployed

simultaneously.

The OTTMAQ is divided into five sections. In the first section, respondents are asked to

rate their level of agreement with a series of statements on a five point Likert scale

(1=disagree strongly to 5=agree strongly). Each statement relates to a key aspect of

theatre team culture, for example: “It makes no difference to me with whom I work”, “To

resolve conflicts, team members should openly discuss their differences with each

other”, “I am less effective when stressed or fatigued”. Items in section one were grouped

into 5 different themes, with some items contributing to more than one theme:

• assessment of the respondents own practice (Items 2, 4, 5, 8, 10, 12, 21, 26, 27, 28,

33, 35, 38, 45)

• the strength of hierarchy within teams (Items 1,3, 9, 11, 13, 14, 34, 40, 41, 43)

• expressing concerns (Items 15, 16, 17, 18, 19, 29, 30, 31, 32, 36)

• attitudes towards the Trust (Items 23, 39, 47)

• attitudes to working as part of a team (Items 6, 7, 20, 22, 25, 24, 37, 42, 44, 46, 48)

Section 2 of the questionnaire was concerned with two statements. The first, “How

frequently, in your work environment, are subordinates afraid to express disagreement

with their superiors?”, which is followed by a choice of five responses, ranging from “very



frequently” to “very seldom”. The second statement, “How often do you feel nervous or

tense at work?” was also followed by a five-point response scale, ranging from “always”

to “never”.

Section 3 examined leadership style, asking the respondents to express their favoured

style, and the style they most often experienced. The four exemplar styles were

described as:

Style 1: Usually makes his/her decisions promptly and communicates them to

his/her subordinates clearly and firmly. Expects them to carry out the decisions

loyally and without raising difficulties.

Style 2: Usually makes his/her decisions promptly but, before going ahead, tries

to explain them fully to his/her subordinates. Gives them the reasons for the

decisions and answers whatever questions they may have.

Style 3: Usually consults with his/her subordinates before s/he reaches his/her

decisions. Listens to their advice, considers it, and then announces his/her

decision. S/he then expects all to work loyally to implement it, whether or not it is

in accordance with the advice they gave.

Style 4: Usually calls a meeting of his/her subordinates when there is an

important decision to be made. Puts the problem before the group and invites

discussion. Accepts the majority viewpoint as the decision.

Section 4 of the questionnaire consists of 16 questions concerning the work goals and

aspirations of the respondents. Section 5 invites respondents to comment upon the two

questions “how can the effectiveness of operating theatre teams be increased?” and

“how can the job satisfaction of operating theatre teams be increased?”. The final section

collects demographic data. The full questionnaire is shown in Appendix 11.

Initial analysis was carried out by calculating the frequency and percentage of each

response to each question, following the same procedure as for the IRAS survey. As

with the IRAS, negatively phrased questions were re-worded in the analysis to be



positive, and the responses reversed to reflect this. The reworded statements are given

in Appendix 12. Responses for each topic theme were summarised in identical fashion to

IRAS.

12.3 Results from the Paediatric Cardiac Unit

Out of approximately 50 clinical staff working in the department whose job directly relates

to cardiothoracic surgery, 19 questionnaires were returned. A detailed breakdown of the

responses received is shown in Appendix 13.

The section of the questionnaire giving details of the staff member’s job description,

length of service in the Trust and their overall length of career was filled in satisfactorily

by 74% of those who responded. While the sample size is too small to analyse potential

differences in responses between gender, professional groups or total working time, the

surveys showed that there was an acceptable spread in the employees completing the

questionnaire. Scrub nurses, operating department practitioners, perfusionists, surgeons

and anaesthetists of registrar and consultant grade (ranging from 28 years experience in

the specialty to 1 year’s experience) all participated in the study.

Table 12.1 indicates that responses were diverse for each of the items relating to

respondents' assessment of their own practice. There is broad disagreement with the

statements “I am equally effective when stressed or fatigued” and “I am no more likely to

make errors in tense or hostile situations” (items 33 & 21), which may indicate that staff

generally have realistic expectations of their own performance. The attitude that

emergency situations do not affect performance (item 12), yet an inferior team member

can (item 35), suggests some potentially inconsistent and unrealistic perceptions.

There was consensus amongst the respondents on a number of questions concerning

hierarchy within theatre teams and leadership styles. Table 12.2 shows that respondents

agreed that good communication, including briefing and debriefing is important to team

performance and patient safety (items 13, 14 and 43). It is also clear that the

respondents would not favour a very strong hierarchy within theatre teams where input

from junior team members is discouraged (items 3, 11 & 41). Another interesting finding

is that, whilst the majority of respondents acknowledge that personal problems can affect



his or her performance (item 45), a large majority think that a true professional should be

able to leave personal problems outside the operating theatre (item 40).

Table 12.3 indicates that, in general, the respondents believe that staff should raise

concerns that they have irrespective of hierarchy or professional boundaries (items 16,

17, 18, 19, 29, 30, 31 & 32), though there is a perception that questioning of senior staff

should not be a usual occurrence (item 15). However, the majority of respondents do not

agree with the statement “To resolve conflicts, team members should openly discuss

their differences with each other” (item 36).

Table 12.4 indicates that the respondents have a positive view of their jobs and of

working for the Trust (items 23, 39 & 47). Table 12.5 indicates that the respondents

believe that team working has a valuable and important role to play in the work of the

department (items 6, 7, 20, 22, 24, 25, 37, 44 & 46). However, a minority of respondents

agreed with the statement “The work of each team member is sufficiently appreciated by

the other team members” (item 48). The equivocal and varying responses to the need for

written procedures suggests that this issue is controversial.

Responses to section 2 of the questionnaire, concerned with conflict and stress in the

workplace, are given in table 12.6. Table 12.7 shows the preferences and experiences of

respondents to section 3, which related to leadership style. Of note is the desire for style

four, with no style one, which conflicts with the perception of a considerable amount of

style one, and no style four leadership.

Responses from section 4, relating to work goals and aspirations, are shown in table

12.8. The responses are generally all in considerable agreement, though clearly some of

those sampled preferred routine and predictability in their work far more than others

(items 58 and 64). Finally, tables 12.9 and 12.10 list the verbatim responses to two open-

ended questions regarding the improvement of effectiveness and job satisfaction in

operating theatre teams. The clearest theme to emerge is that of the need for

communication, with teamwork a related theme. Reward and recognition, support from

the management, the need for debriefings, and social activities are also themes that

appear in more than one response.



100806040200

2) It makes no difference to me with whom I work.

5) I do not work better when I operate according to set guidelines.

8) During periods of low work activity, I would rather keep busy with small tasks.

12) My decision making ability is as good in emergencies as in routine situations.

4) Even when fatigued, I perform effectively during critical phasesof operations.

21) I am no more likely to make errors in tense or hostile situations.

45) Personal problems cannot adversely affect my performance.

27) I am ashamed when I make a mistake in front of my otherteam members.

28) In critical situations, I do not rely on my superiors to tell me what to do.

26) If I perceive a problem with the management of a patient, I will speak out, regardless of who might be affected.

10) I let other team members know if my workload is becomingtoo excessive.

33) I am equally effective when stressed or fatigued.

38) I do not become irritated when I have to work with inexperienced medical staff.

35) My performance is not adversely affected by workingwith an inexperienced or less capable team member.

Questionnaire item

100806040200


5) I do not work better when I operate according to set guidelines.



4) Even when fatigued, I perform effectively during critical phasesof operations.




28) In critical situations, I do not rely on my superiors to tell me what to do.






Questionnaire item

Table 12.1: The respondents own practice (cardiac surgery) Proportion of respondents agreeing with the statement (with exact 95% confidence intervals) for the questions associated with the respondents own practice.

100806040200

1) The senior person, if available, should take over and make all decisions in life-threatening emergencies.

3) Senior staff should not encourage questions from junior medical and nursing staff during operations, even if appropriate.

9) Senior staff deserve extra benefits and privileges.

11) Doctors who encourage suggestions from operating team members are weak leaders.

13) A regular debriefing of procedures and decisions after an operating session or shift is an important part of developing and maintaining effective crew coordination.

34) Leadership of the operating theatre should rest with the medical staff.

40) A truly professional operating theatre team member can leave personal problems behind when working in the operating theatre.41) There are no circumstances where a junior team member should assume control of patient management.43) Good communication and team coordination are as important as technical proficiency for operationalefficiency and patient safety.

14) Team members in charge should verbalise plans for proceduresor actions and should make sure that the information is understood and acknowledged by the others.

Questionnaire item

100806040200

1) The senior person, if available, should take over and make all decisions in life-threatening emergencies.

3) Senior staff should not encourage questions from junior medical and nursing staff during operations, even if appropriate.



13) A regular debriefing of procedures and decisions after an operating session or shift is an important part of developing and maintaining effective crew coordination.




Questionnaire item

Table 12.2: Hierarchy and leadership in teams (cardiac surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with team hierarchies and leadership.

3) Senior staff should encourage questions from junior medical and nursing staff if appropriate



100806040200

15) Junior operating team members should not usually question the decisions made by senior personnel.

16) Anaesthetists should express concerns about actions by surgeons when they have doubts.

17) Surgeons should express concerns about actions by anaesthetists when they have doubts.

18) Scrub nurses should express concerns about actions by anaesthetists when they have doubts.

19) It is not better to agree with other operating theatre team members than voice a different opinion.

29) Operating team members should not be afraid to question the decisions or actions of senior staff except when they threaten the safety of the operation.

30) Anaesthetists should feel able to question the actions of other anaesthetists in front of team members.

31) Surgeons should feel able to question the actions of other surgeons in front of team members.

32) Scrub nurses should feel able to question the actions of surgeons.

36) To resolve conflicts, team members should openly discuss their differences with each other.

Questionnaire tem

100806040200

15) Junior operating team members should not usually question the decisions made by senior personnel.










Questionnaire tem

Table 12.3: Expressing concern at the actions of other team members

(cardiac surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with expressing concern at the actions of other team members .

100806040200

23) Working for this hospital is likebeing part of a large family.

39) I am proud to work for thishospital.

47) I like my job.

Questionnaire item

100806040200



47) I like my job.

Questionnaire item

Table 12.4: Job and workplace satisfaction (cardiac surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with job and workplace satisfaction.



100806040200

6) We should be aware of and sensitive to the personal problems of other operating team members.

7) Morale and productivity would be improved if surgeons, anaesthetists, anaesthetic assistants and scrub nurses consider themselves part of one team.

20) The pre-session Joint Case Conference is important for safety and effective team management.

22) The doctor’s responsibilities include coordination of his/her work team and other support areas.

24) Operating team members share responsibility for prioritising activities in high workload situations.

25) Successful operating theatre management is primarily a function of the doctor’s medical and technical proficiency.37) Operating theatre team members should monitor each other for signs of stress or fatigue.

42) Written procedures are needed for operating theatre situations.

44) The concept of all operating theatre personnel working as a team does work in our hospital.46) Effective operating theatre team coordination requires members to take into account the personalities of other team members.

48) The work of each team member is sufficiently appreciated by the other members of the team.

Questionnaire item

100806040200



20) The pre-session Joint Case Conference is important for safety and effective team management.

22) The doctor’s responsibilities include coordination of his/her work team and other support areas.


25) Successful operating theatre management is primarily a function of the doctor’s medical and technical proficiency.37) Operating theatre team members should monitor each other for signs of stress or fatigue.


44) The concept of all operating theatre personnel working as a team does work in our hospital.46) Effective operating theatre team coordination requires members to take into account the personalities of other team members.


Questionnaire item

Table 12.5: Working as a team (cardiac surgery). Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with working as part of a team. No.

Statement Very frequently

Frequently

Sometimes

Seldom

Very seldom

49. How frequently, in your work environment, are subordinates afraid to express disagreement with their superiors?

2 (10.5%)

2 (10.5%)

12 (63.2%)

1 (5.3%)

5 (10.5%)

No.

Statement Always

Usually

Sometimes

Seldom

Never

50. How often do you feel nervous or tense at work?

0

0

12 (63.2%)

7 (36.8%)

0

Table 12.6: Confl ict and stress at work (cardiac surgery) Number of responses and percentage response for each question from the second section of the OTTMAQ survey. 51. Which one of the above styles of

leadership would you most prefer to employ or experience?

Style 1: 0

Style 2: 6 (31.6%)

Style 3: 7 (36.8%)

Style 4: 6 (31.6%)

52. In your hospital, which one of the above styles do you find yourself most often experiencing?

Style 1: 6 (31.6%)

Style 2: 6 (31.6%)

Style 3: 6 (31.6%)

Style 4: 1 (5.3%)

Table 12.7: Preferred and experienced leadership styles (cardiac surgery) Number of responses and percentage response for each question from the third section of the OTTMAQ survey, concerning leadership styles.



Responses (19 unless indicated) No.

Question Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

53. Maintain good interpersonal relationships with all other operating theatre personnel?

0 1 (5.3%)

0 5 (26.3%)

13 (68.4%)

54. Have an opportunity for advancement to higher level jobs? (18 responses)

0 0 1 (5.6%)

4 (22.2%)

13 (72.2%)

55. Have security of employment? 0 0 0 4 (21%)

15 (79%)

56. Work in an environment where the group’s achievements are valued over your individual success?

0 0 1 (5.3%)

5 (26.3%)

13 (68.4%)

57. Live in an area desirable to you and your family?

0 1 (5.3%)

1 (5.3%)

3 (15.8%)

14 (73.7%)

58. Have a changing work routine with new, unfamiliar tasks?

2 (10.5%)

5 (26.3%)

0 4 (21%)

8 (42.1%)

59. Have the time to consider more than one solution to a problem?

0 0 1 (5.3%)

5 (26.3%)

13 (68.4%)

60. Have a warm relationship with your direct superior?

2 (10.5%)

0 0 8 (42.1%)

9 (47.4%)

61. Have considerable freedom to adopt your own approach to the job?

0 0 3 (15.8%)

6 (31.6%)

10 (52.6%)

62. Have an opportunity for high earnings?

0 0 2 (10.5%)

6 (31.6%)

11 (57.9%)

63. Have challenging tasks to do, from which you get a personal sense of accomplishment? (18 responses)

0 0 1 (5.6%)

3 (16.7%)

14 (77.8%)

64. Know everything about the job, to have no surprises?

1 (5.3%)

7 (36.8%)

2 (10.5%)

5 (26.3%)

4 (21%)

65. Have sufficient time left for your personal or family life?

0 1 (5.3%)

1 (5.3%)

2 (10.5%)

15 (79%)

66. Work with people who co-operate well with one another?

0 0 0 3 (15.8%)

16 (84.2%)

67. Have a job or career that will bring you prestige and recognition from others? (18 responses)

0 1 (5.6%)

3 (16.7%)

7 (38.9%)

7 (38.9%)

Table 12.8: Work goals (cardiac surgery) Number of responses and percentage response for the fourth section of the OTTMAQ survey, concerning work goals.



“How can the effectiveness of operating theatre teams be increased?” “JCC, good communication” “Make anaesthesia and operating room part of same management team with cardiac unit” “1. Communication. 2. Value staff hard work. 3. Respect.” “Better communication. More teamwork. Less moaning. Better managerial support”. “More open communication: anaesthetists rarely involved in pre-operative discussions. Senior involvement in patient preparation”. “Communication and good support, with the multidisciplinary forum”. “Work as a team at the same purpose”. “Good management in booking office department and therefore effective communication management.” “Better communication”. “Better communication at all levels. Integrated decision making”. “Improved communication; go through case with staff beforehand and de-brief after if case was difficult” Table 12.9: Free text responses to the question “How can the effectiveness of operating theatre teams be increased?” “How can the job satisfaction of operating theatre teams be increased?” “Flexible working, team involvement” “Implement briefing/de-briefing. More social time. More staff.” “More humour, improved communication, good joint social activities.” “1. Communication. 2. Value staff hard work. 3. Respect.” “Formal debriefing; value opinions of different members of the team.” “Appreciation for all and more money”. “Can’t”. “A friendly working environment and a pay increase”. “By increasing pay. By increasing effectivity”. “Increased responsibilities. Increased communications. Team decision making taking into account all members of the multi-disciplinary team.” “More appreciation for each other’s efforts and difficulties”. Table 12.10: Free text responses to the question “How can the job satisfact ion of operating theatre teams be increased?”

12.4 Results from the Orthopaedic Centre

Out of approximately 70 clinical staff working in the department, 19 questionnaires were

returned. This represents a response rate of approximately 27%. A detailed breakdown

of the responses received is shown in Appendix 13.

The section of the questionnaire giving details of the staff member’s job description,

length of service in the Trust and their overall length of career was filled in satisfactorily

by 100% of those who responded. It was apparent that a reasonable cross-section of

clinical staff had participated in the survey. Respondents included: surgeons

(consultants and registrars), anaesthetists (consultants and registrars), scrub nurses,



and operating department practitioners. The respondents also encompassed a wide

range of experience, from less than one year working in their current specialty up to one

individual who had accumulated 30 years of practice. Several of the staff members had

spent over ten years working in their particular field.

Table 12.11 indicates that responses were diverse for most of the items relating to

respondents' assessment of their own practice and no clear picture emerges. There is

broad disagreement with the statements “I am equally effective when stressed or

fatigued” and “I am no more likely to make errors in tense or hostile situations” (items 33

& 21), which may indicate that staff have realistic expectations of their own performance.

Respondents indicated that their performance can be influenced by other team members

(item 35), that they like to work with certain people (item 2), and that the lack of the right

people could be a source of irritation (item 38). Again, though stress and fatigue were

clearly recognised to have an influence on performance (item 35), there was less

recognition that emergency situations might also be influential (item 12).

There was consensus amongst the respondents on a number of questions concerning

hierarchy within theatre teams and leadership styles. Table 12.12 shows that the

respondents agreed that good communication, including briefing and debriefing is

important to team performance and patient safety (items 13, 14 and 43). It is also clear

that the respondents would not favour a very strong hierarchy within theatre teams

whereby input from junior team members is discouraged (items 3, 11 & 41). However,

even though there is a recognition that personal problems can affect performance (item

45), there is a widely held view that a true professional should be able to leave personal

problems behind when in the operating theatre (item 40).

Table 12.13 indicates that, in general, the respondents believe that staff should raise

concerns that they have over the actions of others irrespective of hierarchy or

professional boundaries (items 15, 16, 17, 18, 19, 30, 31 & 32). Also, a large majority of

respondents agree that “To resolve conflicts, team members should openly discuss their

differences with each other” (item 36). There does, however, appear to be some fear in

questioning the decisions of senior team members (item 29).



As shown in Table 12.14, a large majority of the respondents have a positive view of

their jobs (item 47). Fewer have a positive view of working for the Trust (items 23 & 39).

Table 12.15 indicates that the respondents believe that team working has a valuable and

important role to play in the work of the department (items 6, 7, 20, 22, 24, 25?, 37, 44 &

46). However, a minority of respondents agreed with the statement “The work of each

team member is sufficiently appreciated by the other team members” (item 48).

The responses to section 2 are given in table 12.16. Table 12.17 shows the preferences

and experiences of respondents to section 3, which related to leadership style. There

seems to be an even spread of preference, but a clear lack of experience of leadership

style 4, which is the least hierarchical of the leadership styles. Responses from section

4, relating to work goals and aspirations, are shown in table 12.18, and display a

considerable uniformity of positive responses. Finally, tables 12.19 and 12.20 list the

verbatim responses to two open-ended questions regarding the improvement of

effectiveness and job satisfaction in operating theatre teams. The themes that are

apparent from both questions relate to teamwork (with one response suggesting a failure

of some anaesthetists to work within the team), reward and incentives, career

progression, better provision of resources (including staff and equipment), the restrictive

nature of some management practices, and other issues associated with organisation,

planning and the timing of operations.




5) I do not work better when I operate according to set guidelin es.



4) Even when fatigued, I perform effectively during critical pha sesof operations.




28) In critical situations, I do not rely on my superiors to tel l me what to do.






Questionnaire item

0 20 40 60 80 100

upperlowermean















Questionnaire item

0 20 40 60 80 100

upperlowermean















Questionnaire item

0 20 40 60 80 100

upperlowermean















Questionnaire item

0 20 40 60 80 100

upperlowermean

Table 12.11: The respondents own pract ice (orthopaedic surgery) Proportion of respondents agreeing with the statement (with exact 95% confidence intervals) for the questions associated with the respondents own practice.

1) The senior person, if available, should take over and make al l decisions in life -threatening emergencies.

3) Senior staff should not encourage questions from junior medical and nursing staff during operations, even if appropriate .



13) A regular debriefing of procedures and decisions after an operating session or shift is an important part of developing an d maintaining effective crew coordination.




Questionnaire item

0 20 40 60 80 100

upperlowermean









Questionnaire item

0 20 40 60 80 100

upperlowermean









Questionnaire item

0 20 40 60 80 100

upperlowermean









Questionnaire item

0 20 40 60 80 100

upperlowermean

Table 12.12: Hierarchy and leadership in teams (orthopaedic surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with team hierarchies and leadership.



15) Junior operating team members should not usually be afraid to questionthe decisions made by senior personnel.










Questionnaire tem

0 20 40 60 80 100

upperlowermean

the decisions made by senior personnel.










Questionnaire tem











Questionnaire tem

0 20 40 60 80 100

upperlowermean











Questionnaire tem











Questionnaire tem

0 20 40 60 80 100

upperlowermean











Questionnaire tem











Questionnaire tem

0 20 40 60 80 100

upperlowermean

Table 12.13: Expressing concern at the actions of other team members

(orthopaedic surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with expressing concern at the actions of other team members .



47) I like my job.

Questionnaire item

0 20 40 60 80 100

upperlowermean



47) I like my job.

Questionnaire item

0 20 40 60 80 100

upperlowermean

Table 12.14: Job and workplace satisfaction (orthopaedic surgery) Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with job and workplace satisfaction.





20) The pre -session Joint Case Conference is important for safety and for effective team management.

22) The doctor ’s responsibilities include coordination of his/her work team and other support areas.


25) Successful operating theatre management is primarily a function of the doctor ’s medical and technical proficiency.

37) Operating theatre team members should monitor each other for signs of stress or fatigue.


44) The concept of all operating theatre personnel working as a team does work in our hospital.

46) Effective operating theatre team coordination requires members to take into account the personalities of other team members.


Questionnaire item

0 20 40 60 80 100

upperlowermean



20) The pre -session Joint Case Conference is important









Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean












Questionnaire item












Questionnaire item

0 20 40 60 80 100

upperlowermean

Table 12.15: Working as a team (orthopeadic surgery). Proportion of respondents agreeing with the statement (with 95% confidence intervals) for the questions associated with working as part of a team. Responses (19 total) No. Statement Very

frequently Frequently

Sometimes

Seldom

Very seldom


4 (21%)

5 (26%)

6 (32%)

3 (16%)

1 (5%)

No. Statement Always

Usually

Sometimes

Seldom

Never


0 2 (10%)

7 (37%)

9 (47%)

1 (5%)

Table 12.16: Confl ict and stress at work (orthopaedic surgery) Number of responses and percentage response for each question from the second section of the OTTMAQ survey. Responses (19 total) 51. Which one of the above styles of


Style 1: 6 (32%)

Style 2: 6 (32%)

Style 3: 4 (21%)

Style 4: 3 (16%)


Style 1: 7 (37%)

Style 2: 5 (26%)

Style 3: 6 (32%)

Style 4: 1 (5%)

Table 12.17: Preferred and experienced leadership styles (orthopaedic surgery) Number of responses and percentage response for each question from the third section of the OTTMAQ survey, concerning leadership styles.



Responses (19 total) No.

Question Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


0 0 0 5 (26%)

14 (74%)

54. Have an opportunity for advancement to higher level jobs?

0 1 (5%)

1 (5%)

1 (5%)

16 (84%)

55. Have security of employment? 0 1 (5%)

1 (5%)

3 (16%)

14 (74%)


1 (5%)

2 (10%)

2 (10%)

6 (32%)

8 (42%)


0 0 0 1 (5%)

18 (95%)


1 (5%)

2 (10%)

5 (26%)

6 (32%)

5 (26%)


0 1 (5%)

2 (10%)

7 (37%)

9 (42%)


0 0 3 (16%)

5 (26%)

11 (58%)


0 1 (5%)

2 (10%)

6 (32%)

10 (53%)


0 0 1 (5%)

3 (16%)

15 (79%)

63. Have challenging tasks to do, from which you get a personal sense of accomplishment?

0 0 1 (5%)

1 (5%)

17 (89%)


1 (5%)

3 (16%)

4 (21%)

6 (32%)

5 (26%)


0 0 0 1 (5%)

18 (95%)


0 0 0 1 (5%)

18 (95%)

67. Have a job or career that will bring you prestige and recognition from others?

0 2 (10%)

3 (16%)

3 (16%)

11 (58%)

Table 12.18: Work goals (orthopaedic surgery) Number of responses and percentage response for the fourth section of the OTTMAQ survey, concerning work goals.



“How can the effectiveness of operating theatre teams be increased?” “Most causes of inefficiency are outside the control of the operating room” “Work as a team, reward team workers” “Good anaesthetists working within the team” “This place requires: 1) dedicated porter/orderly to collect patients on time. 2) Prompt 08:00/08:30 starts” “More staff, more equipment, better relationships” “Communication skills, listening skills, teamwork” “Compulsory individual team briefings prior to start of the list and debriefing at the end of it” “By allowing working conditions to be better. Giving staff a longer lunch break (an additional half an hour) and recognising hard work with praise and support will increase working efficiency. Too much belittling and undervaluing happens – rid this to get a better atmosphere” “Stringent policies” “Better pre-op planning of timing, induction, anaesthetic duration and surgical requirements” “Streamline efficiency, less protocols” “Better communication and more co-operation and consideration between members. Respect for each other and each other’s roles. Support from management” “By everybody being punctual and polite and smiling in the first instance” “By using team work” “More beds, less time-wasting” “By making the atmosphere very friendly and by giving incentive to people who work hard” Tab le 12 .19: Free text responses to the question “How can the effect iveness of operating theatre teams be increased?”



“How can the job satisfaction of operating theatre teams be increased?” “By team work and finishing on time” “Developing and expanding roles in line with capabilities and experience (increased responsibility). Feedback and appreciation for achievements from management. Treated as professionals, e.g. nurses not used as cleaners etc.” “Providing adequate staffing to avoid unnecessary sickness” “More cases and efficiency” “Less following of the book and bureaucratic red tape” “By feeling secure and having the chance for advancement in their career” “Better coffee room allowing for greater social interaction” “Reward incentives” “By providing good working conditions and support network. Allowing opportunities to progress in jobs, and implementing a system of recognition/reward for hard work and good progress.” “Greater appreciation of middle-level managers of the efforts of the workforce. Promotion of ‘bottom-up’ decision making. Less autocratic management style” “More equipment. More staff. Resources to finance new technologies. More salaries and private practice” “To be valued by management in terms of: 1) public recognition. 2) Renumeration. 3) Flexibility of rotas. 4) Accommodating staff difficulties which require time off” “Better communication. Better flexibility. Better role definitions and guidelines. More time for social activities” “Defined teams. Designated theatres and nurses” “Better pay” “Hard work should be well rewarded!” “Improved environment. Better HDU/ICU facilities” “Acknowledge other team members. Appraisals. Incentives, i.e. bonuses” Table 12.20: Free text responses to the question “How can the job sat isfact ion of operating theatre teams be increased?”

12.5 Discussion

Again, the response rate to the questionnaire survey in the orthopaedic unit was

disappointingly low given the attempts made to recruit respondents. However, the 38%

response rate in the paediatric cardiac unit was encouraging, as was the representation

from all main professional groupings within both units. With the caveat that this is not

large sample of the population, general thematic impressions from both surgical units

may be identified.

In contrast with previously reported work, the detrimental effect of individual stressors –

most notably fatigue – on performance was recognised. In orthopaedics there was

perhaps a greater diversity of opinion related to the perception of personal practice, but

attitudes in both units were generally very similar. There were, however, questionable

perceptions. The attitude that personal problems may affect performance, but can be left



outside operating theatre probably reflects the misplaced view that surgical teams should

be “able to cope with anything”. The second questionable perception was that junior

team members affect performance, but emergencies do not. This would be consistent

with a prevalent tendency to blame, or fear being blamed, following intraoperative errors

that occur even in the most difficult of circumstances. It was unfortunate in the

orthopaedic unit that there was a fear of questioning senior members of the team, given

that asking questions serves as both an important teaching mechanism, and a useful

safety mechanism. However, the general recognition of the need to encourage

participation from all members of the team in both centres demonstrated a good

understanding of the role of teamwork in safety.

Amongst respondents there was a clear difference between desired leadership style and

leadership experienced by team members. In the paediatric cardiac unit, style 4, the

least hierarchical leadership style, was desired by almost one-third of the respondents,

and style 1, the most hierarchical, was not preferred by any respondent. In contrast, the

leadership experienced was far more frequently reported as style 1 (steep hierarchy)

than style 4 (shallow hierarchy). Though it was apparent that a high proportion of the

orthopaedic respondents also experienced a great deal of style 1 leadership, and very

little style 4, this in general reflected the preferred leadership style in this unit. Given that

the process was led by different members (anaesthetist, surgical registrar, consultant

surgeon and perfusionist) at different times over the course of an operation, and the

utilisation of a range of expertise was important for a successful operation, it was natural

that a less hierarchical team structure within the paediatric cardiac operating team was

perceived as being desirable. Unfortunately, a few respondents felt that this ideal was

not always experienced. Though the steep hierarchy both preferred and experienced in

orthopaedics was perhaps more appropriate given the limited requirements for the

participation of a range of experts for successful completion of the surgery, the steep

hierarchy places more reliance on one potentially fallible individual, and can reduce the

ability of the team to identify and recover from errors (section 6.9). It is therefore not

always preferable, particularly during periods of increased uncertainty. The complex

relationship between performance, safety and hierarchies in relation to team composition

and task complexity limits any further speculation. Further pursuit of these issues would

require consideration of the perceived ideal hierarchy, the hierarchy experienced, and

the most desirable configuration in relation to other measures of safety and performance.



It was noteworthy that the free-text answers in both surgical centres reflected

independently the findings from Part I of these studies, which suggested communication

and equipment problems regularly influence intraoperative performance. Communication

improvement was noted as being the most common recommendation for improving

patient safety, as has been found in other operating theatres.119 The need for good

teamwork was also recognised in this section, though since this is a clear theme in the

OTTMAQ it undoubtedly primed some of the respondents.

Despite demonstrating some variability in understanding the role that teamwork can play

in patient safety, the attitudes to safety and teamwork found here reflect favourably upon

both the units when compared with the previously reported study. Given the generally

positive attitudes to safety and error management already examined though the CAIR

and IRAS, this is perhaps unsurprising. The general discussion in the next section will

examine this combined evidence in more detail.

12.6 Summary

Response rates to the QTTMAQ were again low in the orthopaedic centre (27%) which

makes conclusions difficult. A response rate of 38% in the paediatric cardiac centre was

more encouraging, but still precludes any firm conclusions. While the results compare

favourably with similar studies conducted internationally, and reflect realistic attitudes to

safety and teamwork, there were two clear gaps in the perception of safety:

• The attitude that personal problems affect performance, but can be left outside

the operating theatre by true professionals is flawed.

• The attitude that emergency situations do not affect performance, but junior team

members can, is also flawed.



13. General Findings from the Assessment of Organisational Safety Health

The CAIR, IRAS and OTTMAQ were all successfully deployed to assess organisational

safety health. Due to the time-consuming task of gathering data, it was possible to

deploy the CAIR only at the paediatric cardiac surgery centre. The IRAS and OTTMAQ

were used in both surgical centres, though only approximately one quarter of the

potential respondents completed either questionnaire in the orthopaedic centre.

Response to the IRAS in the paediatric cardiac centre was similar, though a greater

proportion responded to the OTTMAQ. This was disappointing given the additional effort

expended in the deployment of the surveys in the orthopaedic centre following the low

response rate in the paediatric cardiac centre. It is worth considering why the response

rates were low, and coupling the key themes that emerged with evidence from other

parts of our studies.

Though these response rates reflect below average response for similar studies119, they

are certainly within the range that might be expected. There may be many reasons for

this. IRAS and OTTMAQ can take as much as 25 minutes each to complete, which may

have been too long for the target population. Issuing the two questionnaires at different

times might also have helped, but would have made it more difficult to consider the two

questionnaires as representing the same population. The uncomfortable subject matter

of the questionnaires, and the culturally specific nature of some of the language used

("..to err is human, to report divine.", "I don't want to blot my copybook", "A 'no blame'

culture is a utopia" sic) may have also acted as an obstacle to a high response rate. The

language in particular would deserve attention before utilisation of the IRAS and

OTTMAQ across the full spectrum of staff in the NHS. Unfortunately, low response rates

are common for questionnaire surveys, which ultimately limits the suitability of this

technique for making comparisons between surgical departments, or between trusts.

Aside from the problems of achieving a high enough response rate, the high level of

agreement between the results from these two very different surgical centres suggests

that these tools may not be differential at Trust or department level. It is also important to

consider that the part of the population who are not familiar with, or may not see the

value of, patient safety research, were less likely to complete the questionnaires. The

respondents were a self-selecting sample already pre-disposed to involve themselves



with patient safety, and the results presented here may reflect more positive attitudes to

safety than would be found with a comprehensive population sample. Consequently,

while these assessment tools may be suitable for comparisons at an international level,

these findings suggests that they may not be useful for the comparison of attitudes on a

smaller scale.

The CAIR and IRAS identified a key deficiency in safety management; failures by

management to use, or at least to be seen to use, reported incidents. In this respect,

differences between policy and practice were also noted. As a result, there was both a

mistrust of the use to which reporting is put, and the feeling that reporting is ineffective. It

is therefore likely that incidents are frequently not reported, which may explain why the

minor failures detailed in Part I were allowed to recur and why the major failures were

not detailed in any event reporting system. Our studies focused on the practitioners, and

consequently reflect their feelings on the effectiveness of management. However, it is

not only the management who are responsible for the current gaps in safety. The

mistrust of management led the healthcare practitioners in one centre to deal with the

near-misses or repeating problems themselves, without involving the perceived ‘red

tape’ of management. While providing timely solutions, this contributed to a lack of

coherence in the management of safety. In particular, there appeared to be little

accumulation of knowledge regarding patient safety, even though a considerable amount

of information was available within the organisation. Within the other centre, the divide

between management and practitioners was more pronounced.

The organisational division between healthcare practitioners and management may

contribute to the poor implementation of patient safety. Management is largely unaware

of the implications of their decisions upon patient safety, because practitioners are

reluctant to report or discouraged from reporting the consequences of those decisions.

At the least, presentation of safety-related data by management would result in the more

appropriate reporting and use of safety-related knowledge. However, attendance at

those meetings by clinical practitioners would also be needed, and would benefit from

audit trails of the problems identified, solutions generated, and their eventual success.

Finally, resources would need to be allocated specifically for these purposes.



The appropriate utilisation of CAIR, IRAS and OTTMAQ can provide a rich set of data on

safety management and the cultures of safety found in a healthcare organisation.

Unfortunately, the labour-intensive nature of their application, and the low response rate

from safety questionnaires may limit their widespread utility as a comparator between

trusts. Though this low sample size means caution must be urged in the consideration of

the data collected here, current deficiencies in safety management in the healthcare

trusts examined here largely related to an uneasy relationship between healthcare

management and healthcare practitioners. These issues might be helped by

improvements in the involvement of healthcare practitioners in the analysis and feedback

of incident reports, which would lead to more comprehensive patient safety related

reporting from practitioner to management.

13.1 Summary

• The Checklist for Assessing Institutional Resilience (CAIR) was successfully piloted in

a department of paediatric cardiac surgery.

• The Incident Reporting and Attitude Survey (IRAS) and the Operating Theatre Team

Management Attitudes Questionnaire (OTTMAQ) were successfully deployed in

departments of paediatric cardiac surgery orthopaedic surgery at different UK

hospitals.

• While the combination of CAIR, IRAS and OTTMAQ undoubtedly provides a rich

source of information, in practice it may prove an unwieldy tool for making such an

assessment.

o Data collection required for the CAIR was intensive. Identifying and

obtaining copies of relevant policies proved difficult, and arranging and

conducting the necessary interviews was time consuming. Difficulties with

the present scoring method were identified.

o Low response rates were observed despite considerable efforts to

encourage staff to participate. This limited the conclusions that could be

drawn from the collected data.

• Though the small sample size requires caution with the conclusions, deficiencies were

identified in safety management related to an uneasy relationship between healthcare



management and healthcare practitioners. Practitioners felt that information regarding

incident reports would either be inappropriately used or not used at all.

• The OTTMAQ results compare favourably with similar studies conducted

internationally, and reflect realistic attitudes to safety and teamwork. However, staff

failed to understand some important aspects of safety and performance.

• The suggested solution is to involve healthcare practitioners more in the analysis and

feedback of incident reports. This might also lead to more comprehensive patient

safety related reporting from practitioner to management.



PART III: MATHEMATICAL MODELLING OF THE IMPACT OF LATENT

SYSTEM CONDITIONS ON PATIENT SAFETY

14. Introduction

14.1 General Introduction

Simulation modelling has a rich history in relation to health care, and a broad scope for

application to patient safety. A range of issues within health care have been addressed

previously using simulation modelling121-133 and work related to errors in the health care

process has been pioneered in relation to pharmacy errors134-137. In this section of the

report the application of Operational Research techniques to examine linkages between

latent conditions and patient safety in surgery is addressed. An introduction to computer

based simulation modelling is provided in this section, followed in subsequent sections

by a description of the development of mathematical models to examine these linkages

and to enable the potential impact of latent conditions on patient safety to be explored.

14.2 Simulation Modelling for Analysing Complex Systems

An Operational Research view of health care systems draws an analogy between clinical

processes and industrial processing. Though hospitals are staffed by a dedicated work

force who labour long hours tending for the sick, in an attempt to understand the key

processes and their interaction, the Operational Researcher sets aside these human and

emotional issues, and considers a hospital as a ‘factory’ that processes ‘widgets’ (known

as patients). Widgets of many sorts are delivered to the “factory”, where they undergo a

range of tests and processes in a variety of sites and most are eventually dispatched.

When considering quality of care, it is as unrealistic to expect a hospital to eradicate all

instances of substandard care or miscommunication as it is for the factory manager to

spend his entire operating budget to ensure at all washers leaving the plant are precisely

the same size. Investment in quality must be weighed against other demands. In the

case of the factory, washers must be produced and sold at a rate that makes the

business viable. Elaborate quality control procedures may well improve the

standardisation of the product, but if this is costly, or slows production unduly, then more

rudimentary quality control measures might, quite rightly, be adopted. Since the

processes in a hospital may be complex, the analogy with an industrial processing plant



provides a useful method to consider the a range of interactions between elements of

the health care process, and how they can influence system performance. An approach

that accounts for many inter-related factors when considering one element of a process

is helpful in establishing optimal system configuration. Unfortunately, the investigation of

industrial processes is not amendable to examination by controlled trials and these are

rarely used. Simulation provides an alternative, lower cost, method for examination of

the operation of a complex system.

Computer simulation is a method for predicting system behaviour using a mathematical

model of the interaction between key components. In industrial manufacturing,

processes are undertaken on a variety of components as they are fashioned, finished

and then assembled. Simulation is a method by which flows of components (or widgets)

and delays in the system can be forecast. The effect of changing, for example, the

number or efficiency of components at different points of the process can then be

predicted. Computer simulation therefore provides the facility to examine the impact of

different system configurations prior to making those changes, effectively serving as a

laboratory for examining the consequences of particular interventions. Recording the

effects on hypothetical widgets can provide insight into the behaviour of the real system.

The effects of potential changes that might be made in a process (for example,

introducing additional intensive care beds) can be investigated by observing the effects

of such changes within the simulation, thus estimating how a real-life system would

behave if analogous changes were made.

Simulation can be thought of as having three main components:

• a modelled representation of the key components of a sequential system for

processing ‘widgets’ (washers, engine parts or patients) and the pathways that such

widgets may take through this system, passing between successive “work centres”

(processing machines, inspection points, wards, theatre and recovery.)

• the generation and accumulation of information from a large number of hypothetical

processing journeys of different widgets/patients during the course of their

processing. Each journey is simulated, using a computer model, according to a set of

rules that reflect the nature of the operation of the system as a whole and of individual

processes. Particularly important are issues such as the variation in processing time



between different widgets (or “work items” as they are called in simulation

terminology) and bottle necks, whereby processing of one work item cannot

commence at a particular work centre until processing of the previous work item has

finished.

• assembly of summary information reflecting overall system performance, work item

throughput and delays at each work centre within the system.

This simulation philosophy is illustrated in the schematic diagram shown in Figure 14.1.

Patients are regarded as being processed in the same way that a widget factory would

operate. They pass through a succession of physical locations where various clinical

processes are carried out. In the simulation, each of these phases is described as a

work centre. In addition to considering flows, delays and bottlenecks, simulation can also

take account of other factors, illustrated in Figure 14.2, such as the different teams of

health care professionals who care for a patient at different phases of the patient

journey. To simulate this further detail, each work centre may be thought of as having a

number of characteristics. Although a simple example of a simulation, already this

demonstrates a usefully structured model. For example, this might allow a prediction of

delays in a system that resulted from a nurse shortage. The simulation would reflect this

by having rules that ‘queue’ a patient (the work item) before allowing progress into a

particular health care process (the work centre) unless all appropriate staff were present.

The presence of staff could be modelled simply as a chance event whose likelihood

depends on the number of nurses employed. Thus, a computer model that includes the

structures, detailed modelling of components, and knowledge of the system being

modelled will provide a facility to examine these complex interactions.



Figure 14.1: Patient flow schema. A schematic diagram showing the flows of patients (widgets of work items) passing through various stages of treatment (regarded as work centres).

Figure 14.2: Patient contacts with successive clinical care teams during each phase of the health care process. The numbers indicate different phase of the process (work centre), and red lines illustrate the involvement of different members of the care team.



Since simulations can be extremely detailed and difficult to construct, modern

generations of simulation software allow complex systems to be modelled by building,

testing and analysing sub-models of different parts of overall system prior to bringing the

different parts together. This allows the display and manipulation of different levels of

detail in a model. In the example of a health care process shown in Figure 14.1, the

lowest level of detail might be a single box labelled ‘hospital’ and the graphical display

might illustrate a queue of patients waiting for admission, a set of patient icons for those

in hospital and another set representing those discharged. The next level of detail might

have a display akin to that shown in Figure 14.2, with different patients represented at

different stages of their care journey. Even more detailed levels of an individual

simulation module can be used to represent one of the care processes. To facilitate the

construction of these complex modular structures, simulation software packages

commonly have sophisticated graphical interfaces. The facility to see a visual

presentation of the structure of the simulation greatly facilitates discussion of model with

other model builders and, crucially, non-analytical collaborators such as healthcare

professionals. Since an appropriate structure is key to the success of the model, this

combination of modular structure and graphical presentation in modern simulation

software packages offers considerable enhancements over older technology.

Simulation models offer considerable scope for exploring the effects of variability under a

wide range of assumptions. It is common in the modelling of health care that work items

are heterogeneous and there is unpredictable variability in the performance of work

centres. For example, the time taken to complete a particular diagnostic process can

vary independently of the patient. Such variability is usually taken into account by

assuming that chance is a factor that governs the operation of some of the work centres.

Computer simulation models allow the designer to specify different forms of probability

distribution associated with processing time. In the health care example illustrated in

Figure 14.1, variability in the number of days that a patient spends in intensive care can

be modelled in a variety of ways including assuming no variability at all; assuming a

normal distribution of length of stay; or assuming a log normal distribution of length of

stay. In simulation, therefore, it is possible to examine multiple theories and assumptions

regarding the probability of different events.



Care needs to be taken in the development of a computer simulation. Models should

reflect only the essential features of a system’s operation. However, since modern

simulation tools allow processes to be represented in great detail, it can be tempting to

use this detail when inappropriate to do so. Effort expended to build detail into a model

can make it unwieldy and over-complex, rather than ensuring a higher quality model. A

further problem of designing simulations that are too detailed is that they can be difficult

to calibrate. Every feature that is added to a simulation adds parameters whose values

need to be estimated before the simulation can be judged to be credible. The simple

health care model shown in Figure 14.1 would require 6 parameters (total inflow and the

mean time spent in each care process). The model could be made more detailed by

assuming different elective admission rates for each day of the week (7 parameters)

different emergency admission rate of each day of the week (7 parameters), and a log

normal distribution of processing time for each sub-process (10 parameters). This

modest extension results in a simulation model which requires the estimation of 24

parameters most of which are not likely to be known without extensive observational

studies. The power of modern simulation tools means that it is possible to become

engrossed in the details of model building. Consequently, when using Operational

Research methods, it is essential to concentrate on what the practical problem is before

deciding what particular analysis technique should be used.

14.3 Summary

Computer simulation allows the examination of the reaction of a complex system to

particular configurations, without the expensive process of implementing and monitoring

changes in the real system. This means that large numbers of options can be tested,

with only the best alternatives selected for further work. Modern simulation tools are

both a powerful and flexible tool for the analysis and presentation of complex operational

processes which would not be amenable to other forms of modelling. Given the power of

simulation as a research tool, it is also advisable to be aware of potential difficulties that

simulation can lead to. Appropriate expertise and care ensures that these potential

pitfalls can successfully be negotiated.



15. Modelling of Errors in Surgical Care Processes

15.1 Concepts Underpinning the Model Structure

The modelling framework adopted was based on extending the simple notion of viewing

the hospital process as an analogous to an industrial process. The patient journey can

be thought of as a series of phases of care (Figure 14.1), each associated with a specific

location, a specific mode of care and an associated care team. Within the simulation

model, each of these phases are represented by work centres. In reality, there is

considerably variability both in terms of inter-patient differences in the course that the

journey takes and in terms of the planned and unplanned events that occur during its

course. This is reflected in the way in which the work items, the basic building blocks of

the simulation, are used to represent qualities and quantities associated with individual

patients.

In order to model a complex system, such as the surgical patient pathway in a busy

hospital, it was crucial that the researcher constructing the model became familiar with

the role of key components of the system and how these components interacted. This

familiarisation process, undertaken for the purposes of exploring the potential role of

simulation modelling in addressing issues concerning patient safety, overlapped with the

gathering of evidence to complete the CAIR as described in section 10. While mapping

out the various stages that comprise the care pathways followed by congenital heart

surgery patients, particular attention was paid to identifying opportunities for error in the

system that might constitute latent threats to the safety of the patient.

The focus in the early stage of model development was on individual patients and their

movement through the health care process, but this proved to be restrictive. Although

the patient and the patient’s physical location within the hospital system are of

importance, there are many other patient-related entities that are processed within the

hospital system. Some of these are physical entities (e.g. a blood sample; a set of notes;

an X-ray), while some are more abstract and may be thought of as information (e.g. a

recorded blood type; a working diagnosis written in the notes; a radiologist’s report). It

was convenient to think of both of these as representing generalised information. Errors

associated with the processing of such generalised information was a main focus of the

simulation.



Combining the patient pathway and generalised information approaches, the view of the

patient was expanded from that of purely a physical entity, to include properties of the

patient (blood type, anatomical abnormalities, physiological status) which represent the

information available within the system reflecting beliefs or knowledge about the patient.

The model proposed that some of this information may be wrong (for example a blood

type may have been written down incorrectly), but to the individuals in the system it

would appear to be correct. The list of attributes associated with a patient can be lengthy

and may reflect generalised information of several sorts including:

• the true status of the patient in terms of this aspect of information (e.g. patient

blood type is O)

• information not yet investigated (blood type unknown);

• information investigated but not yet incorporated in patient’s notes;

• information as recorded in the notes (perhaps wrongly entered as blood type A);

• information confirming the availability or otherwise of sufficient cross-matched

blood.

15.2 Mapping the Patient Pathway for Congenital Heart Surgery

The patient journey was now viewed in a more complex way than the simple sequence

of events depicted in Figure 14.1. While the entity representing the physical body of the

patient may progress between different sites within the hospital, entities representing

other generalised information may appear elsewhere and undergo separate activities in

locations remote from the patient. Figure 15.1 shows the patient pathway as a collection

of industrial processes, with the patient represented as a widget that progresses in

space and time. This is the fundamental diagram for the modelling process as it captures

the principal processes in the patient’s journey through the system. The patient

advances through the system and moves to different physical locations (for example

from the ward to the operating theatre), and along the pathway, generalised information

concerning the patient is compiled. This information is transferred between departments,

sometimes independently of the patient (for example, a sample of blood acquired from

the patient and sent to the laboratory for analysis) and sometimes linked to the patient

(such as x-ray films that accompany the patient from the x-ray department back to the

ward). Each of these activities (e.g. cross-matching, X-ray reporting etc) was

represented as a work centre, and in the terms of the simulation, the effect of such work



centres was to change entries in the list of information associated with the patient. This

process of transforming information can incorporate delays, information loss or recording

errors or omissions, and were simulated as being made at pre-specified rates.

The GOSH Cardiac Surgery Patient Pathway

Pre-admission

Admission

Anaesthesia

Surgery

CICU

HDU

Review

Discharge

• Discussion with family, surgeon, cardiacliaison sister, cardiologist (translator)

• Medical tests


• Medical tests• Ward round• Discussion at JCC

• Transfer from ward to induction room• Induction• Transfer from induction room to theatre

• Surgical procedure• Transfer to CICU• Theatre to CICU handover


• Medical tests• Ward round


Pre-admission

Admission

Anaesthesia

Surgery

CICU

HDU

Review

Discharge


• Medical tests








Pre-admission

Admission

Anaesthesia

Surgery

CICU

HDU

Review

Discharge


• Medical tests








Pre-admission

Admission

Anaesthesia

Surgery

CICU

HDU

Review

Discharge


• Medical tests







Figure 15.1: The patient pathway as a series of industrial processes. Department to department and team-to-team interfaces were then modelled along the

pathway. Interfaces between different departments and care teams represent

weaknesses in the system because all of the important information must be transferred

to another physical location and also to different staff teams. Consider the situation

where the patient has been nursed on the ward for a few days prior to having their

operation. In this circumstance, the ward acts as a repository for information associated

with the patient, and all the main sources of information, such as notes, x-rays and hard

copies of blood test results, will be located here. In principle, this should imply that such

information is less likely to be lost. The nursing teams will be familiar with aspects of the

patient’s care and condition, and such information will be distributed to other members of

the ward staff both formally, during nursing handovers and team briefings and informally,

via ad-hoc discussions. Furthermore, the patient’s medical, surgical and anaesthetic

teams will visit the patient on the ward, and details of the changes to the patient’s care,



for example, will be communicated directly to the nursing staff. This information is then

circulated across the nursing teams, as described above. Figure 15.2 represents these

communication interfaces.


Pre-admission Admission Anaesthesia Surgery CICU HDUCardiac liaison sisterCardiology SHOPlay therapistEcho technicianRadiographerPre-admission nursesCardiology techniciansCardiology SpR

Cardiac liaison sisterPlay therapistEcho technicianRadiographerPre-admission nursesCardiology techniciansCardiology SHOCardiology SpRConsultant cardiologistConsultant surgeonSurgical SpRSurgical SHOPhysiotherapistConsultant anaesthetistAnaesthetics SpRPerfusionistStaff nurses

Consultant anaesthetistAnaesthetics SpRPerfusionistStaff nursesTheatre nursesPortersODA’s

Consultant anaesthetistAnaesthetics SpRPerfusionistTheatre nursesODA’sConsultant surgeonSurgical SpRSurgical SHO

CICU nursesConsultant intensivistsSpR intensivistsConsultant cardiologistsSpR cardiologistsConsultant anaesthetistsAnaesthetic SpRConsultant surgeonSurgical SpRSurgical SHOPerfusionist

Cardiac liaison sisterPlay therapistEcho technicianRadiographerCardiology techniciansCardiology SHOCardiology SpRConsultant cardiologistConsultant surgeonSurgical SpRSurgical SHOPhysiotherapistStaff nurses

Cardiac day unitLadybird WardEcho departmentX-ray departmentETT test roomOther departments (e.g. nuclear medicine)


Cardiac day unitLadybird WardTheatre – induction roomTheatre – operating room

Theatre – operating room CICU Cardiac day unitLadybird WardEcho departmentX-ray departmentETT test roomOther departments (e.g. nuclear medicine)

The red region represents a staff-staff or department-department interface





































Figure 15.2: Diagrammatical representation of communication interfaces in the cardiac patient pathway.

The paediatric cardiac surgery centre studied was a tertiary referral centre with no

accident and emergency department, so all admissions were referrals from either other

hospitals or clinics, including emergency admissions, or arranged transfers from other

hospital Trusts. The flowchart shown in figure 15.3 is a simplification of the pathway for

an elective paediatric cardiac surgery patient. The diagram is based upon the standard

structure of diagnose, evaluate, test and commence therapy. Details of the processes at

each stage are given in Appendix 14. To complement the information displayed in figure

15.3, figure 15.4 illustrates the main links via which information is exchanged at different

stages of the patient’s journey. The information network shown relates to an elective



patient that has been through the pre-admission process. The detail of the information

exchange at each stage is given in appendix 15.

Pre-admission procedures

External Transfer or Referral

WARD Discharge

X-rayBloodsECGEchoToEStress EchoEEGETTMRICTBaseline obs

CATHETER:diagnostic

CATHETER:intervention

Recovery

Theatre

ICU/CCU

HDU

Recovery

Cardiac Surgey Pathway:Patient Flowchart

CICU

HDU

Pre-admission procedures

External Transfer or Referral

WARD Discharge

X-rayBloodsECGEchoToEStress EchoEEGETTMRICTBaseline obs

CATHETER:diagnostic

CATHETER:intervention

Recovery

Theatre

ICU/CCU

HDU

Recovery

Cardiac Surgey Pathway:Patient Flowchart

CICU

HDU

Figure 15.3: Descriptive flowchart for the progress of patients through the cardiac pathway.



Consultant’s secretary

Nurse in charge of ward

Patient’s bedside/designated nurse

Nursing handover Ward roundsSpecialist nurse

Non-ward test unit staff (e.g. MRI)

Anaesthetics team

Nurse in charge of CICU

ICU-specific support staff (e.g.perfusionist)

ADMISSION

WARD

THEATRE

CICU

WARD

DISCHARGE

Booking Office

Bed manager (designatedSister/Charge Nurse)


Individual medical/surgical team meetings (e.g. pre-op briefings)

Joint Cardiac Conference

Consultant’ssecretary

Theatre Manager

Theatre support teams:ODP’s, theatre nurses, perfusionist, etc)

Specialist nurse andother support services(e.g. PALS, physiotherapist)



Patient’s bedside/designated nurseNursing handover

Ward rounds

Non-CICU test staff (e.g. MRI)

Discussion with:cardiologist, intensivist,

surgeon

Discharge co-ordinator

Paediatric Cardiac Surgey – Patient Pathway Information Network




Nursing handover Ward roundsSpecialist nurse

Non-ward test unit staff (e.g. MRI)

Anaesthetics team

Nurse in charge of CICU

ICU-specific support staff (e.g.perfusionist)

ADMISSION

WARD

THEATRE

CICU

WARD

DISCHARGE

Booking Office

Bed manager (designatedSister/Charge Nurse)


Individual medical/surgical team meetings (e.g. pre-op briefings)

Joint Cardiac Conference

Consultant’ssecretary

Theatre Manager

Theatre support teams:ODP’s, theatre nurses, perfusionist, etc)

Specialist nurse andother support services(e.g. PALS, physiotherapist)



Patient’s bedside/designated nurseNursing handover

Ward rounds

Non-CICU test staff (e.g. MRI)

Discussion with:cardiologist, intensivist,

surgeon

Discharge co-ordinator

Paediatric Cardiac Surgey – Patient Pathway Information Network

Figure 15.4: A chart of how information flows through the different team members and departments in the cardiac patient pathway.

15.3 Characterising the Paediatric Heart Surgery Patient

The patient was viewed as having a list of attributes which are used to reflect both

properties of the actual patient and generalised information held in the system

concerning that patient, some of which may be erroneous. Within simulation terminology,

these attributes are referred to as labels, and each can take on a range of meanings . In

order to represent paediatric cardiac surgery patient, the labels that were defined

included:

• Hospital number (which acts as a unique identifier)

• Believed diagnosis (which reflects current information)

• True diagnosis (which reflects true patient status)

• Chromosomal disorder

• Other co-morbidities

• Proposed operation type

• History of previous surgery



• Estimated units of blood required (which is derived from diagnosis)

• True blood type

• Believed blood type

• Cross-match status (reflects which of the cross-matching processes have been

completed)

• True fasting status (which reflects whether patient has been starved)

• Believed fasting status

• Status of preoperative checklist

• Status of wristband details

• Status of pre-op blood samples

• CT status

• MRI status

• Pre-op echo status

• Post-op echo status

• Catheterisation status

15.4 Modelling the occurrence of and recovery from errors

With the basic work centre and work item structures of the model in place, the simulation

focused on the propagation of and recovery from errors. The modelling was not

developed to include all possible errors that have ever occurred or which could ever

occur in the future. Instead, the modelling focused on two classes of failures whose

occurrence is relatively frequent: delays in processing, and the introduction of erroneous

or missing information.

Processing delays and bottlenecks are relatively simple to model in a simulation. Since

all hospital processes were represented as work centres, delays were modelled by

adjusting the parameter used to specify the processing time of a given work centre. This

might occur at random, at a pre-specified rate or in response to the conditions within the

system. The introduction of erroneous or missing information was modelled by allowing

the work centre to transform entries in the list of information associated with the patient.

This reflected the consequences of an error where incorrect information is present or a

given piece of information becomes lost. These were instigated as random events, as

errors occurring in response to system conditions (for example, a monitor being mis-



calibrated), and as errors occurring as random events where the probability of error is

dependent on system conditions (for example, modelling the observation that errors are

more likely at times of high work load).

Within the hospital system, remedial actions may be taken to detect errors that have

occurred. Error trapping activity was modelled using work centres to reflect mechanisms

by which errors are amended. Again, the effect was to change entries in the list of

information associated with a patient. A variety of error trapping mechanisms could be

simulated, including the natural detection of errors (for example, a delayed response

from radiography would prompt a phone call and the missing X-ray report found); the

detection of errors by specific action or system defence (for example, a ward pharmacist

visits the ward to check information); and the detection of error conditions following a

subsequent activity.

15.5 Summary

The pathways and networks represent how the system operates when all goes well. A

key aspect of the simulation is to provide a framework for modelling what happens when

these processes do not run smoothly. The use of this model of the patient pathway, and

consideration of the movement of both patient and patient information along it, meant

that the simulation of the occurrence and recovery from errors in the surgical care

system could be examined.



16. Development and Integration of Sub-Simulations

16.1 Introduction

In line with the usual modular model development processes utilised in simulation, the

software package SIMUL8138 was used to construct a series of sub-simulations. In this

section the sub-simulations are described that were developed, based around the

patient pathway of the paediatric cardiac surgery patient. The patient admission and

discharge sub-simulation is described in detail to introduce key features of the

simulation. The other sub-simulations, relating to bed management, diagnostic

processes, and blood cross-matching are briefly illustrated here, and described in

greater detail in Appendix 16. Finally, the integration of these components into a single

model is described, and the utility of the model is examined.

16.2 Patient Admission and Discharge

The admission and discharge sub-simulation modelled the how the patient moves

through the different stages of the cardiac patient pathway (figure 15.1). The sub-

simulation commences with the pre-admission procedures and ends with patient

discharge. Although there is usually a gap of many weeks between pre-admission and

actual admission, the prototype focused on the mechanics of the pre-admission process

and the potential errors that may occur, rather than model the actual process by which

elective admissions are ordered into their actual admission dates. Hence, in the

simulation the patient progressed from pre-admission directly to the queue for their

admission. Two pathways were modelled in the simulation, one for elective admissions

and the other for emergency cases admitted by referral from other centres or from the

acute transport service. Emergency patients take precedence over the elective patients

and an emergency admission will halt elective admissions to the relevant operating

theatre until the emergency case has been finished. Within the simulation model,

patients enter the system at what are referred to as work entry points. The admission

and discharge sub-simulation has the following work entry points (see also figure 16.1):

• Elective admissions. All of the labels associated with the patient are initialised, and

the patient is flagged as an elective admission. Patients arrive, according to a fixed

inter-arrival time set by the user. All elective patients within the simulation that are



waiting for admission are either at the elective entry point or in the elective admission

queue.

• Emergency admissions. At this entry point, all the labels associated with the patient

are initialised and the patient is denoted as an emergency. Patients arrive randomly

according to a Poisson distribution with the mean inter-arrival time set by the user.

When an emergency admission occurs, the elective admission pathway is halted at

the level of the admission ward work centre. This typically causes a backlog of cases

in the system. Hence, although the study hospital had no accident and emergency

department, the “emergency admissions holding area” queue represents the location

at which emergency surgical admission patients are prepared for theatre.

Figure 16.1: Screenshot of the patient admission and discharge sub-simulation.

Each work centre was paired with an appropriate queue, which together represents a

distinct stage in the patient’s journey through the system. The function of each work

centre within the simulation was to model the operation of the relevant hospital

processes that occur at this stage. This was done by applying a set of rules to modify the

patient labels and to determine the time taken for processes to occur. At a work centre



these rules can be complex and errors in the system can be modelled. The work centres

modelled were:

• Preadmission

• Admission Ward

• The Induction Room

• Cardiac Theatre

• ICU

• High Dependency Unit Ward

• Ward Review

Pre-admission ward

Ward

Induction room

Cardiac theatre

CICU

HDU ward

Pre-admission nurse

Radiographer

Echo technician

Cardiologist

Admission ward nurse

Anaesthetist

ODP

Theatre nurse

Perfusionist

Consultant surgeonSurgical SpRSurgical SHO

Intensivist

Porter

CICU nurse

HDU nurse

Pre-admission ward

Ward

Induction room

Cardiac theatre

CICU

HDU ward

Pre-admission nurse

Radiographer

Echo technician

Cardiologist

Admission ward nurse

Anaesthetist

ODP

Theatre nurse

Perfusionist

Consultant surgeonSurgical SpRSurgical SHO

Intensivist

Porter

CICU nurse

HDU nurse

Figure 16.2: Staff groups associated with each work centre within the admission and discharge sub-simulation. The simulation also incorporated different staff members, modelled as resources within

SIMUL8. The staff groups included in the simulation and the corresponding work centres

are summarised in figure 16.2. Within the simulation, each work centre requires a

minimum number of each relevant staff group to function adequately. Furthermore, there

are limited numbers of each type of clinical staff. The working practices of each member



of staff may be incorporated into the model. However, whilst the shift patterns available

in SIMUL8 are quite flexible, it would be difficult to model the effects of fatigue or varying

performance due to staff sickness, for example. Errors within the hospital process were

reflected within the simulation by the changes that are made at each work centre to the

attributes (or labels), associated with a patient (or work item).

16.3 Bed Management on the Ward

The bed management sub-simulation modelled the situation when an in-patient

temporarily leaves the ward for a diagnostic procedure. The patient’s bed is retained for

them until they return from to the ward from the diagnostic department. Figure 16.3

provides an overview of this sub-simulation.

Figure 16.3: Screenshot of the Bed Management sub-simulation.

16.4 Diagnostic Procedures

The diagnostic section of the simulation, shown in figure 16.4, consists of a closed loop

of work centres, representing several investigative procedures that are commonly

utilised for a typical congenital cardiac patient. These are echocardiography, MRI

scanning and CT scanning. The patient travels around the loop, progressing through



several different diagnostic procedures until enough clinical information has been

acquired that they may then progress to the operating theatre.

Figure 16.4: Screenshot of the diagnostic procedures sub-simulation.

16.5 Cross-Matching of Blood

Open-heart paediatric cardiac surgery procedures require the patient to undergo a blood

transfusion during the operation. Because of this, the procedure cannot commence

without the cross-matching process being completed. During informal discussions with

staff and the formal interviews carried out as part of the gathering of evidence required to

complete the CAIR (section 10), one particular latent threat was the cross-matching

process for elective surgical patients. Since changes to the attributes, or labels,

associated with a patient can be made to reflect both local processes (those that occur in

the environment where the patient physically is) and remote processes (such as a blood

test in a laboratory) it was possible to simulate the process of cross-matching of blood in

this way, occurring in parallel to other local diagnostic processes that the patient

undergoes.



Cross-matching starts with a cardiologist, usually the SHO, taking a sample of blood

specifically for this purpose. A precise set of protocols exists for this procedure, which

entail that the cross-match form must be fully completed, otherwise the sample will be

rejected by the laboratory. The blood sample is, whenever possible, acquired as part of

the pre-admission process, since the units of blood may then be prepared well in

advance of any potential operation date. Such units are valid as long as the patient has

not been transfused in the interim period before their procedure. Once the cross-

matching process has been completed by the laboratory, the appropriate units of blood

are placed in the blood fridge near the cardiac operating theatres, ready for the day of

the procedure. Although a critically important process in the patient pathway, the cross-

matching procedure has been omitted from figure 16.4. It was considered to be distinct

from the other communication flows, because there is no direct communication between

the staff members involved in the series of processes. Unless a problem is encountered

at some stage, the cardiologist responsible for acquiring the blood sample does not

communicate with the laboratory staff, and they do not communicate with the ODP who

collects the units of blood on the day of the patient’s operation. Cross-matching was

modelled for this particular reason.

Figure 16.5: Screenshot of the cross-matching of blood sub-simulation.



The sub-simulation of the processes for diagnostic testing and cross-matching of blood

is shown in figure 16.5. To provide a simple illustration of how this was achieved in

SIMUL8, the work item representing the patient is copied, and the relevant labels for one

copy are then changed in accordance with the diagnostic processes modelled. In

parallel, other labels relevant to the status of the patent with respect to the cross-

matching process are changed for the other copy. The two copies are then rejoined

prior to surgery.

16.6 Integrating the Sub-Simulations

The patient admission and discharge sub-simulation, representing the overall patient

pathway, formed the basis for the integrated simulation model. Construction of the final

model involved integrating each sub-simulation to the admission and discharge model in

turn. At each step, the simulation model was rigorously tested and, if necessary,

amended until it functioned as intended. Resources and shift patterns were then added

to the simulation, including assigning staff resources to the diagnostic loop in the pre-

surgical part of the simulation. Unfortunately, the bed management sub-simulation

proved incompatible with the other sub-simulations due to the complex structure required

to reflect this aspect of hospital operation, and was not incorporated into the final model.

Full calibration of the simulation was beyond the scope of the project. However, the

parameters that would need to be estimated as part of the calibration process are given

in Appendix 17, along with a judgement as to how the parameter estimates could be

obtained. None of the parameters were judged to be inestimable.

The code used in the final simulation can be found in Appendix 18.

16.7 Simulation Modelling for Reducing Errors in the Operating Theatre

Having illustrated how it was possible to construct a computer simulation of the surgical

care process from admission through to discharge, and the propagation of error through

the system, it is useful to examine one application of this model. Cross-matching of

blood had been identified by members of the surgical unit as being particularly

problematic in terms of the delays and other threats that it often posed to the surgical

process. As a result, the process for cross-matching was incorporated into the model,

and has been previously described. Now, the utility of the model is examined.



The situation in the trust was as follows. Two Medical Laboratory Scientific Officers

(MLSO) were required to perform the correct safety protocols for cross-matching blood

samples, which did not present a problem during normal working hours. However, at

weekends and evenings there was only one MLSO on duty at any time, despite efforts

over the course of several years to recruit further MLSO’s to cover out-of-hours shifts.

Hence, cross-matched samples received by the laboratory out-of-hours could not be

processed until the next working day, when additional MLSO’s are present. As a result,

long operative delays, when clinical staff send blood samples for cross-matching the

evening before the patient is due for surgery, were not an unusual occurrence, despite

the increasing number of patients that undergo pre-admission processes. It was possible

to reflect this within the simulation.

One solution is to acquire new cross-matching equipment, which is designed to safely

allow the cross-matching process to be carried out by a single MLSO. The cost of the

equipment is significant, however, and laboratory staff would need to be trained on the

new machines. The simulation could be used to evaluate this proposed intervention.

Data obtained from transfusion department relating the frequency with which out-of-

hours blood samples for cross-matching are received, could be used to calibrate a

simulation to estimate how effective the new laboratory equipment would be in reducing

delays and erroneous cross-matches which can have severe implications for patient

safety.

16.8 Summary

The successful construction of the admission and discharge, diagnostic procedures, and

blood cross-match sub-simulations provided a final, integrated model. Unfortunately, the

bed management sub-simulation proved incompatible with the otherwise integrated

model.



17. General Findings from the Mathematical Modelling

• A simulation approach to modelling the care processes comprising the delivery of

surgical services was feasible.

• A key part of this modelling work involved devising a framework whereby the effect of

latent safety conditions on a patient or on the information held in the system

concerning a patient can be reflected within a computer simulation. This allows the

modelling of the propagation, concatenation or resolution of errors at different stages

of the patient pathway.

• A number of sub-simulations that reflect different care processes specific to the

delivery of paediatric cardiac surgery services at a tertiary referral centre have been

developed. With one exception, these were successfully integrated into a single

model.

• The simulation models developed can be used to help evaluate alternative strategies

for reducing threats to patient safety in circumstances where, due to the complexities

and potential interactions of the care processes concerned, simpler analysis

techniques cannot be used.

• The use of computer based simulation modelling may be of considerable value as an

aid to decision making in health care organisations regarding improvements to patient

safety.



PART IV: REDUCING ERRORS IN THE OPERATING THEATRE

“It is easier to perceive error than to find truth, for the former lies on the surface and is

easily seen, while the latter lies in the depth, where few are willing to search for it.”

(Goethe, 1749-1832)

18. Summary of the Research Findings

Direct observation of errors in surgery found that commonly recurring minor failures in

operating theatres were frequent, usually tolerated, and almost never reported. They are

prospectively identifiable, more common in higher risk operations, and can be associated

with human errors predisposed by threats residing in the system. Adverse events in

surgery are likely to be associated with a co-incidental accumulation of a number of

these minor recurring failures, rather than individual incompetence or negligence.

The assessment of attitudes and safety management mechanisms in the organisations

studied showed little difference between the trusts, but identified a deficiency in the

management of both trusts appropriately to use reported incident data. This may suggest

why the observed intraoperative failures, which were generally outside the bounds of

any reporting system, were recurrent and remained unresolved.

The construction of a mathematical model of the care processes comprising the delivery

of surgical services was found to feasible. This allowed modelling of the propagation,

concatenation or resolution of errors at different stages of the patient pathway, and can

be used to help evaluate alternative strategies for reducing threats to patient safety.

Thus, the use of computer based simulation modelling may be of considerable value as

an aid to decision making in health care organisations regarding improvements to patient

safety.

We have already discussed local solutions to many of the specific problems that have

been identified. In this section, we bring these concepts together, to promote

suggestions for the global possibilities for reducing errors in the operating theatre.



19. Learning from Other Industries

The challenges of improving safety in healthcare are similar to those faced in other high-

risk industries, or the quest for improved quality in complex production processes. The

universal component is the human, and the application of human factors principles

derived from other industries, or from generic considerations of the human in the system,

reveals that humans in healthcare are faced with similar problems to those found

elsewhere. Similarly, the application to healthcare of operational research techniques

originally derived for the optimisation of manufacturing processes demonstrates that the

healthcare system usefully bears comparison with a production line. Thus, the

application of extensive knowledge of the human, the system, and their interaction in

relation to quality and safety, derived from other industries can clearly be used to

improve patient safety.

As well as combining ethnography with the experience of civil aviation professionals

throughout these studies, as an early part of this research it was possible to interview a

senior member of the Ferrari Formula 1 motor racing team regarding their approaches to

safety and reliability. A number of practices were cited as having substantially enhanced

the performance of the team over the past several years. In particular:

• A no-blame culture that was led from the highest level of management.

• A detailed database for recording errors or faulty equipment, allowing access

to information at several levels and performance tracking of every element in

the system.

• Clear procedures and extensive use of checklists.

• Well-rehearsed contingency plans for time-critical events and strategic

changes.

• Extensive auditable briefings at a wide variety of levels conducted before,

during and after race weekends to plan for future events and learn from past

experiences.

• Quiet and calm leadership, and a management team able to step back from

proceedings to assess the situation.



• The ability to learn from other teams, particularly where major accidents had

occurred.

These features are similar to those found in many other industries, and are also

remarkably similar to the Seven Steps to Patient Safety139. Since the basic components

– humans, technology, safety criticality, task complexity and system complexity – are the

same, experiences from high-risk industries must be informative. However, heavy

reliance on multiple teams of differing specialties, the lack of specific procedural

protocols or standard operating procedures, and the inability to rely on technological

solutions in the foreseeable future make many approaches inappropriate in surgery that

have been successful elsewhere. Furthermore, the volume of cases, range of goals, and

conflicting demands placed on the system make the challenge of improving patient

safety far more complex than the challenges faced in most other industries.

It is therefore important that healthcare learns the right lessons from other industries,

and understanding what is and is not applicable is key. If approaches to patient safety

derive only from healthcare practitioners, or from safety professionals without healthcare

experience, the best solutions may not be found, since neither have sufficient

experience of the complexities of the other. The ethnographic technique employed in the

present studies was a vital strength, and the ability to identify failures throughout this

complex system demonstrates that healthcare is not so complex as to be inaccessible

without healthcare training, but requires sensitivity and understanding in applying safety

principles from other industries. In the first instance, the establishment and retention of

centres of excellence for safety research and development that cross these disciplinary

boundaries would provide by far the best supported and most effective solutions for

improving patient safety.

Healthcare practitioners must recognise the value and applicability of experiences in

other industries, and safety experts must recognise the complexities of and multiple

goals in healthcare. Future work that keeps these considerations in mind may not only

improve surgical quality, but may also add new perspectives to the growing body of

research concerning safety, system design and human performance in a broad range of

complex environments.



20. Organisational Genesis of Adverse Events in Surgery.

The concept of organisational safety and the contribution that an organisation makes to

an accident is central to the systems model of safety that has been adopted here118. One

feature of the organisational accident is a chain of events that breach defences or reveal

existing, tolerated deficiencies within the system. Since responses to safety following

such events are always too late, and can be negatively influenced by hindsight, it is far

better to identify system deficiencies before they occur. As sources of failure exist within

the healthcare organisation, and not simply within different trusts, it is useful to extend

discussions beyond those directly evidenced here.

The present research has investigated how chains of events can accumulate within the

surgical process, and where the sources of deficiencies in the system may be found. By

employing the range of techniques used here to study errors, system design, and

attitudes to and management of errors, for the first time it is been possible to build a

picture of the propagation of error from the “blunt” end of management and culture to the

“sharp” end of individual performance.

From the outset, it was clear that many of the deficiencies that were associated with the

events at the Bristol Royal Infirmary16 were to be found in both of the organisations

studied here. One of the most striking features of the evidence we collected was the

similarity between the two very different surgical organisations. While there were

differences in safety attitudes and in the range, type and severity of intraoperative

failures, by and large both surgical centres were faced with similar challenges to safety.

Both exhibited problems in the dissociation between management and practitioner that

compromised effective safety management, and indirectly predisposed errors in theatre.

For example, organisational divisions between anaesthetists and surgeons, a relaxed

attitude to intraoperative safety, failures in communication and in planning for and

learning from operations allowed recurrent failures. These were not solved because of

failures in the existing reporting processes - either a reluctance to report, or a failure to

address the reported problem - and failure of the reporting mechanisms themselves to

collect sufficiently detailed information. This meant that many failures were not reported,

with a minority being addressed by the practitioners on an ad-hoc basis. Procedures in

place for managing reported incidents were poorly supported by both management and

practitioner, presumably because sufficient resources were not provided. Failure to



collect sufficient data or audit the frequent recurrence of a wide number of systems

problems allowed their perpetuation. To summarise this chain of deficiencies from the

“sharp” end of the operating theatre to the “blunt” end of the organisation and culture, we

believe we have presented evidence for:

• The predisposal of errors through:

o unusual or unexpected features of the patient;

o risk, uncertainty and variation in the tasks required of surgery;

o equipment failures, poor equipment design and training, diagnostic

deficiencies, and lack of resources to support surgery;

o interruptions, absences, and a culture tolerant of violations of safety

practice.

• The exacerbation of errors through:

o non-technical performance deficiencies;

o team hierarchies that rely on the most senior team member not to fail;

o failure to plan for contingencies or avoid error-inducing situations;

o failure to learn from previous operations or events.

• The perpetuation of errors through:

o toleration of failures;

o poorly defined or inconsistent safety management practices;

o failures in the reporting or management of incidents;

o insufficiently sensitive or detailed error reporting systems;

o ethical and cultural barriers to safety research and the understanding of

error.

It is our belief that this predisposition, exacerbation, and perpetuation of errors is a

system-wide problem, and that these difficulties afflict most – and probably all – health

service providers in the UK. The evidence presented here demonstrates that they can

directly affect events in theatre. Viewed in the growing body of knowledge regarding the

aetiology of organisational failure from seemingly innocuous safety lapses to major

catastrophes, these deficiencies in National Heath Service practice must be addressed if

errors in surgery are to be reduced.



21. Reducing Errors in the Operating Theatre

Fundamental changes will be required in healthcare if there is to be any major reduction

in the incidence of adverse events related to surgery. While the scale of the problems

means that some local improvements in safety may be possible in the near term, the

cultural changes required means that no permanent global fix will be likely in a short

timescale. To summarise the suggestions for reducing errors from the three sections:

• More evidence is needed to inform healthcare practitioners of the prevalence and

nature of error in surgery, and the benefits of adhering to safety practices.

• Understanding how errors in surgery occur can help the anticipation and

avoidance of error-inducing situations.

• The direct observation of events in theatre, utilising a systems approach to the

assessment of errors, would increase the body of safety-related knowledge in

operating theatres. It also allows the prospective identification of deficiencies in

the system that create errors, and the evaluation of future safety solutions.

• User-centred equipment design and procurement, coupled with improved

maintenance standards, could reduce equipment-related problems.

• Standard operating procedures and checklists should be used, where possible, to

reduce variability between practitioners, departments and trusts, and enhance

the predictability and safety of healthcare processes.

• Briefing and debriefing can help avoid, capture, and mitigate present and future

failures, as well as aiding better team cohesion.

• Non-technical skills training could provide the mechanism by which safety

practices can be managed and communicated effectively within varying teams.

• Safety managers should ensure healthcare practitioners are involved in the

analysis and feedback of incident reports. This could encourage more

comprehensive patient safety related reporting.

• Assessments of safety processes and attitudes may not distinguish between

healthcare trusts within the UK, and more consideration should be given to the

value of and methods for making these comparisons.



• The use of computer based simulation modelling may be of considerable value

as an aid to configuring the processes in health care organisations to create

better patient safety.

• The right lessons from other industries should be learned. Close co-operation

between healthcare practitioners and human factors or other safety specialists

should be encouraged.

It is reasonable to believe that the findings of this research are applicable to many

surgical centres in the NHS, and we recommend the deployment of these approaches to

improve surgical standards and enhance patient safety across the UK.



CONCLUSIONS AND RECOMMENDATIONS The observational studies found evidence for:

• a high number of small, seemingly innocuous intraoperative failures in both

orthopaedic and paediatric cardiac surgery.

• the accumulation of those minor failures into more major events that can be

dangerous.

• the influence of risk on the accumulation of errors.

• the predisposition of intraoperative failure by patient, task, environment (including

equipment and resources), culture and organisation.

• clear deficiencies in non-technical skills that create, and prevent the capture of,

failures.

The assessment of safety culture and institutional resilience found:

• the Checklist for Assessing Institutional Resilience (CAIR) could be usefully

deployed in the centre for paediatric cardiac surgery, but proved labour-intensive

to use, and difficult to score.

• the Incident Reporting and Attitude Survey (IRAS) and the Operating Theatre

Team Management Attitudes Questionnaire (OTTMAQ) were successfully

deployed in departments of paediatric cardiac surgery orthopaedic surgery at

different UK hospitals, though low response rates limited the conclusions that

could be drawn, and makes comparisons between trusts unlikely to be useful.

• deficiencies in safety management were largely related to an uneasy relationship

between healthcare management and healthcare practitioners. Practitioners felt

that information regarding incident reports would either be badly used or not used

at all.

The mathematical modelling of the impact of latent system conditions found that:

• a simulation approach to modelling the care processes comprising the delivery of

surgical services is feasible. This allows modelling of the propagation,

concatenation or resolution of errors at different stages of the patient pathway.



• since the simulation models developed can be used to help evaluate alternative

strategies for reducing threats to patient safety, they may be of considerable

value as an aid to decision making in health care organisations regarding

improvements to patient safety.

The recommendations for reducing errors in operating theatres are summarised thus:

• The opportunity for errors should be reduced. Checklists and standard operating

procedures could promote consistency of care. A user-centred approach to the

design, procurement and maintenance of medical equipment could reduce

equipment related problems. Interruptions, absences, and violations of safety

practice should be discouraged.

• The response to errors and the ability to avoid error-inducing situations should be

enhanced. The value of pre-operative briefings, post-operative debriefings, and

non-technical skills training should be evaluated as a method for improving the

ability of teams to anticipate and respond to safety threats. Involving healthcare

practitioners more in the management of and response to incident data may

encourage better incident reporting, but more information about unsafe situations

is also needed.

• There should be continued efforts to understand how, why and when errors occur

in operating theatres. Intraoperative error observation techniques should be used

to further enhance the body of safety-related knowledge. The ethical and cultural

barriers to effective safety research should be examined and removed where

possible. Knowledge transfer from other industries should be encouraged.

• Computer simulation methods have the potential to complement other empirical

methods in the study of systemic errors.



ACKNOWLEDGEMENTS

We would like to extend our thanks to the patients and their parents for allowing us to

involve them in our studies, and to the staff participants who allowed us to observe and

video them at work. We are also grateful to Giles Peek, of Leicester Royal Infirmary for

his help in understanding and reviewing early results from the paediatric cardiac

observational studies. Research at the Institute of Child Health and Great Ormond Street

Hospital for Children NHS Trust benefits from research and development funding

received from the NHS Executive.

Contributions of the Authors

Dr Ken Catchpole Principal research fellow. Development of human factors models, task analysis, FMEA, and observational data collection techniques. Staff, patient and parental consent. Collection, analysis and reporting of observational data. Preparation and editing of report (all sections).

Professor Marc de Leval Consultant surgeon. Project lead. Lead for cardiac surgery. Participant liaison. Review and approval of process and final report.

Mr Tony Giddings Consultant surgeon (retd.). Project lead. Lead for integration of human factors in surgery. Participant liaison. Review of process and final report.

Captain Guy Hirst Aviation human factors consultant. Specialist non-technical skills advice and practical guidance on systems safety. Development of surgical NOTECHS scale. Additional non-technical skills details to final report. Review of final report.

Captain Trevor Dale Aviation human factors consultant. Specialist non-technical skills advice and practical guidance on systems safety. Development of surgical NOTECHS scale. Additional non-technical skills details to final report. Review of final report.

Dr Paul Godden Research fellow. Development of simulation models and collection of observational data. CAIR, IRAS and OTTMAQ collection and analysis. Preparation of report parts II and III.

Dr Martin Utley Deputy Director, Clinical Operational Research Unit. Support to operational research studies. Support to analysis of observational data. Preparation of report parts II and III. Review of final report

Professor Steve Gallivan Director, Clinical Operational Research Unit. Lead for operational research studies. Preparation of report part III. Review of final report.



REFERENCES 1. Dekker SW. The Field Guide to Human Error Investigations. Aldershot: Ashgate,

2002.

2. Department of Health. An Organisation with a Memory: Report of an expert group on learning from adverse events in the NHS. 2000. London, HMSO.

3. Kohn LT, Corrigan JM, Donaldson MS. To Err is Human: Building a Safer Health System. Washington, D.C.: National Academy Press, 2000.

4. Vincent C, Neale G, Woloshynowych M. Adverse events in British hospitals: Preliminary retrospective record review. BMJ 2001;322:517-9.

5. Shahian DM, Normand SL, Torchiana DF, Lewis SM, Pastore JO, Kuntz RE et al. Cardiac surgery report cards: comprehensive review and statistical critique. Ann.Thorac.Surg. 2001;72:2155-68.

6. Cook RI, Woods D. Operating at the Sharp End: The Complexity of Human Error. In Bogner MS, ed. Human Error in Medicine, pp 255-310. Hillsdale, N.J.: Lawrence Erlbaum Associates Inc., 1994.

7. de Leval MR, Carthey J, Wright DJ, Reason JT. Human Factors and Cardiac Surgery: A Multicenter Study. Journal of Thoracic and Cardiovascular Surgery 2000;119:661-72.

8. Way LW, Stewart L, Gantert W, Liu K, Lee CM, Whang K et al. Causes and prevention of laparoscopic bile duct injuries: analysis of 252 cases from a human factors and cognitive psychology perspective. Ann.Surg. 2003;237:460-9.

9. Blackstone EH. Monitoring surgical performance. Journal of Thoracic and Cardiovascular Surgery 2004;128:807-10.

10. Gaba DM. Human error in anesthetic mishaps. Int.Anesthesiol.Clin. 1989;27:137-47.

11. Reason JT. Human Error. Cambridge: University Press, 1990.

12. Fennell, D. Investigation into the King's Cross Underground Fire. Cm 499. 1998. London, HMSO.

13. Carthey J, de Leval MR, Reason JT, Leggatt A, Wright DJ. Adverse events in cardiac surgery: The role played by human and organisational factors. In Vincent C, de Mol B, eds. Safety in Medicine, pp 117-38. Oxford: Pergamon, 2000.

14. Woolf SH, Kuzel AJ, Dovey SM, Phillips RL. A String of Mistakes: The Importance of Cascade Analysis in Describing, Counting, and Preventing Medical Errors. Annals of Family Medicine 2004;2:317-26.

15. Reason J. Human error: models and management. BMJ 2000;320:768-70.



16. Kennedy, I. Learning from Bristol: the report of the public inquiry into children's heart surgery at the Bristol Royal Infirmary 1984 -1995. Command Paper: CM 5207. 2001.

17. Gaba DM. Anaesthesiology as a model for patient safety in health care. BMJ 2000;320:785-8.

18. Helmreich RL. On error management: lessons from aviation. BMJ 2000;320:781-5.

19. Spencer FC. Human error in hospitals and industrial accidents: current concepts. J.Am.Coll.Surg. 2000;191:410-8.

20. Carthey J, de Leval MR, Reason JT, Farewell VJ, Wright DJ. Human factors research in cardiac surgery: Opportunities and pitfalls. Clinical Risk 2001;78:85-90.

21. Wieman TJ,.Wieman EA. A systems approach to error prevention in medicine. J.Surg.Oncol. 2004;88:115-21.

22. Willeumier D. Advocate health care: a systemwide approach to quality and safety. Jt.Comm J.Qual Saf 2004;30:559-66.

23. Cook RI,.Woods DD. Adapting to new technology in the operating room. Hum.Factors 1996;38:593-613.

24. Kirklin JW, Blackstone EH, Tchervenkov CI, Castaneda AR. Clinical outcomes after the arterial switch operation for transposition: Patient, support, procedural, and institutional risk factors. Circulation 1992;86:1501-15.

25. Keogh B, Spiegelhalter D, Bailey A, Roxburgh J, Magee P, Hilton C. The legacy of Bristol: public disclosure of individual surgeons' results. BMJ 2004;329:450-4.

26. Fischhoff B. Hindsight does not equal foresight: The effect of outcome knowledge on judgement under uncertainty. Journal of Experimental Psychology: Human Perception and Performance 1975;1:288-99.

27. Woods D, Cook RI. Hindsight biases and local rationality. In Durso FT, Nickerson RS, Schvaneveldt RW, Dumais ST, Lindsay DS, Chi MTH, eds. Handbook of Applied Cognition, pp 141-71. John Wiley and Sons Ltd., 1999.

28. Berlin L. Hindsight bias. American Journal of Roentgenology 2000;175:597-601.

29. Carthey J, de Leval MR, Reason JT. Institutional resilience in healthcare systems. Qual.Health Care 2001;10:29-32.

30. Vincent C. Analysis of clinical incidents: a window on the system not a search for root causes. Qual Saf Health Care 2004;13:242-3.

31. Weinger MB, Gonzales DC, Slagle J, Syeed M. Video capture of clinical care to enhance patient safety. Qual Saf Health Care 2004;13:136-44.



32. Weinger MB, Reddy SB, Slagle JM. Multiple measures of anesthesia workload during teaching and nonteaching cases. Anesth.Analg. 2004;98:1419-25, table.

33. Bann S, Khan MS, Datta VK, Darzi AW. Technical performance: relation between surgical dexterity and technical knowledge. World J.Surg. 2004;28:142-6.

34. Kaplan HS, Battles JB, van der Schaaf TW, Shea CE, Mercer SQ. Identification and classification of the causes of events in transfusion medicine. Transfusion 1998;38:1071-81.

35. Vincent C, Taylor-Adams S, Stanhope N. Framework for analysing risk and safety in clinical medicine. BMJ 1998;316:1154-7.

36. Carthey J, de Leval MR, Reason JT. The human factor in cardiac surgery: errors and near misses in a high technology medical domain. Ann.Thorac.Surg. 2001;72:300-5.

37. Xiao Y, Hunter WA, Mackenzie CF, Jefferies NJ, Horst RL. Task Complexity in Emergency Medical Care and Its Implications for Team Coordination. Hum.Factors 1996;38:636-45.

38. Proctor ML, Pastore J, Gerstle JT, Langer JC. Incidence of medical error and adverse outcomes on a pediatric general surgery service. J.Pediatr.Surg. 2003;38:1361-5.

39. Feinglass J, Amir H, Taylor P, Lurie I, Manheim LM, Chang RW. How safe is primary knee replacement surgery? Perioperative complication rates in Northern Illinois, 1993-1999. Arthritis Rheum. 2004;51:110-6.

40. Lilford RJ, Mohammed MA, Braunholtz D, Hofer TP. The measurement of active errors: methodological issues. Qual Saf Health Care 2003;12 Suppl 2:ii8-12.

41. Helmreich RL, Klinect JR, and Wilhelm JA. Models of Threat, Error, and CRM in Flight Operations. Proceedings of the Tenth International Symposium on Aviation Psychology, 667-682. 1999. Columbus, OH, The Ohio State University.

42. Klinect JR, Wilhelm JA, and Helmreich RL. Threat and error management: Data from line operations safety audits. 683-688. 1999. Proceedings of the 10th International Symposium on Aviation Psychology.

43. Vincent C, Moorthy K, Sarker SK, Chang A, Darzi AW. Systems approaches to surgical quality and safety: from concept to measurement. Ann.Surg. 2004;239:475-82.

44. Tang B, Hanna GB, Bax NM, Cuschieri A. Analysis of technical surgical errors during initial experience of laparoscopic pyloromyotomy by a group of Dutch pediatric surgeons. Surg.Endosc. 2004;18:1716-20.

45. Helmreich RL, Schaefer H-G. Team Performance in the Operating Room. In Bogner MS, ed. Human Error in Medicine, pp 225-53. Hillsdale, N.J.: Laurence Erlbaum Associates, 1994.



46. Fletcher GCL, Flin RH, Glavin RJ, Maran NJ, Patey R. Anaesthetists' Non-Technical Skills (ANTS): Evaluation of a behavioural marker system. British Journal of Anaesthesia 2003;90:580-8.

47. Nolan, T., Resar, R., Griffin, F. A., and Gordon, A. B. Improving the Reliability of Healthcare. 2004. Cambridge, MA., Institute for Healthcare Improvement.

48. Jenkins KJ, Gauvreau K, Newburger JW, Spray TL, Moller JH, Iezzoni LI. Consensus-based method for risk adjustment for surgery for congenital heart disease. Journal of Thoracic and Cardiovascular Surgery 2002;123:110-8.

49. Avermaete, J. A. G. and van Kruijsen, E. A. C. NOTECHS. The evaluation of non-technical skills of multi-pilot aircrew in relation to the JAR-FCL requirements. 1998. Amsterdam: EC NOTECHS Project final.

50. Flin RH, Fletcher GCL, McGeorge P, Sutherland A, Patey R. Anaesthetists' attitudes to teamwork and safety. Anaesthesia 2003;58:233-42.

51. Lodge M, Fletcher GCL, Russell S, Goeters K-M, Hoermann H, Nijhuis H, Flin RH, O'Connor P, Ellis A, Valot C, Polo L, and Boiardi A. Results of the Experiment. JARTELWP3 Final Report. JARTEL/BA/WP3/D5_20. 2001. The JAR TEL Consortium, for the European Commission, DG TREN.

52. Fletcher GCL, Flin RH, Glavin RJ, Maran NJ, Patey R. Rating non-technical skills: Developing a behavioural marker system for use in anaesthesia. Cognition, Technology and Work 2004;6:165-71.

53. Andlauer E and Delsart MC. Operational validation of NOTECHS. JARTEL WP4 Final Report. JARTEL/SOF/WP4/D6_22. 2001. The JAR TEL Consortium, for the European Commission, DG TREN.

54. Woods D, Patterson E. How Unexpected Events Produce An Escalation Of Cognitive And Coordinative Demands. In Hancock PA, Desmond PA, eds. Stress Workload and Fatigue., pp 290-304. Hillsdale, NJ: Lawrence Erlbaum, 2004.

55. Mrowczynski W, Wojtalik M, Zawadzka D, Sharma G, Henschke J, Bartkowski R et al. Infection risk factors in pediatric cardiac surgery. Asian Cardiovasc.Thorac.Ann. 2002;10:329-33.

56. Nateghian A, Taylor G, Robinson JL. Risk factors for surgical site infections following open-heart surgery in a Canadian pediatric population. Am.J.Infect.Control 2004;32:397-401.

57. O'Brien MM, Shroyer AL, Moritz TE, London MJ, Grunwald GK, Villanueva CB et al. Relationship between processes of care and coronary bypass operative mortality and morbidity. Med Care 2004;42:59-70.

58. Follis FM, Pett SB, Miller KB, Wong RS, Temes RT, Wernly JA. Catastrophic hemorrhage on sternal reentry: still a dreaded complication? Ann.Thorac.Surg. 1999;68:2215-9.



59. Reitz BA. What's new in cardiac surgery. J.Am.Coll.Surg. 2004;198:784-97.

60. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: a study of human factors. Anesthesiology 1978;49:399-406.

61. Dain S. Normal Accidents: Human Error and Medical Equipment Design. The Heart Surgery Forum 2002;5:254-7.

62. Ward JR,.Clarkson PJ. An analysis of medical device-related errors: prevalence and possible solutions. Journal of Medical Engineering and Technology 2004;28:2-21.

63. Morgan JA, Peacock JC, Kohmoto T, Garrido MJ, Schanzer BM, Kherani AR et al. Robotic techniques improve quality of life in patients undergoing atrial septal defect repair. Ann.Thorac.Surg. 2004;77:1328-33.

64. Suematsu Y,.del Nido PJ. Robotic pediatric cardiac surgery: present and future perspectives. Am.J.Surg. 2004;188:98S-103S.

65. Frommater D, Marshall D, Halford G, Rimmasch H, Coons MC. How a three-campus heart service line improves clinical processes and outcomes. Jt.Comm J.Qual Improv. 1995;21:263-76.

66. Null SL, Bonser K. Multiple changes, multiple opportunities in the ambulatory surgery department. J.Perianesth.Nurs. 1997;12:321-8.

67. Cannon-Bowers JA, Salas E, Blickensderfer E, Bowers CA. The impact of cross-training and work-load on team functioning: A replication and extension of initial findings. Hum.Factors 1998;40:92-101.

68. Eaves RH,.Flagg AJ. The U.S. Air Force pilot simulated medical unit: a teaching strategy with multiple applications. J.Nurs.Educ. 2001;40:110-5.

69. Veith FJ. Turf issues: how do we resolve them and optimize patient selection for intervention and ultimately patient care? J.Vasc.Surg. 1998;28:370-2.

70. Dixon T, Shaw M, Ebrahim S, Dieppe P. Trends in hip and knee joint replacement: socioeconomic inequalities and projections of need. Ann.Rheum.Dis. 2004;63:825-30.

71. Friedman RJ,.Poss R. Revision total knee arthroplasty in patients with osteoarthritis. Rheum.Dis.Clin.North Am. 1988;14:537-44.

72. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.

73. Gao T, Lu H, Zhou D, Guan Z. [Risk factors for nosocomial infections after total knee replacement]. Zhonghua Wai Ke.Za Zhi. 2000;38:256-8.

74. McLure HA, Talboys CA, Yentis SM, Azadian BS. Surgical face masks and downward dispersal of bacteria. Anaesthesia 1998;53:624-6.



75. Mitchell NJ,.Hunt S. Surgical face masks in modern operating rooms - a costly and unnecessary ritual? Journal of Hospital Infection 1991;18:242.

76. Woodhead K, Taylor EW, Bannister G, Chesworth T, Hoffman P, and Humphreys H. Behaviours and Rituals in the Operating Theatre: A report from the Hospital Infection Society Working Group on Infection Control in the Operating Theatres. 2004. The Hospital Infection Society.

77. Baird C, Nunn T, Gregori A. Inadequate standards of hygiene in an operating theatre changing room. Journal of Hospital Infection 2005;59:268-9.

78. Fletcher GCL, Flin RH, and McGeorge P. Review of Human Factors Research in Anaesthesia. SCPMDE Project: RDNES/991/C. 2000. Industrial Psychology Group, University of Aberdeen. The Identification and Measurement of Anaesthetists' Non-Technical Skills.

79. Helmreich RL. Anatomy of a system accident: The crash of Avianca Flight 052. International Journal of Aviation Psychology 1994;4:265-84.

80. Helmreich RL, Foushee HC. Why Crew Resource Management? Empirical and theoretical bases of human factors training in aviation. In Wiener E, Kanki B, Helmreich RL, eds. Cockpit Resource Management, pp 3-45. San Diego: Academic Press, 1993.

81. Helmreich RL, Butler RE, Taggart WR, and Wilhelm JA. The NASA/University of Texas/FAA Line/LOS Checklist: A behavioral marker-based checklist for CRM skills assessment. Aerospace Crew Resource Management Project Technical Report 94-04. 1995. Austin: University of Texas.

82. Gaba DM, Howard SK, Flanagan B, Smith BE, Fish KJ, Botney R. Assessment of Clinical Performance during Simulated Crises using both Technical and Behavioral Ratings. Anesthesiology 1998;89:123-47.

83. Gaba DM, Maxwell M, DeAnda A. Anesthetic mishaps: breaking the chain of accident evolution. Anesthesiology 1987;66:670-6.

84. Schaefer HG, Helmreich RL, Scheidegger D. Safety in the operating theatre--part 1: interpersonal relationships and team performance. Curr.Anaesth.Crit Care 1995;6:48-53.

85. Sexton JB, Thomas EJ, Helmreich RL. Error, Stress and Teamworking in Aviation and Medicine. BMJ 2000;320:745-9.

86. Klampfer B, Flin RH, Helmreich RL, Hausler R, Sexton JB, Fletcher GCL, Field P, Staender S, Lauche K, Dieckmann P, and Amacher A. Enhancing Performance In High Risk Evironments: Recommendations for the use of Behavioural Markers. Paper from Behavioural Markers Workshop, Swissair Training Centre, Zurich, 5-6 July 2001. 6-7-2001. Ladenburg: Daimler-Benz Shiftung.

87. Carthey J. The role of structured observational research in health care. Qual Saf Health Care 2003;12:13ii-16.



88. Helmreich RL, Merritt AC. Culture at Work in Aviation and Medicine. Aldershot: Ashgate, 1998.

89. Bann S,.Darzi AW. A protocol for the reduction of surgical errors. Qual Saf Health Care 2004;13:162-3.

90. Carthey J, de Leval MR, Wright DJ, Farewell VJ, Reason JT. Behavioural markers of surgical excellence. Saftey Science 2003;41:409-25.

91. Musson DM, Helmreich RL. Team training and resource management in healthcare: Current issues and future directions. Harvard Health Policy Review 2004;5:25-35.

92. Leape LL, Berwick DM, Bates DW. What practices will most improve safety? Evidence-based medicine meets patient safety. JAMA 2002;288:501-7.

93. Meding JB, Keating EM. Overview: long-term followup: predicting the success of a total knee arthroplasty. Clin.Orthop.Relat Res. 2001;7-9.

94. Nizard R. Computer assisted surgery for total knee arthroplasty. Acta Orthop.Belg. 2002;68:215-30.

95. Nizard RS, Porcher R, Ravaud P, Vangaver E, Hannouche D, Bizot P et al. Use of the Cusum technique for evaluation of a CT-based navigation system for total knee replacement. Clin.Orthop.Relat Res. 2004;180-8.

96. Weinger MB. Anesthesia equipment and human error. J.Clin.Monit.Comput. 1999;15:319-23.

97. Weinger MB, Pantiskas C, Wiklund ME, Carstensen P. Incorporating human factors into the design of medical devices. JAMA 1998;280:1484.

98. Weinger MB. Proposed new alarm standards may make a bad situation worse. Anesthesiology 1991;74:791-2.

99. Lieberman JR, Wenger N. New technology and the orthopaedic surgeon: are you protecting your patients? Clin.Orthop.Relat Res. 2004;338-41.

100. Bainbridge L. Ironies of Automation. Automatica 1983;19:775-9.

101. Parasuraman R, Riley V. Humans and Automation: Use, Misuse, Disuse, Abuse. Hum.Factors 1997;39:230-53.

102. Woods D, Sarter N, Billings C. Automation Surprises. In Salvendy G, ed. The Handbook of Human Factors, 2nd Edition, 1997.

103. Miller DP, Swain AD. Human Error and Human Reliability. In Salvendy G, ed. Handbook of Human Factors, pp 219-50. Chichester: John Wiley & Sons, 1987.

104. Department of Health. Design for Patient Safety. Reference number 32768. 2004. Departement of Health & The Design Council.



105. Tucker AL. The impact of operational failures on hospital nurses and their patients. Journal of Operations Management 2004;22:151-69.

106. Tucker AL, Edmondson AC, Spear S. When problem solving prevents organizational learning. Journal of Organizational Change Management 2002;15:122-37.

107. Hofer TP,.Hayward RA. Are Bad Outcomes from Questionable Clinical Decisions Preventable Medical Errors? A Case of Cascade Iatrogenesis. Annals of Internal Medicine 2002;137:327-33.

108. Healy AN, Undre S, Vincent C. Developing observational measures of performance in surgical teams. Qual Saf Health Care 2004;13:i33-i40.

109. Battles JB, Lilford RJ. Organizing patient safety research to identify risks and hazards. Qual Saf Health Care 2003;12 Suppl 2:ii2-ii7.

110. Hormann HJ. Cultural variation of perceptions of crew behaviour in multi-pilot aircraft. Trav.Hum. 2001;64:247-68.

111. Mackenzie CF, Xiao Y. Video techniques and data compared with observation in emergency trauma care. Qual Saf Health Care 2003;12 Suppl 2:ii51-ii57.

112. Mackenzie CF, Hu FM, Xiao Y, Seagull FJ. Video acquisition and audio system network (VAASNET) for analysis of workplace safety performance. Biomed.Instrum.Technol. 2003;37:285-91.

113. Flanagan B, Nestel D, Joseph M. Making patient safety the focus: crisis resource management in the undergraduate curriculum. Med Educ. 2004;38:56-66.

114. Mackenzie CF, Jefferies NJ, Hunter WA, Bernhard WN, Xiao Y. Comparison of self-reporting of deficiencies in airway management with video analyses of actual performance. LOTAS Group. Level One Trauma Anesthesia Simulation. Hum.Factors 1996;38:623-35.

115. Mackenzie CF, Martin P, Xiao Y. Video analysis of prolonged uncorrected esophageal intubation. Level One Trauma Anesthesia Simulation Group. Anesthesiology 1996;84:1494-503.

116. Weinger MB, Slagle J. Human factors research in anesthesia patient safety. Proc.AMIA.Symp. 2001;756-60.

117. Weinger MB, Slagle J, Jain S, Ordonez N. Retrospective data collection and analytical techniques for patient safety studies. J.Biomed.Inform. 2003;36:106-19.

118. Reason J. Managing the Risks of Organisational Accidents. Aldershot: Ashgate, 1997.

119. Sexton JB, Thomas EJ, Helmreich RL. Error, stress and teamwork in medicine and aviation: cross sectional surveys. BMJ 2000;320:745-9.



120. Armitage P, Berry G. Statistical methods in medical research. Oxford: Blackwell Scientific Publications, 1985.

121. Bagust A, Place M, Posnett JW. Dynamics of bed use in accommodating emergency admissions: stochastic simulation model. BMJ 1999;319:155-8.

122. Bensley DC, Watson PS, Morrison GW. Pathways of coronary care-a computer simulation model of the potential health gain. IMA J Math Appl Med Biol 1995;12:315-28.

123. Brailsford SC, Schmidt B. Towards incorporating human behaviour in models of health care systems: an approach using discrete event simulation. European Journal of Operational Research 2003;150:19-31.

124. Brailsford SC, Shahani AK, Basu Roy R, Sivapalan S. Simulation Modelling for HIV infection and AIDS. International Journal of Biomedical Computing 1992;31:83-8.

125. Costa AX, Ridely SA, Shahani AK, Harper PR, de Senna V, Nielsen MS. Mathematical Modelling and Simulation for Planning Critical Care Capacities. Anaesthesia 2003;58:320-7.

126. Davies R, Roderick P, Raftery J. The evaluation of disease prevention and treatment using simulation models. European Journal of Operational Research 2003;150:53-66.

127. Davies R,.Davies HTO. Modelling patients flows and resource provision in health systems. Omega 1994;22:123-31.

128. Davies R, Roderick P, Canning C, Brailsford SC. The evaluation of screening policies for diabetic retinopathy using simulation. Diabetic Medicine 2002;19:762-70.

129. El-Darzi E, Vasilakis C, Chaussalet T, Millard PH. A simulation modelling approach to evaluating length of stay, occupancy, emptiness and bed blocking in a hospital geriatric department. Health Care Mangement Science 1998;1:143-9.

130. Fetter RB, Thompson JD. The simulation of hospital systems. Operations Research 1965;13:689-711.

131. Harper PR, Shahani AK. Modelling for the planning and management of bed capacities in hospitals. Journal of the Operational Research Society 2002;2:11-8.

132. Harper PR, Phillips S, Gallaguer JE. Geographical simulation modelling for the regional planning of oral and maxillofacial surgery across London. Journal of the Operational Research Society 2005;56:134-43.

133. Harper PR, Gamlin H. Reduced Outpatient Waiting Times with Improved Appointment Scheduling: a Simulation Modelling Approach. Operational Research Spectrum 2003;25:207-22.



134. Dean, B. Hospital medical administration errors: their simulation, observation and severity assessment. 1999. The School of Pharmacy, University of London.

135. Dean B, Gallivan S, Barber ND, van Ackere A. Mathematical modelling of pharmacy systems. American Journal of Health System Pharmacy 1997;54:2491-9.

136. Dean B, van Ackere A, Gallivan S, Barber ND. When should pharmacist visit their wards?. An application of simulation to planning hospital pharmacy services. Health Care Mangement Science 1999;2:35-42.

137. Dean B, Barber ND, van Ackere A, Gallivan S. Can simulation be used to reduce errors in health care delivery? The hospital drug distribution system. Journal of Health Service Research Policy 2001;6:32-7.

138. SIMUL8 Professional Edition (educational site license version), V11.0, build 11.0.0.774. 2004. The SIMUL8 corporation.

139. National Patient Safety Agency. Seven Steps to Patient Safety. 2004. London, National Patient Safety Agency.

140. Iyer RS, Jacobs JP, de Leval MR, Stark J, Elliott MJ. Outcomes after delayed sternal closure in pediatric heart operations: a 10-year experience. Ann.Thorac.Surg. 1997;63:489-91.



GLOSSARY OF TERMS

General Terms

Adverse event Occurrences during clinical care that result in physical or psychological injury or harm to a patient or harm to the mission of the organization.

Authority Gradient The power-distance relationship between two individuals, which influences how they interact. For example, a junior nurse will find it more difficult to suggest that a senior consultant has made an error than vice-versa.

Error An undesirable deviation from intent, or a human action or set of actions that exceed some limit of acceptability.

Failure Modes and Effects Analysis (FMEA)

An analytical process to identify possible failures of a product or service and then determine the frequency and impact of the failure.

Human Factors The scientific discipline concerned with the understanding of interactions among humans and other elements of a system, and the profession that applies theory, principles, data, and other methods to design in order to optimize human well-being and overall system performance.

Major Failures Events that come close to an incident or accident.

Minor Failures Small undesirable process events.

Operational Research The use of mathematical or computer models, and other analytical approaches, to understand and assess the processes in an organisation.

Task Analysis The study of actions and/or cognitive processes that are required in order to achieve a task.

Threats An element of the system that predisposes error.

Work centre The unit that represents an individual component of the healthcare process in a computer simulation.

Work item The unit that represents an item that passes through work centres in a computer simulation – in this case, the patient and patient related information.

Paediatric Cardiac Surgery Terms

Activated Clotting Time (ACT)

A commonly used measure of blood coagulation time in cardiovascular surgery.

Anastomosis The surgical opening or joining of two normally separate spaces, organs, or vessels.

Arterial Switch Operation (ASO)

An operation to correct transposition of the great arteries.



Atrial appendage A small projection from the atrium with a cavity of its own, often used as a cannulation site for cardio-pulmonary bypass.

Atrial septostomy A communication between atria made to allowing the mixing of oxygenated and de-oxygenated blood. It provides support for the first few days of life in patients with transposition of the great arteries.

Cannulation Insertion of a tube (known as a cannula) into a duct or cavity. In the present studies it exclusively refers to the insertion of aortic and venous perfusion cannulae.

Cardiopulmonary Bypass (CPB)

The extracorporeal membrane oxygenation support of a patient in order to bypass the heart and lungs, and allow open-heart surgery.

Cross Clamp The occlusion of the aorta during CPB using a pinching instrument.

Deep Hypothermic Circulatory Arrest (DHCA)

The complete cessation of blood circulation and membrane oxygenation. The patient is protected by lowering their core temperature to approximately 18ºC.

Delayed Sternal Closure (DSC)

A technique to allow the closure of a midline sternotomy to be conducted several days after the operation, and avoid compression of the intathoracic space. The sternum is held open with stents, and a silicone membrane is sutured to the wound140

Echocardiography Non-invasive imaging of heart using ultrasound to determine anatomy and delineate the defects.

Epicardial [….] Echocardiography conducted by placing sensor on the surface of the heart.

Transesophageal [..] Echocardiography conducted via a senor on an endoscope inserted into the oesophagus.

End Tidal Carbon Dioxide (ETCO2)

Carbon dioxide exhaled at the end of a normal expiration. A measure of oxygen exchange.

Exsanguination Draining of blood from the patient.

Haematocrit (HCT) The relative volume of blood occupied by red blood cells.

Heparin Drug that increases blood clotting time to prevent embolism and allow support by CPB.

Homograft Section of tissue obtained from one human and implanted in another.

Hypoplastic Left Heart Syndrome (HLHS)

A congenital cardiac condition, with 100% mortality if left untreated, where the left side of the heart is underdeveloped.

Intermural coronary artery pattern

An unusual configuration of coronary arteries which has considerable implications for the success of the arterial switch operation due to the increased surgical difficulty in managing dissection and reimplantation.

Ischeamia Low oxygen state in tissues indicating poor oxygen supply.



Used in these studies to refer to a hypoxic condition of the heart tissue.

LeCompte manoeuvre A procedure toward the end of an arterial switch operation where the pulmonary artery is passed behind the aorta.

Midline sternotomy Incision through the sternum along the midline.

Modified Ultrafiltration (MUF) or Haemofiltration

Processes by which red blood cells are concentrated, using the perfusion system, following cardio-pulmonary bypass.

Norwood Procedure A two- or three-stage palliative surgical procedure for children with hypoplastic left heart syndrome.

Oxygen Saturation A measure of the current level of oxygen carried by the blood.

Perfusion The act of passing liquid through an organ or tissue. In the present study it exclusively refers to the support of the patient by cardio-pulmonary bypass.

Protamine Drug that reverses the effects of heparin, allowing the formation of clots and cessation of bleeding around the treatment areas.

Transposition of the Great Arteries (TGA)

A congenital cardiac condition with 100% mortality if left untreated, where the aortic and pulmonary arteries are connected to the heart in the opposite way to normal.

Orthopaedic Surgery Terms

Intramedullary rod Rod inserted into the femoral bone marrow.

Femoral locating device Template fitted to the femur utilising the intramedullary rod. Allows subsequent removal of the rod and placement of femoral cutting blocks.

Femoral sizing guide Device used to measure the size of the femur for femoral prosthesis.

Distal femoral cutting block

Support for oscillating bone saw in order to make distal femoral cut.

A/P chamfer cutting block

Support for oscillating bone saw in order to make anterior, posterior and chamfer femoral cuts.



APPENDIX 1: Tasks completed and variations from proposed plan of work Planned Tasks Status Notes IRAS - Cardiac Completed Low response rate IRAS - Orthopaedic Completed Low response rate IRAS - Vascular Not

completed Outside timescale

OTTMAQ - Cardiac Completed Low response rate OTTMAQ - KCH Completed Low response rate OTTMAQ - Vascular Not


CAIR - Cardiac Completed Poor response CAIR - Orthopaedic Not


CAIR - Vascular Not completed

Outside timescale

Complete theoretical training Completed [See NOTE below] Task analysis: neonatal cardiac surgery Completed Key surgical

processes Task analysis: orthopaedic surgery Completed Key surgical

processes Task analysis: vascular surgery Not


FMEA: neonatal cardiac surgery Completed FMEA: orthopaedic surgery Completed FMEA: vascular surgery Not


Event classification from the three FMEAs. Completed Create error framework via FMEA data Completed Develop theatre observation data collection form

Completed

Complete instruction for both observers 10 pilot cases: neonatal cardiac surgery Completed 38 required 10 pilot cases: orthopaedic surgery Completed 10 pilot cases: vascular surgery 3 completed Outside timescale Adjust error framework via results obtained in the pilot cases

Completed Implicit observational techniques required

MacSHAPA: familiarisation Not completed

Outside timescale

MacSHAPA: programming Not completed

Outside timescale

Finalise Cardiac post-op questionnaire Completed Finalise Orthopaedic post-op questionnaire Completed Finalise Vascular post-op questionnaire Not


Acquire Cardiac theatre event data (50 cases) 25 completed



Acquire Orthopaedic theatre event data (50 cases)

20 completed Difficulties in recruitment, ethics and patient attrition

Acquire Vascular theatre event data (50 cases)

Not completed

Outside timescale

Acquire Cardiac theatre behavioural marker data (50 cases)

25 completed

Acquire Orthopaedic theatre behavioural marker data (50 cases)

20 completed

Difficulties in recruitment, ethics and patient attrition

Acquire Vascular theatre behavioural marker data

Not completed

Outside timescale

Draw conclusions between observer-identified errors and video-recorded errors

Complete

Formalise structure of post-op debrief Not completed

Perform debrief at Cardiac (50 cases?) Not completed

Perform debrief at Orthopaedic (50 cases?) Not completed

Perform debrief at Vascular (50 cases?) Not completed

Task prevented by ethical and organisational issues.

Note on Theoretical Training The original proposal envisaged a period of theoretical training in the features of surgical practice relevant to making informed observations. Unfortunately, the member of the original project team that was to deliver this training left the Institute of Child Health for a position at the National Patient Safety Agency. Less formal training based on reading surgical texts and talking to surgeons took place. Fortunately the Human Factors researcher employed on the project was more experienced than originally envisaged. Also, the Operational Researcher employed on the project had considerable experience of working in clinical environments, including assisting with minor surgical procedures and observing major surgery.



APPENDIX 2: Detail of task analyses

Generic Task Analysis

1.0 Transfer to Theatre This is the task that defines the entry of the patient to theatre from the induction room, and the installation of the patient on the operating table.

1.1 Move patient in The patient is moved on a wheeled bed from induction room into a position where they can be lifted onto the operating table. Difficulties encountered mostly relate to physical obstructions that prevent the movement of the bed, or undesirable interaction between the bed and the theatre equipment.

1.2 Transfer patient to table The patient is lifted across from the bed to the operating table. This task requires a co-ordinated team effort to ensure that the patient is moved appropriately and efficiently. Attention must also be paid to the critical pump and monitoring lines.

1.3 Transfer monitoring lines Lines from the patient attached to the anaesthetist’s monitoring equipment must be transferred from portable or induction room equipment to the surgical anaesthetic workstation. This may also require the equipment to be zeroed or configured appropriately. Again, it is important to ensure that tension is minimised, and can be complicated by the theatre layout. 1.4 Transfer pump lines Lines for the administration of drugs are attached to the patient and to portable pumps. A number of these pumps may be installed on a stand. Sometimes the stand is portable, and the pumps can be wheeled in with the patient as a set. This requires the dedication of one individual to the task of wheeling the stand in, alongside the patient. Sometimes these pumps must be individually transferred from one stand in the induction room, brought in with the patient, and installed individually on a second stand in theatre. Again, a common problem is tension on the lines, though it is also possible that the pumps may be dropped.

2.0 Access This task defines the process from the moment the transfer of the patient is complete, to the moment at which the treatment area is fully exposed.

2.1 Preparation Preparation describes all the stages required before the first incision when the patient is on the table. This includes generic tasks such as; cleaning and applying iodine to the skin; positioning the patient in the best way; ensuring the patient will maintain that position; and covering the patient with sterile green sheets. However, there will also be small procedure-specific variations, such as installing warming blankets around the patient. Major risks are task omission, or a sterile violation.



2.2 Incision This is the moment when the first incision is made. It is usually required that the senior anaesthetist is present, or at least nearby. Usually, but not always, the surgeon will indicate his readiness to start before making the cut. Sometimes the consultant surgeon is not present during this time, and the task is conducted by the assistant surgeon. 2.3 Exposure of treatment area Highly variable according to the procedure, this sub-task describes the progress from first incision to the treatment area. It requires particular attention to the control of bleeding, but is usually routine and relatively invariant for a given procedure. It may be more difficult if the patient has had similar previous treatments. For example, a redo sternotomy can be particularly prone to bleeding, and adherence to the back of the sternum requires greater surgical care than a first sternotomy.

3.0 Preparation This task defines all the procedures that are performed between the exposure of the treatment area, and the commencement of the high-risk treatment part of the procedure.

3.1 Anatomical Inspection During this phase, the surgical team will examine the treatment area to ensure that it matches their pre-procedural expectations. They may also indicate specific anatomical features to other members of the team or observers.

3.2 Finalise the treatment plan. Many consultant surgeons will discuss the final plans for the treatment with the assistant surgeons, and other members of the team. This provides the opportunity to address any patient-specific issues on top of the more general aspects of the procedure, as well as ensuring a shared understanding of the process, and a capture of any diagnostic inaccuracies, prior to entering the most difficult stage of the procedure.

3.3 Stabilise patient for treatment. This sub-task will vary according to the physiological strain that the patient will be put under. For example, a hip replacement should not provide a significant risk to the cardio-vascular condition of the patient beyond the loss of blood that the treatment entails. Thus, the patient should already be in a suitably stable condition for treatment. However, an arterial switch requires that the patient be supported extra-corporeally for minutes or hours. Thus, to achieve stability, the patient must be cannulated, perfused and cooled on Cardio-Pulmonary Bypass (CPB) before the treatment commences.

4.0 Treatment This is the beginning of the high-risk phase of the procedure, and will be entirely procedure-specific.

4.1 Execution of treatment. This task analysis can be adapted for particular procedures. Figure A1.1 shows a sub-task analysis for the Arterial Switch Operation. There will also be patient-



specific variations, both planned and unplanned. Variations from standard procedures will add risk, and unplanned variations will add the most risk.

Figure A1: Treatment sub-task for the arterial switch operation 5.0 Assessment Once the treatment has been completed, before the procedure can be completed it is important to assess the success of the operation. The extent and range of assessments are both procedural and patient specific, and often the sub-tasks are conducted simultaneously. Should any of the assessments not show the treatment to be suitable, further surgical intervention may be necessary.

5.1 Stabilise patient Once the treatment is complete, it is important to ensure that the condition of patient is stable. Again, this sub-task varies according to the physiological strain that the patient has been subject to, and is of greatest importance during CPB procedures, where the surgeon and anaesthetist will agree when it is safe to take the patient off bypass. Sometimes pacing wires or other equipment are used to ensure that the patient can be successfully stabilised. In total hip or knee replacement operations, this period is where the prosthetic is cleaned and the team wait for the cement to cure.

5.2 Assess treatment area This sub-task describes the visual inspection of the treatment site, for example, checking for bleeding at suture lines or the fit of the prosthetic.

5.3 Assess treatment success This is the process that surgeons go through to check their own work, and uses procedure specific protocols or tools. In hip or knee replacement, the leg is extended and flexed to assess joint mobility, and leg length is checked. For cardio-vascular operations, echocardiographic or blood doppler measuring equipment can be used to provide post-procedural haemodynamic assessments. If such specialist assessments are required, they will often need to be anticipated prior to the operation, and in cardiac surgery usually benefit from the assistance of a cardiological specialist.

6.0 Closure Once the patient is stable, and the treatment has been assessed as successful, closure can begin. The surgeon may not be present for this part of the procedure.

6.1 Count equipment

Treatment

Aortic Cross Clamping

Transection of ascending aorta & main PA

Coronary artery buttons excised

Buttons re-implanted

LeComte Manoeuvre

Aorta anastomosed to new aortic root

Reconstruction of main PA

Closure of VSD

Cross-clamp removed



During the procedure, the runner keeps a tally of the number of needles and swabs used. Prior to and during the closure process, the scrub nurse and runner perform an audit to ensure that there is no chance of leaving anything behind. 6.2 Close treatment site Though this is a generic task, in that all operations will lead to closing of the incisions made, components of this task are highly specific depending upon the location of the treatment. This task may also require the removal of CBP cannulae, and the insertion of monitoring lines or pacing wires. 6.3 Close skin The completion of this task signifies the completion of the surgical aspects of the procedure.

7.0 Transfer from Theatre The final part of the procedure describes the move of the patient from the operating table to the ICU.

7.1 Prepare for transfer. Once skin closure is complete, the patient must be readied for transfer to a bed. This involves removing of the green sterile sheets, preparing drug and monitoring lines for the transfer, placement of a bed beside the operating table, and the assembly and direction of a team to conduct the transfer. It also requires that on the bed is appropriate portable equipment to allow monitoring of the patient during the transfer to ICU, and in some cases a portable de-fib unit is required.

7.2 Transfer patient to bed This is the task involved in moving the patient from operating table to bed. It is a team task requiring co-ordination, and care with the patient, and the monitoring and drug lines.

7.3 Transfer monitoring lines This is almost exactly the same as task 1.3 in reverse, with the exception that portable monitoring equipment is always used during the transfer to ICU.

7.4 Transfer pump lines This is almost an exact repeat of task 1.4 in reverse. Sometimes pumps are simply placed on the bed. Where a separate portable pump stand is used, one individual will be required to push this alongside the bed to ICU. 7.5 Move patient out Once on the bed, and all preparations for the move have been made, the bed can be wheeled out of surgery to the ICU.




1.0 Access This task defines the process from the moment just prior to the first incision to the moment at which the patient is placed on full flow bypass.

1.1 First Incision 1.1.1 Surgeon seeks visual or verbal confirmation of readiness to start 1.1.2 Incision with scalpel or diathermy 1.1.3 Bleeding is controlled 1.2 Exposure of treatment area 1.2.1 Incision is carried to periosternum 1.2.2 Sternal saw is prepared 1.2.3 Anaesthetist disconnects ventilation and sternal saw is used. 1.2.3a (Redo sternotomy). Sternum is split with scissors 1.2.4 Bone wax is used to control sternal bleeding. 1.2.5 Retractor is used to slowly spread chest. 1.2.6 Further control of bleeding. 1.2.7 Thymus is removed. 1.2.8 Pericardium is opened. 1.3 Anatomical inspection 1.3.1 Arteries are exposed and dissected according to requirements 1.3.2 Treatment area is examined for consistency with expectation 1.4 Prepare for Treatment 1.3.1 Pericardium is harvested if needed. 1.3.2 Heparin is administered 1.3.2.1 Surgeon asks for heparin 1.3.2.2 Anaesthetist confirms & ensures perfusionist is present 1.3.2.3 Anaesthetist prepares and administers heparin 1.3.2.4 Anaesthetist announces heparin administration 1.3.2.5 Blood is taken for ACT after 3 minutes. 1.3.2.6 ACT is confirmed to be >200. 1.3.3 Cannulation 1.3.3.1 Surgeon confirms readiness to cannulate 1.3.3.2 Arterial cannulation 1.3.3.3 Venous cannulation 1.3.4 Bypass is initiated. 1.3.4.1 Surgeon confirms readiness for bypass. 1.3.4.2 Perfusionist & surgeon co-ordinate to initiate bypass 1.3.4.3 Perfusionist confirms full flow bypass 1.3.4.4 Anaesthetist disconnects ventilation. 1.3.4.5 Temperature plan is made. 1.3.4.6 Variations to venous cannulae according to treatment.

1.3.4.7 Vents and suction lines are placed.



2.0 Treatment This task describes the period the patient is on bypass. Universal procedures are described in sections 2.1-2.5. Common, but not universal procedures are described in sections 2.6-2.8. 2.1 Surgical Tasks 2.1.1 Utilise psychomotor and cognitive skills to complete surgical tasks. 2.1.2 As far as is safe, treatment should be as fast as possible.

2.2 Maintenance of effective perfusion 2.2.1 Full flow is maintained where required 2.2.2 Surgical field is kept free from blood 2.2.3 Reservoir is kept at a sufficient level 2.2.4 Problems resolved quickly 2.2.5 ACT checked regularly

2.3 Regular bloodgas check 2.3.1 Perfusionist takes bloodgas sample from perfusion system 2.3.2 ODA takes blood for analysis and brings back results 2.3.3 Perfusionist and anaesthetist examine results 2.3.4 Bloodgas results are affixed to patient notes. 2.4 Temperature control 2.4.1 Temperature is maintained as planned. 2.4.2 Rewarming begins once requested by surgeon. 2.4.3 Temperature is 36°C prior to weaning from bypass.

2.5 Preparation for weaning from bypass 2.5.1 Cross-clamp is removed. 2.5.2 Heart is de-aired. 2.5.3 Heart is observed to be beating.

2.5.4 Pacing wires attached. 2.5.4a OR heart is in sinus rhythm. 2.5.5 Filtration (where required) has been completed. 2.5.6 Ventilation is on. 2.5.7 Sufficient ionotrope support is available 2.5.8 Confirmation that team is ready to wean patient from bypass.

2.6 Cross-clamp1

2.6.1 Cross clamp is applied. 2.6.2 Cross clamp is removed.

2.7 Cardioplegia administration2

2.7.1 Cardioplegia is primed 2.7.2 Cardioplegic needle is inserted 2.7.3 Cardioplegia is run according to dosage required by surgeon 2.7.4 Plan is made for next cardioplegic cycle 1 where cross-clamp is required 2 where cardioplegia is required



2.8 DHCA Requirements3 2.8.1 Target temperature (<25°C) is reached. 2.5.2 Patient has been on bypass for 20 mins. 2.8.3 Ice packs used at head to minimise neurological damage. 2.8.4 Perfusionist periodically announces time on DHCA. 3.0 Closure

3.1 Stabilise patient 3.1.1 Weaning from bypass

3.1.1.1 Perfusionist and surgeon co-ordinate reduction in flow 3.1.1.2 Anaesthetic alarms switched on. 3.1.1.3 Venous perfusion line is clamped 3.1.1.4 Blood is transfused. 3.1.1.5 Venous cannulae are removed

3.1.2 Modified Ultrafiltration 3.1.2.1 MUF plan is agreed 3.1.2.2 Hotline is placed in atrial appendage 3.1.2.3 Filtration begins 3.1.2.4 Bloodgas is used to check HCT 3.1.2.5 Filtration finishes

3.1.3 General Stability 3.1.3.1 Sinus rhythm is achieved 3.1.3.2 Suitable hemodynamic stability is achieved 3.1.3.3 Oxygen saturations acceptable 3.1.3.4 Bleeding is controlled 3.1.3.5 Acidosis is controlled 3.1.3.6 Chest drains inserted 3.1.3.7 Peritoneal dialysis catheter inserted4

3.2 Assess treatment

3.2.1 Echocardiographic investigation 3.2.2 Bleeding sites examined 3.3 Count Equipment 3.3.1 Count before closure of chest cavity

3.3.1.1 2 members conduct the counts aloud, and in unison 3.3.1.2 Recommence if there is an interruption 3.3.1.3 Verbal confirmation of the success of the count.

3.3.2 Count before skin closure 3.3.2.1 2 members conduct the counts aloud, and in unison 3.3.2.2 Recommence if there is an interruption 3.3.2.3 Verbal confirmation of the success of the count.

3.4 Skin Closure 3.4.1 Arterial cannula is removed.

3.4.2 Heparin reversal 3.4.2.1 Surgeon asks for protamine 3.4.2.2 Protamine is given by anaesthetist 3 where DHCA is required 4 where peritoneal dialysis is required



3.4.2.3 Confirmation of protamine administration by anaesthetist 3.4.2.4 ACT returns to pre-operative level 3.4.3 Clots observed to be forming

3.4.4 Retractor removed 3.4.5 Sternal Closure

3.4.5.1 Sternum closed with heavy sutures 3.4.5.2 Deep fascia closed 3.4.5.3 Subcutaneous tissue closed 3.4.5.4 Skin closed

3.4.5a Delayed Sternal Closure 3.4.5a.1 Two stents applied crossways 3.4.5a.2 Stents sutured to sternum

3.4.5a.3 Transparent window stitched to wound

Total Hip Replacement Surgery

1.0 Access 1.1 First Incision

1.1.1 Surgeon seeks visual or verbal confirmation of readiness to start 1.1.2 Incision with scalpel or diathermy over hip joint. 1.2 Exposure of treatment area

1.2.1 Gluteus maximus divided 1.2.2 Release of psoas & capsulotomy 1.2.3 Trochanter with its muscle attachments removed5. 1.2.4 Full exposure of acetabulum and proximal femur 1.2.5 Bleeding is controlled

2.0 Treatment

2.1 Femoral resection 2.1.1 Hip capsule opened. 2.1.2 The femoral head is dislocated. 2.1.3 Power saw is prepared 2.1.4 Femoral neck removed.

2.2 Femoral Preparation 1 2.2.1 Leg held at 45° by assistant 2.2.2 Femoral elevator and retractor used to expose femur 2.2.3 Opening in the femoral neck is developed 2.2.4 Cavity is opened using a gouge or chisel 2.2.5 Taper pin reamer used to develop femoral slot 2.2.6 Traebecular bone removed 2.2.7 Rasp/broach of correct size is used6

2.3 Acetabular preparation 2.3.1 Lip of acetabulum is identified 2.3.2 Osteophytes are removed 2.3.3 Grater reamer is used to remove cartilage and shave cavity 2.3.4 Gouge and curette used to create ischial and pubic cement pits 2.3.5 Multiple 10mm holes are drilled in illium

5 Executed only on some patients, to allow better access to treatment sites. 6 Very important not to remove too much. Where excess force required a lower size should be used.



2.3.6 Supplementary 6mm holes are also drilled 2.3.7 Trial cup is inserted to check size and fit 2.3.8 Rim seal is checked and packed with autograft chips if necessary 2.3.9 Interface is cleaned

2.4 Acetabular cementing 2.4.1 Acetabulum packed with swabs while the cement is prepared

2.4.2 Scrub nurse mixes cement for 1 minute 2.4.3 Cement is left to stand for 1 minute 2.4.4 Cement is thumbed into pits 3-4 minutes after mixing starts 2.4.5 Sample of cement is kept and checked by surgeon periodically 2.4.6 Pressure is applied using pressuriser, gauze, and force 2.4.7 Pressure continues until bleeding pressure is reduced

2.5 Acetabualar implantation 2.5.1 Prosthetic is obtained by runner

2.5.2 Surgeon confirms correct size. 2.5.3 Prosthetic is unpacked and given to scrub nurse. 2.5.4 Implantation begins when cement is at right consistency 2.5.5 Cup implanted with introducer

2.5.6 Considerable force is applied 2.5.7 Rods on introducer used to set correct angle 2.5.8 Cup is fixed and checked

2.6 Trial reduction 2.6.1 Trial head/neck placed onto rasp 2.6.2 Length of leg checked 2.6.3 Trial femoral head removed 2.6.4 Leg length set using spigot

2.7 Femoral prepration 2 2.7.1 Intramedullary plug size is selected7 2.7.2 Femoral seal & backing plate positioned on cement gun nozzle 2.7.3 Assembly is tested over the end of the femur 2.7.4 Femoral canal is cleaned 2.7.5 Suction catheter is placed in canal 2.7.6 Gauze is used to pack wound & removed just before cementing

2.8 Femoral cementing 2.8.1 Cement is mixed for 1 minute 2.8.2 Cement gun is filled 2.8.3 Assembly left to stand for 30 seconds 2.8.4 Nozzle is fixed and gun is primed 2.8.5 Cement introduced 2½ - 3 minutes after mixing begins 2.8.6 Femoral seal is impacted 2.8.7 Impacting continues for approximately 6 minutes 2.8.8 Femoral stem is inserted as late as possible

2.9 Femoral stem implantation 2.9.1 Femoral stem prosthetic is obtained by runner 2.9.2 Surgeon confirms correct size. 2.9.3 Prosthetic is unpacked and given to scrub nurse. 2.9.4 Selected stem mounted to introducer 2.9.5 Stem is introduced as late as possible in cement curing process 2.9.6 Stem is driven into canal

7 Using either plug trials, or flexible reamers



2.9.7 Introducer is removed 2.9.8 Firm pressure is maintained until the cement fixes.

2.10 Reduction 2.10.1 Spigot protector removed and cleaned 2.10.2 Femoral head prosthetic is obtained by runner 2.10.3 Surgeon confirms correct size 2.10.4 Prosthetic is unpacked and given to scrub nurse. 2.10.5 Femoral head is fitted & impacted with hand

3.0 Closure

3.1 Stabilise patient 3.1.1 Tissues are cleaned with sterile saline solution

3.2 Assess treatment 3.1.1 Surgeon ensures the integrity of the hip abductor mechanism 3.1.2 Iliotibial band is checked

3.3 Count Equipment 3.3.1 2 members conduct the counts aloud, and in unison 3.3.2 Recommence if there is an interruption 3.3.3 Verbal confirmation of the success of the count.

3.4 Close Skin 3.4.1 The trochanter (if removed) is wired back in place. 3.4.2 Suction drains are inserted. 3.4.3 Deep tissues are repaired 3.4.4 The incision is closed with sutures and/or metal clips 3.4.5 Skin is closed.

Total Knee Replacement Surgery

1.0 Access 1.1 First Incision

1.1.1 Tourniquet cuff inflated. Runner notes time. 1.1.2 Surgeon seeks visual or verbal confirmation of readiness to start

1.1.3 Incision with scalpel or diathermy over hip joint. 1.2 Exposure of treatment area

1.2.1 Muscles and tissues are separated and spread 1.2.2 Knee capsule exposed 1.2.3 Cruciate ligaments removed. 1.2.4 Collateral ligaments released

1.2.5 Patella everted 1.2.6 Patellofemoral ligament cut with diathermy 1.2.7 Bleeding is controlled

2.0 Treatment

2.1 Femoral Preparation 2.1.1 Drill inserted into bottom of femur 2.1.2 Intramedullary rod inserted into hole 2.1.3 Femoral locating device locked in place on rod 2.1.4 Femoral locating device tapped into position 2.1.5 Distal femoral cutting block secured with drill bits or pins 2.1.6 Rod and locating device removed 2.1.7 Distal femoral cut: Osciallating blade saw placed through slot



2.1.8 Cutting block removed 2.1.9 Treatment area tidied with nibblers 2.1.10 Femoral sizing guide centred on prepared surface 2.1.11 Stylus used to check and fix position 2.1.12 Rotational alignment determined & femoral sizing noted 2.1.13 Femoral sizing guide removed 2.1.14 A/P chamfer cutting block selected (based on femoral size) 2.1.15 A/P chamfer cutting block secured with pins 2.1.16 Assistant surgeon holds cutting block firmly. 2.1.17 Anterior, posterior and chamfer cuts with oscillating saw 2.1.18 Treatment area tidied with nibblers 2.1.19 Prosthetic template hammered into place on femur 2.1.20 Lug drill advanced through appropriate holes on template. 2.1.21 Led is extended and checked. 2.1.22 Intramedullary hole is plugged with autograft tissue

2.2 Tibial Preparation 2.2.1 Knee placed in maximal flexion 2.2.2 Tibia is levered from underneath femur 2.2.3 Malleolar clamp of tibial alignment device placed 2.2.4 Upper cutting platform is positioned 2.2.5 Alignment is checked

2.2.6 Tibial stylus used to check and fix position 2.2.7 Cutting platform secured with pins or drill bits. 2.2.8 Oscillating saw used to make tibial cut 2.2.9 Cutting block and pins removed 2.2.10 Cut tidied with nibblers 2.2.11 Tibial tray selected and handle fitted. 2.2.12 Tibial tray positioned onto prepared surface 2.2.13 Prosthetic template is fixed to top of tray

2.2.13 Fitting is tested by extending leg. 2.2.14 Leg is placed in maximal flexion. 2.2.15 Tray secured with pins. 2.2.16 Handle is removed 2.2.17 Drill bushing size selected & fitted 2.2.18 Drill is used 2.2.19 Keel punch guide size is selected & fitted

2.2.20 Keel punch used 2.3 Patella resurfacing 2.3.1 Template selected 2.3.2 Leg placed in extension 2.3.3 Patella cutting guide used to grip patella 2.3.4 Resection with oscillating saw 2.3.5 Cut tidied with nibblers 2.3.5 Template used to drill holes 2.4 Insertion of Implants 2.4.1 All implant interfaces are cleaned 2.4.2 Surgeon and assistant change gloves 2.4.3 Prosthetics are obtained by runner



2.4.4 Surgeon confirms correct size of components 2.4.5 Prosthetic is unpacked and given to scrub nurse. 2.4.6 Cement is prepared. 2.4.7 Tibial implant assembled into inserter handle 2.4.8 Cement applied to tibial implant 2.4.9 Inserter handle used to place tibial component 2.4.10 Mallet blows are used to seat implant. 2.4.11 Excess cement is removed 2.4.11 Cement applied to femoral interface

2.4.12 Implant assembled into femoral inserter 2.4.13 Femoral implant inserted 2.4.14 Mallet blows are used to seat implant. 2.4.15 Cement applied to patella 2.4.16 Implant fixed to patella 2.4.17 Excess cement is removed 2.4.18 Knee is flexed to check fitting

3.0 Closure

3.1 Stabilise patient 3.1.1 Knee is fully extended. 3.1.2 Tissues are cleaned with sterile saline solution

3.2 Count Equipment 3.3.1 2 members conduct the counts aloud, and in unison 3.3.2 Recommence if there is an interruption 3.3.3 Verbal confirmation of the success of the count.

3.3 Close Skin 3.4.1 Suction drains are inserted. 3.4.2 Deep tissues are repaired 3.4.3 The incision is closed with sutures clips 3.4.4 Skin is closed & dressings applied 3.4.5 Tourniquet deflated. 3.4.6 Runner announces tourniquet time.

215

APPENDIX 3: Failure Modes and Effects Analyses


Tasks Failure Mode Effects Occurrence Severity Detection Criticality Index

(C x S x D) 1 Access

1.1 First Incision

Team left behind 7 1 3 21 1.1.1 Surgeon seeks visual or verbal confirmation of readiness to start

Surgeon starts without confirmation Critical anaesthetic tasks not

complete 3 5 2 30

1.1.2 Incision with scalpel or diathermy

Psychomotor Error Extra scarring 4 6 3 72

Procedure more difficult 3 3 3 27 1.1.3 Bleeding is controlled Bleeding continues

Greater demand for bloods 3 1 2 6

1.2 Exposure of treatment area

1.2.1 Incision is carried to periosternum

Psychomotor Error Procedure more difficult 2 3 5 30

Delay 5 1 1 5 1.2.2 Sternal saw is prepared Not prepared

configuration failure 3 1 6 18

1.2.3 Anesthetist disconnects ventilation and sternal saw is used.

Ventilation not disconnected damage to pleurae 3 4 4 48

Psychomotor Error Rupture of arteries - extensive bleeding

4 6 3 72

1.2.3a (Redo sternotomy). Sternum is split with scissors

Psychomotor Error Delay 2 1 7 14

Procedure more difficult 3 3 3 27 1.2.4 Bone wax is used to control sternal bleeding.

Bleeding continues


1.2.5 Retractor is used to slowly spread chest.

Retactor opened too fast Damage to ribs 2 6 7 84

Procedure more difficult 3 3 3 27 1.2.6 Further control of bleeding. Bleeding continues


Not removed access difficulties for surgeon

2 3 1 6

Procedure more difficult 3 3 3 27

1.2.7 Thymus is removed.

Poorly removed - bleeding


216

1.2.8 Pericardium is opened. Poorly opened (psychomotor error)

difficulty healing 2 3 3 18

1.3 Anatomical inspection

Not properly exposed Delays 3 1 2 6 1.3.1 Blood vessels are exposed and dissected according to requirements

Vessel ruptured Rupture of arteries - extensive bleeding

1 5 2 10

1.3.2 Treatment area is examined for consistency with expectation

Found to be different More difficult operation 7 2 6 84

1.4 Prepare for Treatment

Not harvested Delays while it is harvested 1 3 2 6 1.4.1 Pericardium is harvested if needed. Poor harvest extra pericardial section

required 3 2 1 6

1.4.2 Heparin is administered

Delays 5 1 2 10 1.4.2.1 Surgeon asks for heparin Not asked for

Continues to next task see 1.3.4.3

Delays 5 1 2 10 Not confirmed


1.4.2.2 Anaesthetist confirms & ensures perfusionist is present

Perfusionist not called Delays 5 1 2 10

Delays 1 5 7 35 1.4.2.3 Anaesthetist prepares and administers heparin

Failure to administer heparin


Delays 5 1 7 35 1.4.2.4 Anaesthetist announces heparin administration

Not called


Delays 3 3 7 63 1.4.2.5 Blood is taken for ACT after 3 minutes.

Not taken


Delays 3 3 3 27 1.4.2.6 ACT is confirmed to be >200.

Not confirmed

Poor awareness see 1.3.4.3

1.4.3 Cannulation

Delays 5 1 5 25 1.4.3.1 Surgeon confirms readiness to cannulate

Not confirmed

Continues to next task

Substantial blood loss 6 3 1 18

Poor perfusion 3 2 7 42

1.4.3.2 Arterial cannulation Poor cannulation

Arterial stenosis 1 6 8 48

1.4.3.3 Venous cannulation Poor cannulation Poor perfusion 5 2 7 70

1.4.4 Bypass is initiated.

217

Delays 2 3 3 18 1.4.4.1 Surgeon confirms readiness for bypass.

Not confirmed

Continues to next task

Bypass fails 1 8 1 8 Not co-ordinated

blood loss 4 5 2 40

1.4.4.2 Perfusionist & surgeon co-ordinate to initiate bypass

Poorly co-ordinated Poor bypass 2 5 3 30

poor team awareness 2 2 4 16 Not confirmed

Delays 2 3 3 18

1.4.4.3 Perfusionist confirms full flow bypass

heparin not given Patient death 2 9 8 144

1.4.4.4 Anaesthetist disconnects ventilation.

ventilation not switched off surgical difficulty 3 2 1 6

1.4.4.5 Temperature plan is made. temp plan not made Delays until plan is made 6 2 2 24

1.4.4.6 Variations to venous cannulae according to treatment.

Poor cannulation Poor perfusion 6 3 3 54

1.4.4.7 Vents and suction lines are placed.

Poor vents & suckers Poor perfusion 6 3 3 54

2 Treatment

2.1 Surgical Tasks

2.1.1 Utilise psychomotor and cognitive skills to complete surgical tasks.

tasks omitted patient not properly treated 1 8 7 56

slow treatment greater pysiological stress 2 3 7 42 2.1.2 As far as is safe, treatment should be as fast as possible.

unsafe treatment patient not properly treated 1 8 7 56

2.2 Maintenance of effective perfusion

2.2.1 Full flow is maintained where required

full flow not maintained greater pysiological stress 2 6 4 48

2.2.2 Surgical field is kept free from blood

blood in surgical field greater task difficulty 7 5 1 35

2.2.3 Reservoir is kept at a sufficient level

air embolism patient death 1 9 5 45

2.2.4 Problems resolved quickly problems recurr greater task difficulty 5 2 5 50

2.2.5 ACT checked regularly not checked blood coagulation not monitored - low awareness

2 2 5 20

2.3 Regular bloodgas check

2.3.1 Perfusionist takes bloodgas sample from perfusion system

not taken blood gas not checked - low awareness

2 2 4 16

2.3.2 ODA takes blood for analysis and brings back

not taken blood gas not checked - low awareness

3 2 2 12

218

results

2.3.3 Perfusionist and anaesthetist examine results

not examined low awareness 4 2 4 32

2.3.4 Bloodgas results are affixed to patient notes.

not affixed poor record keeping 3 1 1 3

2.4 Temperature control

2.4.1 Temperature is maintained as planned.

not maintained as planned greater pysiological stress 3 6 3 54

2.4.2 Rewarming begins once requested by surgeon.

not begun when asked delays 5 1 2 10

temperature too low greater pysiological stress 3 6 2 36 2.4.3 Temperature is 36°C prior to weaning from bypass. temperature too high greater pysiological stress 2 3 3 18

2.5 Preparation for weaning from bypass

2.5.1 Heart is de-aired. not de-aired correctly air embolism 2 7 5 70

2.5.2 Heart is observed to be beating.

not beating before weaning from bypass

return to bypass 2 7 1 14

2.5.3 Pacing wires attached. not attached poor cardiac pacing 2 5 2 20

2.5.4 Filtration (where required) has been completed.

not complete greater pysiological stress 7 3 3 63

2.5.5 Ventilation is on. not on sats reduce 2 2 2 8

2.5.6 Sufficient ionotrope support is available

not given greater pysiological stress 2 5 3 30

2.5.8 Confirmation that team is ready to wean patient from bypass.

no confirmation any of above failures 6 1 6 36

2.6 Cross-clamp

2.6.1 Cross clamp is applied. application not announced low awareness 5 2 3 30

2.6.2 Cross clamp is removed. removal not announced low awareness 5 2 4 40

2.7 Cardioplegia administration

2.7.1 Cardioplegia is primed not primed greater pysiological stress 2 5 3 30

2.7.2 Cardioplegic needle is inserted

poor insertion: poor delivery of cardioplegia

greater pysiological stress 2 4 6 48

2.7.3 Cardioplegia is run according to dosage required by surgeon

incorrect dosage greater pysiological stress 1 5 7 35

2.7.4 Plan is made for next cardioplegic cycle

no plan low awareness 4 1 5 20

2.8 DHCA Requirements

2.8.1 Target temperature (<25°C) is reached.

not reached greater pysiological stress 1 5 2 10

219

2.5.2 Patient has been on bypass for 20 mins.

not on bypass for long enough

greater pysiological stress 1 5 4 20

2.8.2 Ice packs used at head to minimise neurological damage.

ice packs not used neurological damage 1 6 3 18

2.8.3 Perfusionist periodically announces time on DHCA.

DHCA time not announced low awareness 5 1 2 10

3 Closure

3.1 Stabilise patient

3.1.1 Weaning from bypass

3.1.1.1 Perfusionist and surgeon co-ordinate reduction in flow

poor co-ordination greater pysiological stress 3 3 6 54

3.1.1.2 Anesthetic alarms switched on.

no alarms low awareness 3 2 4 24

3.1.1.3 Venous perfusion line is clamped

not clamped ex-sanguination 2 5 3 30

3.1.1.4 Blood is transfused. poorly transfused low BP / Delay 3 3 2 18

bleeding 3 3 3 27 3.1.1.5 Venous cannulae are removed

poorly

stenosis 2 6 5 60

3.1.2 Modified Ultrafiltration

3.1.2.1 MUF plan is agreed no plan agreed poor awareness / delays 7 2 4 56

3.1.2.2 Hotline is placed in atrial appendage

not placed in atrial appendage

exsanguination 2 4 4 32

3.1.2.3 Filtration begins not begun delay 3 1 3 9

3.1.2.4 Bloodgas is used to check HCT

not checked poor awareness 3 3 3 27

too soon HCT low: greater physiological stress

3 4 3 36 3.1.2.5 Filtration finishes at right time

late delay 3 1 5 15

3.1.3 General Stability

3.1.3.1 Sinus rhythm is achieved not achieved delay & diagnostics 6 3 1 18

3.1.3.2 Suitable hemodynamic stability is achieved

not achieved delay & diagnostics 6 3 1 18

3.1.3.3 Oxygen saturations acceptable

not acceptable delay & diagnostics 6 3 1 18

3.1.3.4 Bleeding is controlled not controlled delay & diagnostics 6 3 1 18

3.1.3.5 Acidosis is controlled not controlled delay & diagnostics 6 3 1 18

3.1.3.6 Chest drains inserted poorly functioning Poor recovery 3 2 1 6

3.1.3.7 Peritoneal dialysis catheter inserted

poorly functioning Poor recovery 2 2 1 4

220


3.2.1 Echocardiographic investigation

not performed poor awareness 6 3 1 18

3.2.2 Bleeding sites examined not examined poor awareness 3 3 3 27

3.3 Count Equipment

3.3.1 Count before closure of chest cavity

3.3.1.1 2 members conduct the counts aloud, and in unison

mis-count equipment left in 1 3 3 9

3.3.1.2 Recommence if there is an interruption


3.3.1.3 Verbal confirmation of the success of the count.


3.3.2 Count before skin closure

3.3.2.1 2 members conduct the counts aloud, and in unison


3.3.2.2 Recommence if there is an interruption


3.3.2.3 Verbal confirmation of the success of the count.


3.4 Skin Closure

3.4.1 Arterial cannula is removed. poorly removed aterial stenosis 2 6 5 60

3.4.2 Heparin reversal

3.4.2.1 Surgeon asks for protamine not asked for Delays 3 1 2 6

3.4.2.2 Protamine is given by anaesthetist

not given Still heparinised 2 2 1 4

3.4.2.3 Confirmation of protamine administration by anaesthetist

not confirmed poor awareness 4 1 4 16

not checked poor awareness 3 1 3 9 3.4.2.4 ACT returns to pre-operative level not returned more protamine required 1 3 4 12

not observed poor awareness 1 3 7 21 3.4.3 Clots observed to be forming

not forming delays / more bloods required

4 3 3 36

3.4.4 Retractor removed poorly removed sternal compression 1 4 3 12

3.4.5 Sternal Closure

3.4.5.1 Sternum closed with heavy sutures

poorly closed poor recovery 2 2 1 4

3.4.5.2 Deep fascia closed poorly closed poor recovery 2 2 1 4

3.4.5.3 Subcutaneous tissue closed poorly closed poor recovery 2 2 1 4

221

3.4.5.4 Skin closed poorly closed poor recovery 1 2 1 2

3.4.5a Delayed Sternal Closure

3.4.5a.1 Two stents applied crossways

poorly placed poor recovery 2 2 5 20

3.4.5a.2 Stents sutured to sternum poorly secured poor recovery 2 2 5 20

3.4.5a.3 Transparent window stitched to wound

poorly secured poor recovery 3 2 2 12

222

Total Hip Replacement Surgery

Task Failure Mode Effects Occurrence Severity Detection Criticality Index

(C x S x D) 1.0 Access

1.1 First Incision

1.1.1 Surgeon seeks visual or verbal confirmation of readiness to start

Surgeon starts without confirmation

Team left behind 5 1 3 15

Critical tasks not complete 2 2 2 8

1.1.2 Incision with scalpel or diathermy over hip joint.


1.1.3 Bleeding is controlled Bleeding continues Procedure more difficult 2 3 3 18



1.2.1 ligaments and muscles are spread or incised

Damage to tissues Difficult recovery 2 4 5 40

1.2.2 Tissues prepared for later repair. Not prepared Delay 2 2 2 8

1.2.3 Damage to femur Difficult recovery 1 5 7 35

Unnecessary removal More difficult recovery 1 5 7 35

Trochanter with its muscle attachments removed .

Not removed when necessary


1.2.4 Bleeding is controlled Bleeding continues Procedure more difficult 2 3 3 18


2.0 Treatment

2.1 Remove Femoral Head

2.1.1 Hip capsule opened. Damage to joint Procedure more difficult 1 5 5 25

2.1.2 The femoral head is dislocated. Damage to femur/acetabulum


2.1.3 Power saw is prepared Not prepared Delay 5 1 2 10

2.1.4 Femoral neck removed. Psychomotor Error Damage to femur 2 5 3 30

2.2 Prepare Acetabulum

2.2.1 Reamer is prepared using the correct size head.

Incorrect size head Fitting problems 2 4 5 40

2.2.2 Reamer shapes acetabulum to receive prosthetic.

Not completed properly Fitting problems 2 4 4 32

2.2.3 Trial component is used to check size and fit


223

2.2.4 Interface is cleaned, irrigated and dried

Not completed properly Infection 2 7 2 28

2.3 Insert Acetabular Component

2.3.1 Prosthetic is prepared

2.3.1.1 Prosthetic is obtained by runner Incorrect size Delay 5 1 3 15

2.3.1.2 Surgeon confirms correct size. Incorrect size resource wastage/delay 5 1 2 10

2.3.1.3 Prosthetic is unpacked and given to scrub nurse.

Incorrect size resource wastage/delay 5 1 2 10

2.3.2 Cement preparation and application

2.3.2.1 Scrub nurse mixes cement. Incorrect cement resource wastage/delay 4 1 2 8

Cement not cured Fitting problems 2 5 4 40 2.3.2.2 Cement is allowed to partially cure.

Cement not mixed at right time

Delay 3 1 1 3

2.3.2.3 Cement applied to treatment area. poorly applied Poor joint function 2 5 4 40

2.3.3 Acetabular prosthetic is inserted into place.

Pooly inserted Poor joint function 2 5 4 40

Used when not needed Delay 2 2 4 16 2.3.4 Additional screws are used to hold the component firmly to the bone. Not used when needed Poor joint function 2 4 5 40

2.4 Prepare Femoral Canal

2.4.1 Rasps of appropriate size are prepared

Incorrect size Procedure more difficult 2 5 3 30

2.4.2 Femur is shaped using rasps Incorrect shape Procedure more difficult 2 5 3 30

2.3.3 Femur is hollowed out Poorly hollowed Poor seating of prosthetic 2 5 3 30

2.3.4 Trial component is used to check size and fit


2.3.5 Movement of the hip joint is tested. Not completed properly Poor joint function 2 5 7 70

2.5 Insert Femoral Stem

2.5.1 Prosthetic is prepared

2.5.1.1 Prosthetic is obtained by runner Incorrect size Delay 5 1 3 15

2.5.1.2 Surgeon confirms correct size. Incorrect size resource wastage/delay 5 1 2 10

2.5.1.3 Prosthetic is unpacked and given to scrub nurse.


2.5.2 Intramedullary canal is plugged Not plugged Cement mixes with marrow 2 5 7 70

2.5.3 Cement loaded into gun. Poorly loaded Poor cement delivery 3 3 6 54

2.5.4 Cement injected retrogradely into femoral canal with gun.

Poor cement delivery Poor seating of prosthetic 2 6 6 72

2.5.5 Stem is inserted into the femoral canal.

Poorly completed Poor seating of prosthetic 2 6 6 72

224

2.5.6 Stem is centralised in the cement mantle.

Not completed properly Poor seating of prosthetic 2 6 6 72

3.0 Closure


3.1.1 Tissues are cleaned with sterile saline solution

Not completed Infection 2 7 2 28


3.1.1 Surgeon ensures the integrity of the hip abductor mechanism

Not completed properly Poor joint function 2 7 2 28

3.1.2 Iliotibial band is checked Not completed properly Muculo-seletal complications 2 7 3 42

3.1.3 Leg length is checked. Not completed properly Musculo-seletal complications

2 7 3 42

3.1.4 X-ray to check prosthesis is in the correct position

Not completed properly Poor joint function 2 7 3 42

3.3 Count Equipment

3.3.1 2 members conduct the counts aloud, and in unison


3.3.2 Recommence if there is an interruption


3.3.3 Verbal confirmation of the success of the count.


3.4 Close Skin

3.4.1 The trochanter (if removed) is wired back in place.

Not completed Musculo-seletal complications

2 7 3 42

3.4.2 Suction drains are inserted. Not completed properly More difficult recovery 2 4 2 16

3.4.3 Deep tissues are repaired Not completed properly More difficult recovery 3 4 3 36

3.4.4 The incision is closed with sutures and/or metal clips


3.4.5 Skin is closed. Not completed properly Infection 2 6 2 24

225

Total Knee Replacement Surgery

Tasks Failure Mode Effects Occurrence Severity Detection Criticality Index

(C x S x D) 1.0 Access

1.1 First Incision

1.1.1 Tourniquet cuff inflated. Runner notes time.

Time not noted No measure of TT 4 1 8 32

Team left behind 5 1 3 15 1.1.2 Surgeon seeks visual or verbal confirmation of readiness to start

Surgeon starts without confirmation Critical tasks not complete 2 2 2 8

1.1.3 Incision with scalpel or diathermy over knee joint.



1.2.1 Muscles and tissues are separated and spread


1.2.2 Knee capsule exposed Damage to tissues Difficult recovery 2 3 5 30

1.2.3 Cruciate ligaments removed. Damage to tissues Difficult recovery 2 3 5 30

1.2.4 Collateral ligaments released Damage to tissues Difficult recovery 2 3 5 30

1.2.5 Patella everted Damage to tissues Difficult recovery 2 3 5 30

1.2.6 Patellofemoral ligament cut with diathermy


Procedure more difficult 2 3 3 18 1.2.7 Bleeding is controlled Bleeding continues


2.0 Treatment

2.1 Femoral Preparation

2.1.1 Drill inserted into bottom of femur poor fitting poor implantation 2 5 4 40

2.1.2 Intramedullary rod inserted in hole

2.1.3 Femoral locating device locked in place on rod

2.1.4 Femoral locating device tapped into position

poorly fitted poor implantation 2 5 4 40

2.1.5 Distal femoral cutting block secured with drill bits or pins


2.1.6 Rod and locating device removed

2.1.7 Distal femoral cut: Osciallating blade saw placed through slot

poor cut poor implant 2 5 6 60

2.1.8 Cutting block removed

226

2.1.9 Treatment area tidied with nibblers not tidied properly discomfort/infection 2 3 4 24

2.1.10 Femoral sizing guide centred on prepared surface


2.1.11 Stylus used to check and fix position

poorly checked poor implantation 2 5 4 40

2.1.12 Rotational alignment determined & femoral sizing noted

not determined/noted wrong size implantation 3 4 3 36

2.1.13 Femoral sizing guide removed

2.1.14 A/P chamfer cutting block selected (based on femoral size)

Wrong size poor cut 2 3 4 24

2.1.15 A/P chamfer cutting block secured with pins

poorly fitted poor cut 2 3 4 24

2.1.16 Assistant surgeon holds cutting block firmly.

not secured poor cut 4 3 4 48

2.1.17 Anterior, posterior and chamfer cuts with oscillating saw

poor cut poor implantation 2 5 6 60

2.1.18 Treatment area tidied with nibblers not tidied properly Discomfort/infection 2 3 4 24

2.1.19 Prosthetic template hammered into place on femur

badly hammered damage to femur 2 5 3 30

2.1.20 Lug drill advanced through appropriate holes on template.

poor drilling poor implantation 2 5 4 40

2.1.21 Led is extended and checked. not/poorly checked poor implantation 2 5 4 40

2.1.22 Intramedullary hole is plugged with autograft tissue

not plugged Difficult recovery 2 3 5 30

2.2 Tibial Preparation

2.2.1 Knee placed in maximal flexion

2.2.2 Tibia is levered from underneath femur

badly levered damage to femur/tissues 2 4 3 24

2.2.3 Malleolar clamp of tibial alignment device placed

badly placed poor implantation 2 5 6 60

2.2.4 Upper cutting platform is positioned


2.2.5 Alignment is checked poorly checked poor implantation 2 5 4 40

2.2.6 Tibial stylus used to check and fix position

poorly checked poor implantation 2 5 4 40

2.2.7 Cutting platform secured with pins or drill bits.


2.2.8 Oscillating saw used to make tibial cut

poor cut poor implantation 2 5 6 60

2.2.9 Cutting block and pins removed

2.2.10 Cut tidied with nibblers not tidied properly discomfort/infection 2 3 4 24

2.2.11 Tibial tray selected and handle fitted.

227

fitted.

2.2.12 Tibial tray positioned onto prepared surface


2.2.13 Prosthetic template is fixed to top of tray

2.2.13 Fitting is tested by extending leg. not/poorly checked poor implantation 2 5 4 40

2.2.14 Leg is placed in maximal flexion.

2.2.15 Tray secured with pins. poorly fitted poor keel punch 2 3 4 24

2.2.16 Handle is removed

2.2.17 Drill bushing size selected & fitted incorrect bushing poor drilling 2 3 4 24

2.2.18 Drill is used poor drilling poor implantation 2 5 4 40

2.2.19 Keel punch guide size is selected & fitted

poor sizing poor keel punch 2 3 4 24

2.2.20 Keel punch used poor keel punch poor implantation 2 6 4 48

2.3 Patella resurfacing

2.3.1 Template selected incorrect template poor drilling 2 3 4 24

2.3.2 Leg placed in extension

2.3.3 Patella cutting guide used to grip patella

gripped badly poor cut

2.3.4 Resection with oscillating saw poor cut poor implantation 2 5 6 60

2.3.5 Cut tidied with nibblers not tidied properly discomfort/infection 2 3 4 24

2.3.5 Template used to drill holes poor drilling poor implantation 2 5 6 60

2.4 Insertion of Implants

2.4.1 All implant interfaces are cleaned not cleaned infection 2 6 4 48

2.4.2 Surgeon and assistant change gloves

not changed infection 2 6 4 48

2.4.3 Prosthetics are obtained by runner Incorrect size discomfort/infection 5 1 3 15

2.4.4 Surgeon confirms correct size of components


2.4.5 Prosthetic is unpacked and given to scrub nurse.


2.4.6 Cement is prepared. poorly prepared resource wastage/delay 4 1 4 16

2.4.7 Tibial implant assembled into inserter handle

2.4.8 Cement applied to tibial implant poorly applied poor implantation 2 5 6 60

2.4.9 Inserter handle used to place tibial component

2.4.10 Mallet blows are used to seat implant.

228

implant.

2.4.11 Excess cement is removed cement left discomfort/infection 2 3 4 24

2.4.11 Cement applied to femoral interface

poorly applied poor implantation 2 5 6 60

2.4.12 Implant assembled into femoral inserter

2.4.13 Femoral implant inserted

2.4.14 Mallet blows are used to seat implant.

2.4.15 Cement applied to patella

2.4.16 Implant fixed to patella poorly applied poor implantation 2 5 6 60

2.4.17 Excess cement is removed cement left discomfort/infection 2 3 4 24

2.4.18 Knee is flexed to check fitting Not checked fitting incorrect 2 3 3 18

3.0 Closure


3.1.1 Knee is fully extended.

3.1.2 Tissues are cleaned with sterile saline solution


3.2 Count Equipment

3.3.1 2 members conduct the counts aloud, and in unison


3.3.2 Recommence if there is an interruption


3.3.3 Verbal confirmation of the success of the count.


3.3 Close Skin

3.4.1 Suction drains are inserted. Not completed properly More difficult recovery 2 4 2 16

3.4.2 Deep tissues are repaired Not completed properly More difficult recovery 3 4 3 36

3.4.3 The incision is closed with sutures clips


3.4.4 Skin is closed & dressings applied Not completed properly Infection 2 6 2 24

3.4.5 Tourniquet deflated. Not completed properly longer TT 1 2 1 2

3.4.6 Runner announces tourniquet time.

Wrong TT poor info 4 1 8 32


12 August 2005 Final Draft Version 2.0 229

APPENDIX 4: Sample data collection packs

Data Collection Pack for the Arterial Switch Operation

IMPROVING SAFETY IN THE OPERATING THEATRE PEADIATRIC CARDIAC SURGERY CASE SUMMARY SHEET

Date of procedure (DD/MM/YY): __ /__ /__

Video Start Time _________ Video End Time _________

Observation Start Time _________ Observation End Time _________

Patient’s name, age and hospital number (or attach sticker):

______________________________________________________________________

Diagnosis / Procedure: ______________________________________________________________________

Other relevant medical conditions:

______________________________________________________________________

Previous surgery: ______________________________________________________________________

Weight (kg): _____ Number of previous sternotomies: o

Gore-tex membrane at previous surgery: (circle one) YES NO

Time Allocated _________

Time into anaesthetic room (24hr clock) _________

Time into theatre (24hr clock) _________

Surgical start time _________ Surgical end time _________

CPB start time _________ CPB end time _________

X-clamp start time _________ X-clamp end time _________

DHCA start time _________ DHCA end time _________

Heparin in at _________

lowest temp (oC) ________ Rewarm begins at _________

Time arrived in CICU (24hr clock) _________

Surgeon ________ Anaesthetic Consultant ________

1st Assistant ________ Anaesthetic Registrar ________

2nd Assistant ________ Perfusionist ________

Scrub Nurse ________ Observer KC PG



IMPROVING SAFETY IN THE OPERATING THEATRE PEADIATRIC CARDIAC SURGERY POST-OPERATIVE DATA COLLECTION

Date of procedure (DD/MM/YY): __ /__ /__ Patient’s name, age and hospital number (or attach sticker):

Known airway problem (circle one): YES NO

If yes, define: ________________________________________________________

Known access problem e.g. blocked femorals (circle one) YES NO

If yes, define: _________________________________________________________

Echo On Table: None Planned Unplanned PIMS Score _______________ ECMO (circle one) YES NO Chest? (circle one) Open Closed Extubated on table YES NO

Anaesth SpR



SOURCE PROCEDURE CHECKLIST FOR THE ARTERIAL SWITCH OPERATION Procedural Prerequisites The following are also being conducted simultaneously:

Arch hypoplasia or interruption repair Arterial switch VSD Closure RVOTO resection (or atrial septectomy for Double inlet ventricle)

Perfusion (one): Simple TGA: 20°C - 22°C Complex: 15°C -18°C Perfusion Pressures: 25-35mmHg for infants 2-3kg (mark one) 35-45mmHg for infants 3-5kg 45-55mmHg for infants 5-10kg Anaesthesia (all) Temperature probes: nasal, oesophageal & toe Surface electrocardiagraphic electrodes Heating/Cooling blanket Head turned to left

: : : : :

Time + Check



Midline sternotomy8 (see MIDLINE STERNOTOMY procedure) Exposure Silk ligatures stayed to pericardium for harvesting

Anterior Pericardium harvested in two rectangles Aorta mobilised & separated from the main PA to the level of the coronary ostia R and L pulmonary arteries and duct are partially dissected out IF Arch reconstruction: all arch vessels dissected and upper descending aorta mobilised.

Normal Arterial Cannulation: see CANNULATION procedure

IF Arch reconstruction: 2nd arterial cannula placed in the proximal duct & Y-connector used FOR SIMPLE CASES: Rygg cannula placed through R atrial appendage.

FOR COMPLEX CASES: double venous cannulation

Ductus arteriosis dissected out and divided between 2-0 silk ligatures Aortic end secured with 7-0 prolene purse string Right and Left pulmonary arteries surrounded with vessel loops Dissected out until branches clearly seen. EITHER: Implantation sites of coronary arteries are marked with 6-0 prolene suture OR masking suture placed higher circumflex artery arises from the right facing sinus.

ON BYPASS (Refer to separate procedure) IF NOT Arch reconstruction

Cardioplegia is administered IF Arch reconstruction

Ductus arteriosis ligated proximal to ductal cannula. Duct, aortic arch & upper descending aorta mobilised. At temp of 18°C or less, bypass flows reduced to 10% Arch-vessel snares are tightened Ductal cannula clamped & removed Cross-clamp is placed distal to the aortic cannula. All ductal tissue excised Aorta anastomosed to underside of proximal arch. Clamp shifted proximal to aortic cannula Cardioplegia is administered Full bypass is reinstated.

Cardioplegic needle is removed

Cardioplegia plan is negotiated between perfusionist and surgeon Cross- clamp applied VP: Surgeon announces x-clamp Ascending aorta is opened anteriorly Inspected for coronary ostia before transaction is completed

8 Removal of left thymus. Possible removal or right thymus; see midline sternotomy procedure

:

:

:

:

Notes



Excised with a D-shape cuff.9 Excision of Coronary Ostia IF Yacoub type B: two separate cuffs if ostia separated by >2mm.

IF NOT Yacoub type B: Ostia cuffs Left together MPA transacted 2mm above marker sutures Operable LVOTO is resected Hinged trapdoor flaps created in proximal MPA Coronary artery cuffs sewn with 7-0 Prolene into trapdoor flaps IF flap is too long, 1 mm trimmed off before suture line completed

IF Yacoub type B EITHER Both arteries excised together

D-shaped cuff sewn to adjacent edge of proximal MPA Pericaridal patch sewn to part of cuff to create pocket. Ascending aorta sewn to anterior wall of pocket.

OR Coronary arteries left in situ. Aortic wall excised to within 1 mm of ostia. Excision of MPA wall Aorta and MPA sewn together D-shaped cuff excised from aortic sinus, but left hinged Curved part of hinge sewn into the coronary-artery sinus… …creating a pocket Ascending aorta anastomosed to front edge of pocket.

Lecomte manuovre: pulmonary artery brought in front of ascending aorta

Cardioplegic purse-string passed behind PA Purse-string pulled to hold bifurcation against cross-clamp Neo-Aorta

Proximal MPA (bearing coronary arteries) anastomosed to ascending aorta Cardioplegic purse-string passed back up under the MPA bifurcation

Aortic cross-clamp removed VP: Surgeon announces x-clamp removal Major leakages identified Inspection of heart: should start slowly beating Coronary artery branches observed for filling Mycocardium observed to become pink10 Adjustment may be required Massaging may also be required MPA gathered into posterior wall of aorta (some possible variations with complexity)

9 Eccentrically placed ostia treated on their merits 10 Failure in pinkness means likely kinking

Reimplanation of Coronary Arteries

Notes



Coronary sinus sucked placed through aortic valve Repair of Aortic Sinus Defects D-shaped defects treated with pericardial patches11 INTRACARDIAC REPAIR NOTE: Repairs Differ for SIMPLE and COMPLEX TGA

Intracardiac Repair for SIMPLE TGA Aortic Cross-clamp reapplied at 22°C VP: Surgeon announces x-clamp Cardioplegia applied for 2mins (see APPLYING CARDIOPLEGIA procedure) CPB is discontinued Blood is drained from patient Venous cannula removed Coronary sinus sucker placed through right atrial appendage Right atrium opened low down ASD closed

Intracardiac Repair for COMPLEX TGA Lowering temp to 18°C is started during sinus repair

Cross-clamp reapplied VP: Surgeon announces x-clamp Cardioplegia applied for 2mins (see APPLYING CARDIOPLEGIA procedure) CPB is discontinued Blood is drained from patient Venous cannula removed Coronary sinus sucker placed through right atrial appendage RVOTO attacked through aortic root VSD(s) are closed via tricuspid valve with Dacron ASD closed with direct suture

IF Bicaval cannulation: repair conducted on full flow bypass at 22°C CPB reinstated Completion of Intracardiac Repair

Re-warming Heart is de-aired

Cross-clamp removed VP: Surgeon announces x-clamp removal Vessel loops on right and left pulmonary arteries are relaxed Heart is observed for adequate filling & perfusion Right Atrium closed Venous cannula reinserted Pericardial patches trimmed12 Anastamosis between proximal aorta & distal MPA

11 For complex TGA, cooling to 18°C is begun (see Intracardic Repair for COMPLEX TGA) 12 Note: generous pericardial patches important

Notes



Pressure monitor placed in: left atrial appendage Completion of Treatment

Pressure monitor placed in neo-MPA Both pressure monitors tied to heart

RV and RA pacing wires sewn in On rewarming, dopamine infused, CPB discontinued (see OFF BYPASS procedure) Heart observed for perfusion L atrial pressure is monitored13

Control of Bleeding Major bleeding may need further CPB and take down of neo-MPA anastomosis Multiple small bleeding sites managed by use of foam pieces, removed before closure

GOTO Closure of Midline Sternotomy

13 Note 1: myocardic ischemia (poor colour or LA pressure rise) is observed before electrocardiographic changes are noted. Note 2: Rise in LA pressure above 10mmHg requires careful scrutiny of myocardium and coronary vessels

Notes



Opening with Midline Sternotomy Patient placed supine Workspace Preparation Head positioned straight Towel/Supports Underneath Diathermy pads applied Incision line marked with pen Skin preparation: betadine applied Green drapes on

Steri-tape (or other) applied to chest, neck & abdomen Drape barrier set up Light handle covers on Perfusion pipe organiser set up COMMS PROTOCOL: Surgeon: “Okay to start?” Confirmation from anaesthetist First incision with scalpel Incision carried to periosternum Exposure of Sternum Bleeding controlled with diathermy COMMS PROTOCOL: Surgeon: “Stop ventilating.” (or similar) Sternum Split Anasethetist stops ventilation Sternum split with electric saw COMMS PROTOCOL: 1st A: “Start ventilating again” Ventilation started. Bleeding controlled with diathermy & bone wax Retractor inserted Bar inserted superiorly Retractor opened Gradually (to avoid sternal fracture) Pleurae pushed to sides Thymus resected Exposure of Treatment Area Pericardium Opened on midline OR opened from right side (pericardial patch) Edges of pericardium suspended to edges of wound

:

: :

:

:

Notes



Cannulation Preparation for Cannulation

Pericardium between ascending aorta and MPA divided with diathermy. Visceral pericardium overlying RPA is divided just to left of the SVC Surgeon/Assistant Surgeon checks for readiness to Cannulate Anaesthetist responds positively Perfusionist responds positively At some time during this period:

Surgeon: “Heparin in” Anaesthetist read-back Anaesthetist ensures Perfusionist has heard Perfusionist takes blood for ACT Anaesthetist checks ACT ACT >450

Arterial Cannulation

Surgeon decides on cannula size or asks Perfusionist for tables Size of cannula indicated to SN Purse-string in aorta Snugger is snared Scalpel blade advanced through aortic wall within purse-string Flap used to control haemorrhage Cannuala slid into incision Purse string stutures snugged Cannuala tied to snugger Air bubbles tapped out of cannula before connection Perfusionist slowly advances blood through arterial perfusion line Arterial perfusion line clamped at patient end Arterial perfusion line is snipped off just above clamp. Cannula connected to perfusion pipe Cannual tied to snugger Cannula/snugger positioned away from surgical field

Single Venous Cannulation Purse string around atrial appendage Snugger is snared Either Forceps applied to base of appendage & tip cut off Or Scalpel is advanced through purse string & enlarged with scissors Cannula positioned Purse strings snugged Cannula connected to perfusion pipe

Cannula tied to snugger Snugger/Cannulae positioned away from surgical field. Workspace & pipes organised.



Indirect Bicaval Cannulation

Purse string suture in right atrial appendage SVC Cannula Stab incision Cannula inserted14 Pressure on abdomen Blood passes up cannula and down into the Y-connector Cannula attached to pipe On bypass (See “Going on Bypass” Subtask Purse string low in right atrium close to mouth of IVC IVC cannuala Stab incision Cannula inserted15 Cannulae snugged Umbilical tape/Silk Ligature passed around IVC

Direct Bicaval Cannulation

Purse string position on SVC Scalpel blade placed through purse-string SVC Cannula Mosquito clamp or prolene suture controls haemorrhage Mosquito forceps dilate incision Cannula passed into SVC Clamp or suture removed Snuggers snared Cannula attached to pipe Cannula tied to snugger On bypass (See “Going on Bypass” Subtask

Purse string into IVC below pericardial reflection Direct stab incision IVC cannula

Cannula positioned Snuggers snared Cannula tied to snugger

14 Care to avoid pushing tip through atrial wall or through an atrial septal defect. 15 IVC cannula should be short enough to avoid obstructing venous return from liver. Surgeon must avoid catching cannula in Eustachian valve



Variation of Direct Bicaval Cannulation Purse-string on right atrial appendage IVC Cannula Venous cannula connected to Y-connector Purse string around atrial appendage

EITHER: Forceps around base of appendage Tip cut off IVC cannular positioned Forceps released

OR Stab wound in centre of purse string Wound enlarged with pair of scissors.

Purse string position on SVC 0.5-1cm above junction with Right atrium16 Scalpel blade placed through purse-string SVC Cannula Mosquito clamp or prolene suture controls heamorrage Mosquito forceps dialate incision Cannula passed into SVC Clamp or suture removed Snuggers snared Cannula tied to snugger On bypass (See “Going on Bypass” Subtask IVC cannula line clamped IVC Cannula moved Sucker placed in R atrial appendage IVC purse string inserted Cannular positioned in IVC IVC cannula line unclamped

Off Bypass

At end of bypass, SVC cannulation site repaired tying purse-string, or sewing within purse-string.

16 SVC narrowing serious potential complication if position not correct



Closing Midline Sternotomy Control of bleeding Pacing Wires in Chest Drains Retractor removed Sternum closed with heavy sutures (wire/synthetic material) Deep fascia closed Subcutaneous tissue closed Skin closed Wound painted with betadine Wound is dried Dressings applied Green sheets taken down Bed in

IF Delayed closure

2x Stents crossways Sutured to bone

Transparent window stitched onto wound Dressings placed over window



Data Collection Pack for the Total Knee Replacement Operation

IMPROVING SAFETY IN THE OPERATING THEATRE TOTAL KNEE REPLACEMENT CASE SUMMARY SHEET

Date of procedure (DD/MM/YY): __ /__ /__ Observation Start Time _________ Observation End Time _________ Patient Hospital Number:____________ MALE FEMALE DoB: :____________

Diagnosis / Procedure:

_____________________________________________________________________

Other relevant medical conditions:

____________________________________________________________________

Previous surgery:

Time into theatre (24hr clock) _________

Surgical start time ________ Surgical end time ________ Tourniquet Time _______

Time out of theatre (24hr clock) _________

Surgeon ________ Anaesthetist ________ Theatre Number _______ Assistant ________ Scrub Nurse ________ Sales Rep ________ Runner ________ Observers: KC PG Others ________________________

Implant Type: Scorpio PFC Uncemented Computer Navigation: YES NO Anaesthetic Type: Local General

Anaesthetic problems (circle one): YES NO

If yes, define: ________________________________________________________

Pre-Operative Briefing NONE SURGEONS ONLY WHOLE TEAM

EXPECTATIONS THREATS

Post-Operative Briefing NONE SURGEONS ONLY WHOLE TEAM

SUCCESSES FAILURES



1.0 Access 1.1 Preparation

Tourniquet cuff fixed and inflated. Runner notes time.

………………………………………………………………………………………

Leg painted with betadine

………………………………………………………………………………………

Leg cleaned with alcohol

………………………………………………………………………………………

Blue cover placed over leg

………………………………………………………………………………………

Elastic bandage on leg

………………………………………………………………………………………

Drapes used to cover patient

………………………………………………………………………………………

Blue cover opened

………………………………………………………………………………………

Microban film applied

………………………………………………………………………………………

Light handle covers on

………………………………………………………………………………………

Diathermy connected and configured

………………………………………………………………………………………

Interruptions, Distractions & Absences

………………………………………………………………………………………

………………………………………………………………………………………

Theatre Discipline & Safety Awareness

……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:



1.2 First Incision

Surgeon seeks visual or verbal confirmation of readiness to start

………………………………………………………………………………………

Incision with scalpel or diathermy over knee joint.

………………………………………………………………………………………


Muscles and tissues are separated and spread

………………………………………………………………………………………

Knee capsule exposed

………………………………………………………………………………………

Cruciate ligaments removed.

………………………………………………………………………………………

Collateral ligaments released

………………………………………………………………………………………

Patella everted

………………………………………………………………………………………

Patellofemoral ligament cut with diathermy

………………………………………………………………………………………

Bleeding is controlled

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



Access LEADERSHIP & MANAGEMENT Leadership Involves / Reflects on suggestions / Visible / Accessible / Inspires / Motivates / Coaches Maintenance of standards Subscribes to standards / Monitors compliance to standards / Intervenes if deviation

/ Deviates with team approval / Demonstrates desire to achieve high standards Planning & preparation Team participation in planning / Plan is shared / Understanding confirmed / projects / changes in consultation Workload management Distributes tasks / monitors / reviews / tasks are prioritised / Allots adequate time / responds to stress Authority & assertiveness Advocates position / Values team input / Takes control / Persistent / Appropriate assertiveness TEAMWORK & CO-OPERATION Team building/ maintaining Relaxed / Supportive / Open / Inclusive / Polite / Friendly / Use of humour / Does not compete Support of others Helps others / Offers assistance / gives feedback Understanding team needs Listens to others / Recognises abilities of team / Condition of others considered / Gives personal feedback Conflict solving Keeps calm in conflicts / Suggests conflict solutions / Concentrates on what is right

PROBLEM SOLVING & DECISION MAKING Definition & Diagnosis Uses all resources / Analytical decision making / Reviews factors with team Option Generation Suggests alternative options / Asks for options / Reviews outcomes / Confirms opinions Risk Assessment Estimates risks / Considers risk in terms of team capabilities / Estimates patient outcome Outcome Review Reviews outcomes / Reviews new options / Objective, constructive and timely reviews / Makes time for review

/ Seeks feedback from others / Conducts post treatment review

SITUATION AWARENESS Notice Considers all elements / Monitors vital signs & progress of operation / Asks for or shares information

/ Encourages vigilance / Checks and reports changes / Requests reports and updates

Understand Cross-checks above / Shares mental models / Speaks up when unsure / Updates other team members

Think Ahead Identifies future problems / Discusses contingencies / Plans for future patient states / Anticipates high workload / Discusses constraints / Uses low workload periods

Below Standard Basic Standard Standard Exceed Behaviour directly

compromises patient safety and effective teamwork.





Teamwork & Co-Operation


Situation Awareness



2.0 Treatment

2.1 Femoral Preparation

Drill inserted into bottom of femur

………………………………………………………………………………………

Intramedullary rod inserted into hole

………………………………………………………………………………………

Femoral locating device locked in place on rod

………………………………………………………………………………………

Femoral locating device tapped into position

………………………………………………………………………………………

Distal femoral cutting block secured with drill bits or pins

………………………………………………………………………………………

Rod and locating device removed

………………………………………………………………………………………

Distal femoral cut: Osciallating blade saw placed through slot

………………………………………………………………………………………

Cutting block removed

………………………………………………………………………………………

Treatment area tidied with nibblers

………………………………………………………………………………………

Femoral sizing guide centred on prepared surface

………………………………………………………………………………………

Stylus used to check and fix position

………………………………………………………………………………………

Rotational alignment determined & femoral sizing noted

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



Femoral sizing guide removed

………………………………………………………………………………………

A/P chamfer cutting block selected (based on femoral size)

………………………………………………………………………………………

A/P chamfer cutting block secured with pins

………………………………………………………………………………………

Assistant surgeon holds cutting block firmly.

………………………………………………………………………………………

Anterior, posterior and chamfer cuts with oscillating saw

………………………………………………………………………………………

Treatment area tidied with nibblers

………………………………………………………………………………………

Prosthetic template hammered into place on femur

………………………………………………………………………………………

Lug drill advanced through appropriate holes on template.

………………………………………………………………………………………

Led is extended and checked.

………………………………………………………………………………………

Intramedullary hole is plugged with autograft tissue

………………………………………………………………………………………

2.2 Tibial Preparation

Knee placed in maximal flexion

………………………………………………………………………………………

Tibia is levered from underneath femur

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



Malleolar clamp of tibial alignment device placed

………………………………………………………………………………………

Upper cutting platform is positioned

………………………………………………………………………………………

Alignment is checked

………………………………………………………………………………………

Tibial stylus used to check and fix position

………………………………………………………………………………………

Cutting platform secured with pins or drill bits.

………………………………………………………………………………………

Oscillating saw used to make tibial cut

………………………………………………………………………………………

Cutting block and pins removed

………………………………………………………………………………………

Cut tidied with nibblers

………………………………………………………………………………………

Tibial tray selected and handle fitted.

………………………………………………………………………………………

Tibial tray positioned onto prepared surface

………………………………………………………………………………………

Prosthetic template is fixed to top of tray

………………………………………………………………………………………

Fitting is tested by extending leg.

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



Leg is placed in maximal flexion.

………………………………………………………………………………………

Tray secured with pins.

………………………………………………………………………………………

Handle is removed

………………………………………………………………………………………

Drill bushing size selected & fitted

………………………………………………………………………………………

Drill is used

………………………………………………………………………………………

Keel punch guide size is selected & fitted

………………………………………………………………………………………

Keel punch used

………………………………………………………………………………………

2.3 Patella resurfacing

Template selected

………………………………………………………………………………………

Leg placed in extension

………………………………………………………………………………………

Patella cutting guide used to grip patella

………………………………………………………………………………………

Resection with oscillating saw

………………………………………………………………………………………

Cut tidied with nibblers

………………………………………………………………………………………

Template used to drill holes

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



2.4 Insertion of Implants

All implant interfaces are cleaned

………………………………………………………………………………………

Surgeon and assistant change gloves

………………………………………………………………………………………

Prosthetics are obtained by runner

………………………………………………………………………………………

Surgeon confirms correct size of components

………………………………………………………………………………………

Prosthetic is unpacked and given to scrub nurse.

………………………………………………………………………………………

Cement is prepared.

………………………………………………………………………………………

Tibial implant assembled into inserter handle

………………………………………………………………………………………

Cement applied to tibial implant

………………………………………………………………………………………

Inserter handle used to place tibial component

………………………………………………………………………………………

Mallet blows are used to seat implant.

………………………………………………………………………………………

Excess cement is removed

………………………………………………………………………………………

Cement applied to femoral interface

………………………………………………………………………………………

Implant assembled into femoral inserter

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:

:



Femoral implant inserted

………………………………………………………………………………………

Mallet blows are used to seat implant.

………………………………………………………………………………………

Cement applied to patella

………………………………………………………………………………………

Implant fixed to patella

………………………………………………………………………………………

Excess cement is removed

………………………………………………………………………………………

Knee is flexed to check fitting

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

………………………………………………………………………………………

:

:

:

:

:

:

:

:

:



Treatment LEADERSHIP & MANAGEMENT Leadership Involves / Reflects on suggestions / Visible / Accessible / Inspires / Motivates / Coaches Maintenance of standards Subscribes to standards / Monitors compliance to standards / Intervenes if deviation
















Situation Awareness



3.0 Closure


Leg is fully extended

………………………………………………………………………………………

Tissues are cleaned with sterile saline solution

………………………………………………………………………………………

3.3 Count Equipment

2 members conduct the counts aloud, and in unison

………………………………………………………………………………………

Recommence if there is an interruption

………………………………………………………………………………………

Verbal confirmation of the success of the count.

………………………………………………………………………………………

3.4 Close Skin

Suction drains are inserted.

………………………………………………………………………………………

Deep tissues are repaired

………………………………………………………………………………………

The incision is closed with sutures and/or metal clips

………………………………………………………………………………………

Skin begins closure Is closed

………………………………………………………………………………………

Dressings applied

………………………………………………………………………………………

Tourniquet deflated

………………………………………………………………………………………

Runner announces tourniquet time.

………………………………………………………………………………………


………………………………………………………………………………………

………………………………………………………………………………………


……………………………………………………………………………………………….

…………………………………………………………………………………….

:

:

:

:

:

:

:

:

:

:

:

:



Assessment / Closure LEADERSHIP & MANAGEMENT Leadership Involves / Reflects on suggestions / Visible / Accessible / Inspires / Motivates / Coaches Maintenance of standards Subscribes to standards / Monitors compliance to standards / Intervenes if deviation
















Situation Awareness



APPENDIX 5: Checklist for Assessing Institutional Resilience INDICATORS OF RESILIENCE YES ? NO 1) Patient safety is recognised as being everyone’s responsibility, not just that of the risk management team.

2) Top management accepts occasional setbacks and nasty surprises as inevitable. It anticipates that staff will make errors and trains them to detect and recover them.

3) Top managers, both clinical and non-clinical, are genuinely committed to the furtherance of patient safety and provide adequate resources to serve this end.

4) Safety related issues are considered at high level meetings on a regular basis, not just after some bad event.

5) Past events are thoroughly reviewed at high level meetings and the lessons learnt are implemented as global reforms rather than local repairs.

6) After some mishap, the primary aim of top management is to identify the failed system defences and improve them, rather than seeking to pin blame on specific individuals.

7) Top management adapts a proactive stance towards patient safety. It does some or all of the following: Takes steps to identify recurrent error traps and remove them; strives to eliminate the workplace and organisation factors likely to provoke errors; brainstorms new scenarios of failure; conducts regular ‘health checks’ on the organisational processes known to contribute to mishaps.

8) Top management recognises that error-provoking institutional factors (e.g. under-manning, inadequate equipment, inexperience, patchy training, bad human-machine interfaces, etc) are easier to manage and correct than fleeting psychological states such as distraction, inattention and forgetfulness.

9) It is understood that effective management of patient safety, like any other management process, depends critically on the collection, analysis and dissemination of relevant information.

10) Management recognises the necessity of combining reactive outcome data (i.e. from the near miss and incident reporting system) with proactive process information. The latter entails far more than occasional audits. It involves regular sampling of a variety of institutional parameters (e.g. scheduling, roistering, protocols, defences, training).

11) Meetings relating to patient safety are attended by staff from a wide variety of departments and levels within the institution.

12) Assignment to a safety-related function (quality or risk management) is seen as a fast track appointment, not a dead end. Such functions are accorded appropriate status and salary.

13) It is appreciated that commercial goals, financial constraints and patient safety issues can come into conflict and that mechanisms exist to identify and resolve such conflicts in an effective and transparent manner.

14) Policies are in place that encourage everyone to raise patient safety issues.

15) The institution recognises the critical dependence of a safety management system on the trust of the workforce, particularly in regard to



reporting systems. (A safe culture, i.e. an informed culture, is the product of a reporting culture that, in turn, can only arise from a just culture). 16) There is a consistent policy for reporting and responding to incidents across all of the professional groups within the institution.

17) Disciplinary procedures are predicated on an agreed distinction between acceptable and unacceptable behaviour. It is recognised by all staff that a small proportion of unsafe acts are indeed reckless and warrant sanctions, but that the large majority of such acts should not lead to punishment. (The key determinant of blameworthiness is not so much the act itself - error or violation - as the nature of the behaviour in which it was embedded. Did this behaviour involve deliberate unwarranted risk-taking, or a course of action likely to produce avoidable errors? If so, then the act would be culpable regardless of whether it was an error or violation).

18) Clinical supervisors train junior staff to practise the mental as well as the technical skills necessary to achieve safe performance. Mental skills include anticipating possible errors and rehearsing the appropriate recoveries.

19) The institution has in place rapid, useful and intelligible feedback channels to communicate the lessons learnt from both the reactive and proactive safety information systems. Throughout the institution the emphasis is upon generalising these lessons to the system at large rather than merely localising failures and weaknesses.

YES = This is definitely the case in my institution (scores 1); ? = Don’t know, Maybe, or Could be partially true (scores 0.5); NO = This is definitely not the case in my institution (scores 0). Interpreting your score: 16-20 So healthy as to be barely credible 11-15 Moderate to high level of intrinsic resistance 6-10 Considerable improvements needed to achieve institutional resilience 1-5: Moderate to high institutional vulnerability 0: A complete rethink of organisational culture and processes is needed



APPENDIX 6: Evidence gathered for CAIR process

Review of Policies

Several policies directly relating to patient safety were reviewed. A brief description of each is given below.

Incident reporting policy The policy states clearly the relevant staff contacts and lists the external agencies from which specialist advice may be sought in the event of a major adverse incident. It also discusses the basic steps involved in making a report, appropriate disciplinary action and also explains the risk assessment process and how to contact the Risk Management team to request an assessment. The policy also details the procedures for undertaking an assessment of a high-risk activity. The policy also gives clear and detailed instructions as to how and when to make a report and what happens when a report has been submitted. However, aside from stating that “To ensure incidents are followed up in a timely manner” and “issuing quarterly a summary of incidents to the Trust Board, Divisional, and Clinical Unit Staff, Senior Nurses and any others the Patient Safety Manager deems appropriate” there is no explanation of how the outcome of the investigation into the incident and any associated changes of policy are disseminated throughout the Trust. Attitudes amongst staff towards adverse incident reporting were assessed via the Incident Reporting Attitude Survey.

Serious incident policy The serious incident policy is a continuation of the incident reporting policy described above and details the actions required to set up a telephone hotline system in the event of a major incident. It contains detailed and clear information concerning the procedures required to deal with a major event, but again does not detail how any changes in policy are disseminated after an event.

Disciplinary procedure The disciplinary procedure policy details the procedure for disciplinary action, namely: “a) an alleged incident occurs; or the member of staff fails to respond to counselling. b) Suspension if necessary. c) Investigation – to investigate the facts thoroughly and objectively. d) Disciplinary hearing. e) Decision”. The policy makes a clear distinction as to what happens when the staff member concerned is actually in breach of the law. The full disciplinary procedure is detailed extensively and clear distinctions are made between the various stages involved.

Policy for the control of MRSA The minimisation of hospital-acquired infections is paramount in maintaining good standards of patient safety. Low rates of in-patient MRSA infection may be seen as an indicator of good overall infection control procedures and policies. With this in mind, the policy of the control of MRSA was reviewed. The policy details extensively the testing procedures, treatment method and recommended isolation guidelines for MRSA.

Policy for order and purchasing equipment No policies could be found on the hospital intranet directly relating to the ordering and purchasing of equipment, but these issues were raised with appropriate members of staff during the interview process.

Interviews with Staff Members

All potential interviewees were contacted by letter and the interviewer (PG) made great efforts to arrange interviews for those expressing an interest at a time most convenient for them.



The following staff were interviewed:

• three consultant anaesthetists • three consultant cardiothoracic surgeons • two consultant cardiologists • a consultant intensivist • two specialist registrars (cardiology) • the secretary to a consultant cardiothoracic surgeon • the surgical research administrator • a senior operating department practitioner • the booking office manager • the clinical nurse specialist for cardiac transplantation • two theatre nurses • a senior perfusionist • the respiratory services manager • the cardiac respiratory and critical care unit divisional manager • the Trust chief executive.

A script of interview questions was followed for all of the interviews, and this is presented in appendix 2. Within this broad structure, there was scope within the interviews for any emerging issues to be discussed at length. Extensive notes were taken during the interviews and these were typed up to form a permanent record as soon after the interview as possible.

Semi-Structured Interview Script Thank you for taking the time to assist us. This interview is part of an information-gathering exercise for a project on patient safety being conducted by Professor Marc de Leval of Great Ormond Street Hospital for Children NHS Trust. Our role in this study is to apply Operational Research techniques to help understand how the different processes and departments work together, and how information flows between the different hospital departments and personnel and where bottlenecks and glitches occur. To provide a context for our discussions with yourself and other members of staff, we are focusing on the patient pathway for the arterial switch procedure. We are interested in gaining the perspective of members of staff involved in the delivery or organisation of care for this patient population. Process questions

1. What is your job title? What is your role in the organisation? 2. How do you fit into the management structure – who reports to you, and to whom do you

report? 3. What is your role in the organisation and delivery of care for patients undergoing the types of

surgical procedures that fall within your specialty? 4. With which members of staff do you communicate regarding surgical patients? What is their

job title, and in which departments do they work? 5. What information and materials do you receive regarding the surgical patients? Who does it

come from and what form does it take? 6. What information and materials regarding the surgical procedures do you pass to other

people? What form does it take? 7. What occurs upstream that can cause disruptions to your information flow?



8. What can happen downstream to cause problems with the information that you have passed on?

Questions on problems that have been encountered previously We are interested in understanding situations in which problems arise in the patient pathway that have a potential impact upon patient safety The information obtained through this interview is entirely confidential, and will be recorded with complete anonymity. Information acquired via this interview will be used only for research purposes and only on Professor de Leval’s safety project. The information obtained will only be reported in an aggregated form that will render it impossible for the source of the information to be traced or identified.

1. In your personal experience, either at this or other Trusts, what organisational problems have you encountered, at an earlier stage in the patient pathway, that could have an impact upon patient safety, either relating to the information or materials that you receive or otherwise? • How frequently does this problem occur? • How potentially severe is it? • Can it be resolved? If so, what steps would you recommend to minimise its occurrence?

2. In your personal experience, either at this or other Trusts, what organisational problems with

the later stages in the pathway are you aware of that could have an impact upon patient safety? • How frequently does this problem occur? • How potentially severe is it? • Can it be resolved? If so, what steps would you recommend to minimise its occurrence?

3. In your personal experience, either at this or other Trusts, what organisational problems have you encountered that could have an impact upon patient safety, relating to information generated or to other events that occur within your department? • How frequently does this problem occur? • How potentially severe is it? • Can it be resolved? If so, what steps would you recommend to minimise its occurrence?

Examples that may be used to prompt responses: • X-rays stored separately from patient notes, and getting separated from the notes when both are

needed in, for example, the Radiology Department; • A particular medical device that was not well designed being replaced with an alternative that does

the job much better; • A rota system that did not allocate enough time for handovers at the weekend. Questions relating to safety and incident reporting attitude • In terms of receiving reports of adverse incidents, at what forums, colleagues or meetings are these

reports disseminated? • What involvement do you have in taking action regarding these incidents? • How are changes to policy implemented and what is your role in such changes? • What channels are there for reporting adverse incidents? • Are you aware of the procedures and key personnel in your Trust directly responsible for patient

safety? • Have you ever submitted any kind of adverse incident report? • Did you receive feedback? What form did the feedback take?



The Trust Risk Management Team

Factors relating to the risk management team occur repeatedly in the items comprising the CAIR. Due to the importance of this team in maintaining good standards of patient safety, a meeting was held with them and a review of the team and their role in the Trust was carried out. The risk management team is involved in clinical risk assessment, handling of complaints, litigation and health and safety issues. There are two main team members who specialise in clinical risk issues, although all members of the team assist with all aspects of the team’s work. For clinical risk assessment, a specific member of the team acts as a contact for each Trust division,. The team carries out risk assessments, according to the primary risk concerns listed by each Trust division. For minor risk evaluations, each department performs its own risk assessments and contacts the risk management team when expert help or input is required. The three clinical units within the studied division receive a summary of their adverse incident reports, which then get added to the Trust's risk assessment register. The structure of the register and the process by which items are added to it is currently being revised by the team. One of the main functions of the risk management team is to produce root cause analyses for adverse incidents. Root cause analysis is a deductive, retrospective analysis of an incident, designed to understand the factors which contributed to the occurrence of the incident and ensure that steps are taken to minimise the risk of recurrence. To perform a root cause analysis, the team collect information from staff, policies, patient’s notes and other related information. They then construct a chronological sequence of events. The method of root cause analysis that the team uses differs from the standard suggested by the National Patient Safety Agency (NPSA), who recommend that the members of staff that were involved in the incident are excluded from the investigation. The team goes through the analysis with the members of staff involved in the incident because they feel this approach works very well for the division studied, due to the complexity of the patient cases. However, the team always involves some members of staff who were not involved in the incident to sit-in on the analysis process to ensure balance of opinion. The NPSA guidelines suggest that the root cause analysis is performed without interviewing members of staff directly involved in the incident but rather staff members who work in the relevant specialty but were not involved in the particular event of interest. The team attends the Friday morning clinical governance meeting if a member of the team has been involved in a recent risk assessment. This is one means by which the team liaises with the clinical staff in the cardiothoracic unit. Several of the clinical staff members interviewed said that they have found the team supportive and helpful when adverse incidents have occurred. The team also gets involved with medico-legal issues, such as withdrawal of consent and conflicts with families. The team feel that the reporting ethos in the Trust is very good and that staff are very honest and forthcoming with regard to reporting adverse incidents and alerting the team to potential safety problems. This, they believe, has led to a very good, blame-free working atmosphere. A protocol is in place for the production and dissemination of reports concerning adverse incidents. Firstly, the root cause analyses are formulated into a report. This report is then checked for factual accuracy with the members of staff involved in the incident. A summary of this report then goes to the chief executive, the Trust board and the families of the patients concerned. The root cause analyses are then presented at, for example, the Friday morning cardiothoracic unit clinical governance meeting. Next, there is a debrief with the members of staff involved in the incident, the form of which is dependent upon both the nature of the incident and also the staff involved. The team produces



quarterly risk management reports, which are available for all members of staff. Three of these reports were reviewed for the purposes of scoring the Checklist. The reports detail the frequency and type of adverse incidents that have occurred in the quarter, the details of any root cause analyses that have been performed and they also list other current issues related to patient safety.

Reviews of Meetings and Handovers in the Cardiac Surgery Centre

The meetings and handover procedures were reviewed and are discussed below.

The Multidisciplinary Conference Frequency: weekly. Attendees & Format: cardiac surgeons, cardiologists, interventional radiologists specialising in cardiac procedures, from SHO level upwards. A member of secretarial staff enters the decisions into the main database. The meeting is chaired by a senior clinician. At least one senior surgeon must be present for any final decision concerning a patient to be made. It is preferred that the physician in charge of each case is present for a decision to be made, although this is rarely enforced. The meeting is held in a seminar room with access to PC’s and multi-media projection equipment. Information is displayed on large screens. Occasionally remote conferencing is used via the internet; the latter aspect has been problematic in some sessions with poor audio quality causing difficulties (in one conference it had to be abandoned completely due to extremely poor quality). Scope: the meeting takes the form of a diagnostic, decision-making and planning forum. The format is for cases to be presented by cardiology registrars to the senior clinical staff. Decisions regarding diagnosis, further investigations, on-going treatment, surgical options, suitability for interventional techniques carry on until a general consensus regarding the next stage of treatment is reached. Occasionally, a particularly complex or challenging case is deferred for further discussion. Safety-related issues: the multidisciplinary conference may be considered a focal point of pre-operative patient information, and one of the most critical decision-making areas as far as minimising the operative risk to patients goes. Matters directly related to safety include: multiple opinions being carefully considered; discussions referring patients to lower-risk options (e.g. interventional cardiology options) where possible; careful assessment of the diagnostic information relating to the patient’s case and whether such information is comprehensive enough to make the best decision is consistently debated. However, the multidisciplinary conference has no minimum number of attendees: often there is only one consultant surgeon present, which may narrow the overall surgical opinion. There is no minimum discussion time per case, and the process is fairly unstructured. Generally, the registrar will present a diagnosis and history, together with a recent echocardiogram and cardiac angiography (if available). Treatment options are then discussed with the group acting as a forum. However, the members of the conference often leave and return later in the meeting, which implies that there are fluctuating numbers of staff present for different discussions. Since there is no formal structure to the presentation of each case, staff members may enter and be asked for their opinion without being fully briefed on the details of the patient. On one occasion, it was observed that a patient’s angiography results, showing the pressures in various blood vessels and heart chambers, actually had an erroneous value marked on them. This was quickly spotted by a member of the conference and agreed to be incorrect. However, it was not amended on the slide. Hence, when a surgeon entered and was asked for his opinion on the case, he made a recommendation based upon the (wildly) incorrect pressure reading he observed on the slide. Some minutes later, a colleague guessed that this had been the case and informed him of the correct pressure reading. The surgeon, who was to operate on this patient, changed his mind as a result of realising the magnitude of the error. Occurrences like this are rare, although the format of the meeting means that such errors are prone to happen. Another issue relating to the discussion of patients is that often the list of patients is very



long, and some discussions are either withheld to another meeting or occasionally rushed through. The list does follow an ad-hoc order, with the cardiology registrars presenting the cases in the order requested by their respective consultants. Although the focus of the meeting is to select the best treatment plan for each patient, a great deal of emphasis is placed on minimising the operative risk to the patient. A considerable amount of effort is utilised to try and identify patients that are likely to have a difficult path to a safe post-operative recovery and measures are taken to ensure that adequate provision is made as much as possible. It is difficult to assess how much of the information in the multidisciplinary conference reaches other staff directly involved in the care of the patient, for example nursing staff and anaesthetists. Presumably, these groups are informed on a more casual basis. The multidisciplinary conference is a productive forum for concerns regarding patient safety to be brought to the fore, although such issues are discussed elsewhere. A retrospective internal audit on the outcome of the multidisciplinary conference decisions, is currently in progress.

Intensive Care Post-Operative Handover Six post-op handovers, from theatre to ICU, were followed. The categories of staff participating varied and included the consultant anaesthetist supervising the theatre case, the anaesthetic registrar, the perfusionist involved in the case, the ICU consultant, the ICU registrar, the consultant surgeon, the surgical SpR, the bedside nurse, the ODP present at the case and the nurse in charge of ICU. Only the consultant anaesthetist was present at all of the handovers, and they did not always stay for the full course of the handover. The usual course of the handover was for the patient to be brought to ICU and positioned in the correct bedspace. Next, the consultant anaesthetist, anaesthetic SpR and ODA would transfer the patient and, with the aid of the bedside nurses, connect the ICU equipment to the patient. Once this process had started, the consultant anaesthetist would usually commence the verbal handover while the patient was being connected to the ICU equipment. Once the consultant anaesthetist had finished, the surgeon or the surgical SpR would give a brief handover of the surgical details. After this time, there would be frequently be some discussion between the intensivist and the surgeon, or the intensivist and the anaesthetist. During this discussion, the bedside nurse would start reading the surgical and anaesthetic notes and scan the theatre observations. Ventilator settings, for example, would be discussed with the ICU SpR. Generally, the handover process is a fragmented affair, with different parts of the handover targeting different members of staff. The bedside nursing staff in particular seem frequently to rely on “second hand” information. There also seems to be a loose order in which members of staff proceed in the handover, and often other members of staff talk amongst themselves or are engaged with equipment.

Clinical Governance Meeting Frequency: weekly. Attendees & Format: the clinical governance meeting is open to all members of staff. Attendees comprise surgeons, cardiologists and intensivists of all grades as the mainstay; frequently, nurses, anaesthetists and perfusionists will also attend. Depending upon the agenda of the meeting, there



may be invited speakers from other specialties from time to time. A record of attendance is also taken. The meeting has a fairly informal structure Scope: the meeting commences with an overview of the ICU activity for the week. Secondly, patients and events of interest are highlighted and discussed. Post-mortem information is presented next. Important topics and invited speakers are then presented. Finally, there is a brief review of the week’s activity in the cardiac catheter laboratory. When the meeting has been concluded, it is followed by a joint cardiac conference to review patients that have had surgery during the week. Safety-related issues: since the central content of the meeting varies from week to week, the safety-related content is often dependent upon the subject being discussed. However, due to the varied members of staff that attend, the forum provides a good way for safety-related issues to be disseminated, and the meeting is chaired in such a way as to encourage all staff to voice their opinions on sensitive issues. The post-mortem information is presented in an unbiased manner by pathologists, which provides an open and constructive way for all staff to benefit from the results. Staff are also welcome to present brief talks related to clinical issues.

ICU Morning Ward Round Two consultant cardiac surgeons, two consultant anaesthetists, three consultant cardiologists, two consultant intensivists, three SpR cardiologists, two SpR surgeons, two SpR intensivists and several members of nursing staff participated in the ward round observed. The morning handover for each patient was given by the ICU SpR, who had been present on the night shift. The general level of attention paid to the SpR as she handed over was good. The amount of information was quite comprehensive. There seemed to be a general feeling that the ward round was also a time to discuss patients amongst clinicians, although at the actual time of handover focus on the current patient was re-established. The consultant intensivists allowed the SpR to perform the handover, and only added extra information in response to a query from another consultant. The time spent on each patient seemed proportionate to the general condition of the child and seemed to be accepted by all the clinical staff as sufficient.

Cardiac Planning Meeting The cardiac planning meeting occurs twice per day every weekday, at 08:15 and 11:00. The meeting observed consisted of two consultant cardiac surgeons, a member of staff from the booking office, a consultant intensivist and two Sisters (one of whom was the bed coordinator). The meeting started on time and progressed rapidly. All those in attendance showed a very high level of awareness with regard to the status of patients, admissions and beds. Several of the attendees had comprehensive notes to hand which were used throughout the meeting. The patients and their needs were discussed carefully, and the meeting had a very open format whereby all the attendees were free to speak and state their opinions. All the attendees displayed a comprehensive knowledge of the cases discussed, and it appeared that the consequences of the meeting would have a direct real-time effect on the management of the day’s patients. The meeting is clearly a useful nexus for discussion regarding the forthcoming events of the day.



Cardiac Catheter Meeting Frequency: weekly. Attendees & Format: consultant cardiologist (head of cardiac catheter laboratory), consultant cardiologists, cardiology SpR’s, booking office administrator, catheter laboratory technicians, catheter laboratory nursing staff. Scope: the meeting encompasses all cardiology patients that have been referred for a cardiac catheter procedure via either the multidisciplinary conference or from clinic. Safety-related issues: the catheter meeting is the information nexus for patients undergoing any type of radiological intervention. The nature of such procedures involves a different type of risk to open surgery. In general, with routine procedures the overall risk to the patient is less than the corresponding open surgical procedure. However, sometimes interventional procedures are attempted on patients that are unfit for surgery, and in these situations the overall risk can be relatively high. The purpose of the catheter meeting, therefore, is to discuss each individual case, determine when the procedure will be done and plan for any potential contingencies. In case an interventional procedure goes awry, a full theatre and appropriate team is required to deal with the situation. The meeting focused on the status of the patients and which procedures would be appropriate, but only the booking office manager made comments regarding the provision of an operating theatre should an emergency situation arise.



APPENDIX 7: Commentary on CAIR results Item 1: “Patient safety is recognised as being everyone’s responsibility, not just that of the risk management team”. From the interviews: there are departmental sub-committees of the Trust Risk Management team which function to bring awareness of risk management issues to the main Trust team. All of the clinicians that were interviewed stated that they had met with the risk management team at some stage to deal with a clinical issue or to for the team to perform a root cause analysis. This reflects an underlying awareness of the importance of risk. Score = 1 Item 2: “Top management accepts occasional setbacks and nasty surprises as inevitable. It anticipates that staff will make errors and trains them to detect and recover them”. Protocols exist for divisional-level management to put contingency plans in place. Active training of general members of staff was not noted via the interviews. Regular top-level management meetings to assess risk and adverse incident reports occur. Specific training of staff to detect and recover from errors was not noted. Score = 0.5. Item 3: “Top managers, both clinical and non-clinical, are genuinely committed to the furtherance of patient safety and provide adequate resources to serve this end”. Top clinical managers – for example, initiation of the safety project; diversion of core funding to attempt to rectify the situations of blood transfusion and immediate post-op echocardiogram. Score = 0.5 Item 4: “Safety related issues are considered at high level meetings on a regular basis, not just after some bad event”. Regular top-level safety meetings (monthly basis) at divisional and Trust levels, extra meetings scheduled if a major adverse event occurs. Smaller meetings for clinical departments also held at regular (weekly) intervals. Cross-over between departmental and management meetings, although there appears to be little cross-over between clinical department meetings. Score = 1 Item 5: “Past events are thoroughly reviewed at high level meetings and the lessons learnt are implemented as global reforms rather than local repairs”. The risk management team is proactive in ensuring that patient safety issues are discussed at top-level management meetings. There may be a gap in the chain of dissemination from the divisional management level and the actual level of non-management divisional clinicians. For example, the divisional manager reported that dissemination of information from a potential major incident would take place via the Trust intranet after it had been discussed at top management level. Placing information on the Trust intranet could be viewed as a passive mode of dissemination The Risk Management Team is proactive in ensuring that patient safety issues are discussed at top-level management meetings. Score = 1 Item 6: “After some mishap, the primary aim of top management is to identify the failed system defences and improve them, rather than seeking to pin blame on specific individuals”. All levels of interviewees stated that there was a good culture in place for speaking out, and also for reporting in a blame-free manner. The interviews indicated that the aim of top management is definitely to improve defences rather than pin blame. The risk mMnagement tTam is definitely centred on a blame-free response, and this has been confirmed from all of the interviews with clinicians. Score = 1 Item 7: “Top management adapts a proactive stance towards patient safety. It does some or all of the following: Takes steps to identify recurrent error traps and remove them; strives to eliminate the workplace and organisation factors likely to provoke errors; brainstorms new scenarios of failure; conducts regular ‘health checks’ on the organisational processes known to contribute to mishaps”. Taking steps to identify recurrent error traps is currently implemented, from the CEO downwards through the management structure. It is not known whether hypothetical new scenarios of failure are considered – in all likelihood not, since the Trust risk management team is extremely busy producing root cause analyses for existing problems. The regular safety health checks are provided by the Risk



Management Team, who highlight the areas of greatest risk as feedback from individual Trust departments. However, the regular safety health checks seem to be initiated principally by the Risk Management Team, and not top-level management. The latter receive reports from the Team. Score = 0.5 Item 8: “Top management recognises that error-provoking institutional factors (e.g. under-manning, inadequate equipment, inexperience, patchy training, bad human-machine interfaces, etc) are easier to manage and correct than fleeting psychological states such as distraction, inattention and forgetfulness”. The chief executive certainly recognises the value of managing and maintaining a careful control over institutional factors which may be detrimental to patient safety. However, from the survey conducted, it was not clear what the overall attitude of the top-level management is. One of the senior managers interviewed showed a pro-active attitude towards management of institutional factors while another was focused more upon the concerns of individual members of staff. Score = 0.5 Item 9: “It is understood that effective management of patient safety, like any other management process, depends critically on the collection, analysis and dissemination of relevant information”. Regular auditing, for example surgery and CICU morbidity and mortality, is performed. The resulting information is disseminated via an interdisciplinary meeting, departmental meetings and open seminars. The bias is towards patient safety and learning from appropriate incidents so that protocols can be modified. The Risk Management Team conduct regular audits and are proactive in passing the information on to Trust management and individual clinical staff involved in the incident. Score = 1 Item 10: “Management recognises the necessity of combining reactive outcome data (i.e. from the near miss and incident reporting system) with proactive process information. The latter entails far more than occasional audits. It involves regular sampling of a variety of institutional parameters (e.g. scheduling, roistering, protocols, defences, training)”. While the potential to regularly sample proactive process data is in place, a senior clinician in the cardiothoracic unit stated that there are simply not enough correctly qualified and experienced staff available to perform such audits. Reviewing occurs at some points (for example, a new full assessment of the impending TOMCAT booking system. . Information is collected and stored, such as the THEATREMAN database, but there are no known regular samplings of such information. Many staff members revealed in the interviews that they have awareness of such issues, but there is no known formal reporting (with the possible exception of ICU). The Risk Management Team performs audits when a significant risk factor becomes identified at divisional level and data needs to be gathered to assess the risk. Additionally, one of the surgical registrars in the cardiothoracic unit is assigned the task of presenting regular seminars detailing surgical mortality data. Score = 1 Item 11: “Meetings relating to patient safety are attended by staff from a wide variety of departments and levels within the institution”. The multi-disciplinary mortality & morbidity meeting is held on a Friday morning and is open to all members of staff. It is chaired either by a cardiologist or an intensivist. The meeting is prepared and presented by a consultant intensivist from the cardiac intensive care unit. Interviewees commented that they found the meeting a good, open forum. They also said that they found that they could raise points in a blame-free manner. However, there are not always anaesthetists present at the meeting, since they have their own separate departmental M&M meeting. Further, some interviewees also commented that the meeting is not sufficient to disseminate lessons to a wider audience, and that the agenda is focused on intensive care incidents and issues. One consultant cardiologist stated that issues arising in outpatients for example would never get discussed at the Friday morning meeting. Dissemination of information via the Trust intranet has the inherent problem that it i snot possible to know if not knowing whether staff members have even seen, let alone read, the information. Score 0.5 Item 12: “Assignment to a safety-related function (quality or risk management) is seen as a fast track appointment, not a dead end. Such functions are accorded appropriate status and salary”. The Risk



Management Team is seen as a fast-track appointment, it has a high profile and an excellent working relationship with the clinical staff. The clinicians interviewed spoke positively of the risk management team in all cases, suggesting that they are held in quite high esteem. Also, the Trust risk management team liaises with the departmental sub-committees, and also sits on the senior management risk control meetings. It may be inferred, therefore, that the Trust Risk Management Team is held in high regard. Score = 1 Item 13: “It is appreciated that commercial goals, financial constraints and patient safety issues can come into conflict and that mechanisms exist to identify and resolve such conflicts in an effective and transparent manner”. Aside from the senior management meetings, there are no mechanisms that I have observed that would substantiate this. However, the attitude of the CEO and the head of the cardiothoracic department reflect very well their understanding of the potential conflict of interests between financial constraints and patient safety issues. They appear well-versed in assigning appropriate resources to ensuring that these types of conflicts are carefully resolved with as little compromise as possible. This may not be true of all senior management, however. Score 0.5 Item 14: “Policies are in place that encourage everyone to raise patient safety issues”. The incident reporting policy does indeed encourage all members of Trust staff to raise patient safety issues. Score = 1 Item 15: “The institution recognises the critical dependence of a safety management system on the trust of the workforce, particularly in regard to reporting systems. (A safe culture, i.e. an informed culture, is the product of a reporting culture that, in turn, can only arise from a just culture).” Several members of staff interviewed said that they had submitted incident reports, so there exist relatively accessible methods of doing so. The Risk Management Team felt that the reporting culture overall was very good. Score = 1 Item 16: “There is a consistent policy for reporting and responding to incidents across all of the professional groups within the institution”. There is indeed a consistent policy. Score = 1 Item 17: “Disciplinary procedures are predicated on an agreed distinction between acceptable and unacceptable behaviour. It is recognised by all staff that a small proportion of unsafe acts are indeed reckless and warrant sanctions, but that the large majority of such acts should not lead to punishment”. The policy does state clearly what behaviour is considered acceptable and unacceptable. It also lists the categories of behaviour that would constitute gross misconduct. Of the staff interviewed, there was a general agreement made that unsafe actions that appear reckless do warrant sanctions. Score = 1 Item 18: “Clinical supervisors train junior staff to practise the mental as well as the technical skills necessary to achieve safe performance. Mental skills include anticipating possible errors and rehearsing the appropriate recoveries”. This is definitely the case in the cardiothoracic department. All four of the consultants have an appreciation of technical and non-technical mental skill sets, which they pass on to their juniors. It was commented upon by multiple support staff that two of the consultants are very good at informing their juniors of the importance of items like discharge summaries. Score = 1 Item 19: “The institution has in place rapid, useful and intelligible feedback channels to communicate the lessons learnt from both the reactive and proactive safety information systems. Throughout the institution the emphasis is upon generalising these lessons to the system at large rather than merely localising failures and weaknesses”. Firstly, the main dissemination channels for adverse incident reporting in the cardiothoracic unit are via the Friday morning M&M meeting and, from the senior management level downwards, via the hospital intranet. Although many interviewees stated that the Friday morning meeting was a good, open forum, where matters could be discussed in a blame-free



environment, there was some concern that the forum was not wide enough to present all the issues that may occur in the Trust. Further, the fact that the anaesthetic department conducts its own dissemination meetings, and that there is not always an anaesthetist present at the Friday morning meeting, gave some cause for concern. Score = 0.5 Item 20: “The institution has the will and the resources to acknowledge its errors, to apologise for them, and to reassure patients (or their relatives) that the lessons learnt from such mishaps will prevent their recurrence.” Of all of the items present in the checklist, this is the most difficult to assess. The reason for this originates from the fact that the pre-requisite to accurately scoring this item is to have several histories for which major errors have occurred, and to analyse the subsequent response of the Trust to these events. Paradoxically, the greater the organisational safety health in the department concerned, the fewer of these events that are likely to have occurred. In terms of the overall attitude of the staff members, as is clear from the exacting discussions regarding adverse incidents in the JCC and the close integration with which the Risk Management Team operates with Trust staff, it is likely that in the event of major errors the Trust does have the will and resources to acknowledge its errors. However, in the absence of detailed recent cases, this cannot be directly confirmed. In light of this, the score = 0.5.



APPENDIX 8: The IRAS Questionnaire No.

Statement

Strongly Disagree

Disagree

Unsure

Agree

Strongly Agree

1. In the event of an incident, the Directorate seeks to find the individual responsible and discipline them.

1

2

3

4

5

2. The Directorate is proactive regarding incident prevention.

1 2 3 4 5

3. Incident investigations attempt to find out the real causes of an adverse event, rather than just blame those involved.

1

2

3

4

5

4. Directorate managers are not interested in near misses.

1 2 3 4 5

5. Reporting minor incidents (and near misses) is generally perceived as extra work and is avoided where possible.

1

2

3

4

5

6. Directorate managers can be trusted to follow up any incident that is reported, however minor.

1

2

3

4

5

7. Communication between staff and Directorate managers is open and easy.

1

2

3

4

5

8. In this department a forum exists to share experiences and learn from each others’ errors.

1

2

3

4

5

9. When I report an incident I am confident that action will be taken.

1 2 3 4 5

10. The risk management team is there to enforce rules and regulations and to punish staff that have put patients at risk.

1

2

3

4

5

11. The risk management team are interested in the chain of events that led up to the incident, rather than how any one individual acted.

1

2

3

4

5

12. Good clinicians/nurses will always report their mistakes.

1 2 3 4 5

13. Admitting to failure is like telling the medical world you are a bad clinician/nurse.

1 2 3 4 5

14. I feel supported when I admit a mistake. 1 2 3 4 5 15. I am likely to be reprimanded when I admit

to making a mistake. 1 2 3 4 5

16. Good clinicians/nurses will not allow themselves to make mistakes.

1 2 3 4 5



No. Statement Strongly

Disagree Disagree Unsure Agree Strongly

Agree

17. When I report an incident it is never a waste of my time.

1 2 3 4 5

18. Incident reporting is a valuable tool in the drive to continuously improve the safety of the patient.

1

2

3

4

5

19. Clinicians/nurses are only blamed and disciplined when they deserve it.

1 2 3 4 5

20. The attitude in this department is that to err is human, to report divine.

1 2 3 4 5

21. Communication between clinicians and managers is often difficult.

1 2 3 4 5

22. A ‘no blame’ culture is a utopia. 1 2 3 4 5 23. Any policy of naming, blaming and shaming

is unhelpful and out of date. 1

2

3

4

5

24. In this department we are encouraged to disclose our errors in order that everyone may profit from any lessons to be learnt.

1

2

3

4

5

25 In this department I might hesitate to make an incident report because:

a) I will get the blame. 1 2 3 4 5 b) Some things are too minor to warrant the

extra work. 1 2 3 4 5

c) I cannot be sure that any action will be taken anyway.

1 2 3 4 5

d) There is no reward for doing so. 1 2 3 4 5 e) I don’t want to blot my copybook. 1 2 3 4 5 f) The reporting method is too complicated

and/or time consuming. 1 2 3 4 5

g) I get no feedback. 1 2 3 4 5 h) If the consequences are not serious, no-one

need know. 1 2 3 4 5

i) I might get someone else into trouble. 1 2 3 4 5 If you have any further comments, please feel free to add them here.



Please complete the following section to help us with our analysis. JOB TITLE: ___________________________________________ Department: ___________________________________________ Length of time at this hospital: ____________________

Length of time in current department: ____________________

(if different from above, please list other departments in which you have worked) Date of qualification: ________________ Sex (M/F): ________________



APPENDIX 9: Rewording of the IRAS questionnaire for analysis purposes Reworded items are highlighted. Theme 1: attitude to the positive value of reporting incidents. No questions were re-worded related to this theme. Original question Re-

worded question

12. Good clinicians/nurses will always report their mistakes.

17. When I report an incident it is never a waste of my time.

18. Incident reporting is a valuable tool in the drive to continuously improve the safety of the patient.

Theme 2: perceptions that there are adverse personal consequences for reporting incidents Original question Re-worded question 1. In the event of an incident, the Directorate

seeks to find the individual responsible and discipline them.

In the event of an incident, the Directorate does not seek to find the individual responsible and discipline them.

3. Incident investigations attempt to find out the real causes of an adverse event, rather than just blame those involved.


The risk management team is not there to enforce rules and regulations and does not exist to punish staff that have put patients at risk.

13. Admitting to failure is like telling the medical world you are a bad clinician/nurse.

Admitting to failure is not like telling the medical world you are a bad clinician/nurse

14. I feel supported when I admit a mistake. 15. I am likely to be reprimanded when I admit to

making a mistake. I am unlikely to be reprimanded when I admit to making a mistake.

16. Good clinicians/nurses will not allow themselves to make mistakes.

Good clinicians/nurses sometimes make mistakes.

19. Clinicians/nurses are only blamed and disciplined when they deserve it.

22. A ‘no blame’ culture is a utopia. A ‘no blame’ culture can be achieved. 23. Any policy of naming, blaming and shaming

is unhelpful and out of date.



Theme 3: perceptions concerning value of reporting minor incidents and near misses. Original question Reworded question 4.

Directorate managers are not interested in near misses.

Directorate managers are interested in near misses.

5.

Reporting minor incidents (and near misses) is generally perceived as extra work and is avoided where possible.

Reporting minor incidents (and near misses) is not generally perceived as extra work and is done where possible.

Theme 4: attitudes to risk management team and incident investigation. Original question Reworded question 3. Incident investigations attempt to find out the

real causes of an adverse event, rather than just blame those involved.


The risk management team is not there purely to enforce rules and regulations and to punish staff that have put patients at risk.

11. The risk management team are interested in the chain of events that led up to the incident, rather than how any one individual acted.

Theme 5: perceptions that Directorate/management perform appropriately concerning

incidents. Original question Reworded question 2. The Directorate is proactive regarding incident prevention. 6. Directorate managers can be trusted to follow up any

incident that is reported, however minor.

7. Communication between staff and Directorate managers is open and easy.

8. In this department a forum exists to share experiences and learn from each others’ errors.

9. When I report an incident I am confident that action will be taken.

14. I feel supported when I admit a mistake. 20. The attitude in this department is that to err is human, to

report divine.

21. Communication between clinicians and managers is often difficult.

Communication between clinicians and managers is rarely difficult.

24. In this department we are encouraged to disclose our errors in order that everyone may profit from any lessons to be learnt.



APPENDIX 10: Full breakdown of responses to IRAS

Responses to IRAS questionnaire: Peadiatric Cardiac Unit

27 respondents unless otherwise indicated. Item Strongly


Agree 1 In the event of an incident,

the Directorate seeks to find the individual responsible and discipline them.

2 (7.4%)

12 (44.4%)

8 (29.6%)

2 (7.4%)

3 (11.1%)


1 (3.7%)

2 (7.4%)

5 (18.5%)

16 (59.2%)

3 (11.1%)

3 Incident investigations attempt to find out the real causes of an adverse event, rather than just blame those involved.

0 (0%)

0 (0%)

5 (18.5%)

15 (55.5%)

7 (25.9%)

4 Directorate managers are not interested in near misses (26 responses).

2 (7.7%)

12 (46.1%)

3 (11.5%)

6 (23.1%)

3 (11.5%)

5 Reporting minor incidents (and near misses) is generally perceived as extra work and is avoided where possible.

1 (3.7%)

6 (22.2%)

6 (22.2%)

12 (44.4%)

1 (3.7%)

6 Directorate managers can be trusted to follow up any incident that is reported, however minor.

1 (3.7%)

5 (18.5%)

10 (37.0%)

10 (37.0%)

1 (3.7%)

7 Communication between staff and Directorate managers is open and easy.

3 (11.1%)

3 (11.1%)

11 (40.7%)

8 (29.6%)

2 (7.4%)

8 In this department a forum exists to share experiences and learn from each others’ errors.

1 (3.7%)

6 (22.2%)

3 (11.1%)

12 (44.4%)

5 (18.5%)

9 When I report an incident I am confident that action will be taken (26 responses).

0 (0%)

5 (19.23%)

9 (34.6%)

10 (38.4%)

2 (7.7%)

10 The risk management team is there to enforce rules and regulations and to punish staff that have put patients at risk.

9 (33.3%)

9 (33.3%)

2 (7.4%)

6 (22.2%)

1 (3.7%)



Item Strongly Disagree

Disagree Unsure Agree Strongly Agree


1 (3.7%)

1 (3.7%)

7 (25.9%)

11 (40.7%)

7 (25.9%)


0 (0%)

0 (0%)

4 (14.8%)

17 (62.9%)

6 (22.2%)

13 Admitting to failure is like telling the medical world you are a bad clinician/nurse.

6 (22.2%)

15 (55.5%)

5 (18.5%)

1 (3.7%)

0 (0%)

14 I feel supported when I admit a mistake (26 responses).

0 (0%)

5 (19.23%)

7 (26.9%)

12 (46.1%)

2 (7.7%)

15 I am likely to be reprimanded when I admit to making a mistake.

1 (3.7%)

13 (48.1%)

6 (22.2%)

6 (22.2%)

1 (3.7%)

16 Good clinicians/nurses will not allow themselves to make mistakes.

3 (11.1%)

13 (48.1%)

4 (14.8%)

7 (25.9%)

0 (0%)


1 (3.7%)

7 (25.9%)

3 (11.1%)

13 (48.1%)

3 (11.1%)


0 (0%)

0 (0%)

2 (7.4%)

15 (55.5%)

10 (37%)

19 Clinicians/nurses are only blamed and disciplined when they deserve it.

0 (0%)

9 (33.3%)

11 (40.7%)

4 (14.8%)

3 (11.1%)

21 Communication between clinicians and managers is often difficult.

3 (11.1%)

3 (11.1%)

13 (48.1%)

5 (18.5%)

3 (11.1%)


1 (3.7%)

7 (25.9%)

6 (22.2%)

7 (25.9%)

6 (22.2%)

23 Any policy of naming, blaming and shaming is unhelpful and out of date.

0 (0%)

3 (11.1%)

3 (11.1%)

10 (37%)

11 (40.7%)


0 (0%)

5 (18.5%)

4 (14.8%)

12 (44.4%)

6 (22.2%)

Table A10.1: Numbers of responses given to each question for the f irst section of the IRAS survey



Response (27 total unless indicated otherwise)

Reason Strongly Disagree


I will get the blame (26 responses). 3 (11.5%)

11 (42.3%)

9 (34.6%)

3 (11.5%)

0 (0%)

Some things are too minor to warrant the extra work.

1 (3.7%)

8 (29.6%)

3 (11.1%)

14 (51.8%)

1 (3.7%)

I cannot be sure that any action will be taken anyway.

1 (3.7%)

9 (33.3%)

1 (3.7%)

12 (44.4%)

4 (14.8%)

There is no reward for doing so. 3 (11.1%)

6 (22.2%)

11 (40.7%)

7 (25.9%)

0 (0%)

I don’t want to blot my copybook (25 responses).

3 (12%)

12 (48%)

6 (24%)

4 (16%)

0 (0%)

The reporting method is too complicated and/or time consuming.

2 (7.4%)

8 (29.6%)

7 (25.9%)

6 (22.2%)

4 (14.8%)

I get no feedback. 1 (3.7%)

9 (33.3%)

7 (25.9%)

6 (22.2%)

4 (14.8%)

If the consequences are not serious, no-one need know.

3 (11.1%)

16 (59.2%)

7 (25.9%)

1 (3.7%)

0 (0%)

I might get someone else into trouble. 2 (7.4%)

7 (25.9%)

7 (25.9%)

11 (40.7%)

0 (0%)

Table A10.2: Number of responses and percentage response for each section of the second section of the IRAS survey. This section dealt with responses to the statement “In this department I might hesitate to make an incident report because...”



Responses to the IRAS questionnaire: Orthopeadic Surgery

18 respondents unless otherwise indicated. Number Item Strongly


Agree 1 In the event of an incident, the

Directorate seeks to find the individual responsible and discipline them.

0 (0%)

12 (67%)

2 (11%)

3 (16%)

1 (6%)


2 (11%)

4 (22%)

8 (44%)

4 (22%)

0 (0%)

3 Incident investigations attempt to find out the real causes of an adverse event, rather than just blame those involved.

1 (6%)

3 (16%)

0 (0%)

13 (72%)

1 (6%)

4 Directorate managers are not interested in near misses.

1 (6%)

8 (44%)

6 (33%)

3 (16%)

0 (0%)

5 Reporting minor incidents (and near misses) is generally perceived as extra work and is avoided where possible.

0 (0%)

5 (28%)

0 (0%)

11 (61%)

2 (11%)

6 Directorate managers can be trusted to follow up any incident that is reported, however minor.

4 (22%)

8 (44%)

2 (11%)

3 (16%)

1 (6%)

7 Communication between staff and Directorate managers is open and easy.

5 (28%)

9 (50%)

3 (16%)

1 (6%)

0 (0%)

8 In this department a forum exists to share experiences and learn from each others’ errors.

4 (22%)

7 (39%)

2 (11%)

5 (28%)

0 (0%)


6 (33%)

6 (33%)

4 (22%)

2 (11%)

0 (0%)

10 The risk management team is there to enforce rules and regulations and to punish staff that have put patients at risk.

3 (17%)

12 (67%)

2 (11%)

1 (6%)

0


0 3 (17%)

2 (11%)

12 (67%)

1 (6%)


1 (6%)

5 (28%)

2 (11%)

7 (39%)

3 (17%)

13 Admitting to failure is like telling the medical world you are a bad clinician/nurse.

6 (33%)

10 (56%)

1 (6%)

1 (6%)

0


1 (6%)

2 (11%)

9 (50%)

5 (28%)

1 (6%)

15 I am likely to be reprimanded when I admit to making a mistake.

1 (6%)

8 (44%)

4 (22%)

5 (28%)

0

16 Good clinicians/nurses will not allow themselves to make mistakes.

6 (33%)

11 (61%)

1 (6%)

0 0


4 (22%)

3 (17%)

4 (22%)

5 (28%)

2 (11%)



Number Item Strongly Disagree



0 0 3 (17%)

12 (67%)

3 (17%)

19 Clinicians/nurses are only blamed and disciplined when they deserve it.

2 (11%)

7 (39%)

6 (33%)

3 (17%)

0

20 The attitude in this department is that to err is human, to report divine.

2 (11%)

4 (22%)

10 (56%)

2 (11%)

0

21 Communication between clinicians and managers is often difficult.

0 0 4 (22%)

12 (67%)

2 (11%)

22 A ‘no blame’ culture is a utopia. 0 7 (39%)

5 (28%)

4 (22%)

2 (11%)

23 Any policy of naming, blaming and shaming is unhelpful and out of date.

0 3 (17%)

0 11 (61%)

4 (22%)


0 5 (28%)

6 (33%)

6 (33%)

1 (6%)

Table 10.3: Numbers of responses given to each question for the f irst section of the IRAS survey.

Response (18 total unless indicated otherwise) Reason Strongly


Agree I will get the blame. 1

(6%) 13

(72%) 2

(11%) 2

(11%) 0

Some things are too minor to warrant the extra work.

2 (11%)

4 (22%)

2 (11%)

10 (56%)

0

I cannot be sure that any action will be taken anyway.

1 (6%)

2 (11%)

2 (11%)

10 (56%)

3 (17%)

There is no reward for doing so. 4 (22%)

3 (17%)

2 (11%)

8 (44%)

1 (6%)

I don’t want to blot my copybook. 3 (17%)

8 (44%)

3 (17%)

3 (17%)

1 (6%)

The reporting method is too complicated and/or time consuming.

2 (11%)

5 (28%)

2 (11%)

7 (39%)

2 (11%)

I get no feedback. 1 (6%)

1 (6%)

4 (22%)

9 (50%)

3 (17%)

If the consequences are not serious, no-one need know.

1 (6%)

11 (61%)

5 (28%)

1 (6%)

0

I might get someone else into trouble. 2 (11%)

11 (61%)

4 (22%)

1 (6%)

0

Table A10.4: Number of responses and percentage response for each section of the second section of the IRAS survey. This section dealt with potential responses to the statement “In this department I might hesitate to make an incident report because...”



APPENDIX 11: The Operating Team Management Attitude Questionnaire Please answer by circling the appropriate number 1-5 against each of the following statements: No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

1. The senior person, if available, should take over and make all decisions in life-threatening emergencies.

1

2

3

4

5

2. It makes no difference to me with whom I work.

1 2 3 4 5

3. Senior staff should encourage questions from junior medical and nursing staff during operations if appropriate.

1

2

3

4

5

4. Even when fatigued, I perform effectively during critical phases of operations.

1 2 3 4 5

5. I work better when I operate according to set guidelines.

1 2 3 4 5

6. We should be aware of and sensitive to the personal problems of other operating team members.

1

2

3

4

5

7. Morale and productivity would be improved if surgeons, anaesthetists, anaesthetic assistants and scrub nurses consider themselves part of one team.

1

2

3

4

5

8. During periods of low work activity, I would rather relax than keep busy with small tasks.

1 2 3 4 5

9. Senior staff deserve extra benefits and privileges.

1 2 3 4 5

10. I let other team members know when my workload is becoming (or about to become) excessive.

1

2

3

4

5

11. Doctors who encourage suggestions from operating team members are weak leaders.

1 2 3 4 5

12. My decision making ability is as good in emergencies as in routine situations.

1 2 3 4 5

13. A regular debriefing of procedures and decisions after an operating session or shift is an important part of developing and maintaining effective crew co-ordination.

1

2

3

4

5

14. Team members in charge should verbalise plans for procedures or actions and should make sure that the information is understood and acknowledged by the others.

1

2

3

4

5



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

15. Junior operating team members usually should not question the decisions made by senior personnel.

1

2

3

4

5

16. Anaesthetists should express concerns about actions by surgeons when they have doubts.

1 2 3 4 5

17. Surgeons should express concerns about actions by anaesthetists when they have doubts.

1 2 3 4 5

18. Scrub nurses should express concerns about actions by anaesthetists when they have doubts.

1

2

3

4

5

19. It is better to agree with other operating theatre team members than to voice a different opinion.

1

2

3

4

5

20. The pre-session Joint Case Conference is important for safety and for effective team management.

1 2 3 4 5

21. I am more likely to make errors in tense or hostile situations.

1 2 3 4 5

22. The doctor’s responsibilities include co-ordination of his/her work team and other support areas.

1

2

3

4

5

23. Working for this hospital is like being part of a large family.

1 2 3 4 5

24. Operating team members share responsibility for prioritising activities in high workload situations.

1

2

3

4

5

25. Successful operating theatre management is primarily a function of the doctor’s medical and technical proficiency.

1

2

3

4

5

26. If I perceive a problem with the management of a patient I will speak out, regardless of who might be affected.

1

2

3

4

5

27. I am ashamed when I make a mistake in front of my other team members.

1 2 3 4 5

28. In critical situations, I rely on my superiors to tell me what to do.

1 2 3 4 5

29. Operating team members should not question the decisions or actions of senior staff except when they threaten the safety of the operation.

1

2

3

4

5

30. Anaesthetists should never question the actions of other anaesthetists in front of team members.

1 2 3 4 5

31. Surgeons should never question the 1 2 3 4 5



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

actions of other surgeons in front of team members.

32. Scrub nurses should never question the actions of surgeons.

1 2 3 4 5

33. I am less effective when stressed or fatigued.

1 2 3 4 5

34. Leadership of the operating theatre team should rest with the medical staff.

1 2 3 4 5

35. My performance is not adversely affected by working with an inexperienced or less capable team member.

1

2

3

4

5

36. To resolve conflicts, team members should openly discuss their differences with each other.

1

2

3

4

5

37. Operating theatre team members should monitor each other for signs of stress or fatigue.

1

2

3

4

5

38. I become irritated when I have to work with inexperienced medical staff.

1 2 3 4 5

39. I am proud to work for this hospital.

1 2 3 4 5

40. A truly professional operating theatre team member can leave personal problems behind when working in the operating theatre.

1

2

3

4

5

41. There are no circumstances where a junior team member should assume control of patient management.

1

2

3

4

5

42. Written procedures are needed for operating theatre situations.

1 2 3 4 5

43. Good communication and team co-ordination are as important as technical proficiency for operational efficiency and patient safety.

1

2

3

4

5

44. The concept of all operating theatre personnel working as a team does not work in our hospital.

1 2 3 4 5

45. Personal problems can adversely affect my performance.

1 2 3 4 5

46. Effective operating theatre team co-ordination requires members to take into account the personalities of other team members.

1

2

3

4

5

47. I like my job.

1 2 3 4 5

48. The work of each team member is 1 2 3 4 5



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

sufficiently appreciated by the other members of the team.

Note different scale for the next item:

No.

Statement

Very frequently

Frequently

Sometimes

Seldom

Very seldom


1

2

3

4

5

Note different scale for the next item: No.

Statement

Always

Usually

Sometimes

Seldom

Never


1

2

3

4

5

Please read the following descriptions of four different leadership styles, and answer the questions that follow by circling the appropriate style. Style 1. Usually makes his/her decisions promptly and communicates them to his/her

subordinates clearly and firmly. Expects them to carry out the decisions loyally and without raising difficulties.

Style 2. Usually makes his/her decisions promptly but, before going ahead, tries to

explain them fully to his/her subordinates. Gives them the reasons for the decisions and answers whatever questions they may have.

Style 3. Usually consults with his/her subordinates before s/he reaches his/her decisions.

Listens to their advice, considers it, and then announces his/her decision. S/he then expects all to work loyally to implement it, whether or not it is in accordance with the advice they gave.

Style 4. Usually calls a meeting of his/her subordinates when there is an important

decision to be made. Puts the problem before the group and invites discussion. Accepts the majority viewpoint as the decision.

51. Which one of the above styles of


Style 1

Style 2

Style 3

Style 4


Style 1

Style 2

Style 3

Style 4



WORK GOALS Please think of your ideal job – disregard your present job. In choosing an ideal job, how important would it be to you to:

No. Question Disagree

Strongly Disagree Slightly

Neutral

Agree slightly

Agree strongly


1

2

3

4

5

54. Have an opportunity for advancement to higher level jobs?

1 2 3 4 5

55. Have security of employment? 1 2 3 4 5


1 2 3 4 5


1 2 3 4 5


1

2

3

4

5


1

2

3

4

5


1 2 3 4 5


1 2 3 4 5


1 2 3 4 5

63. Have challenging tasks to do, from which you get a personal sense of accomplishment?

1 2 3 4 5


1 2 3 4 5


1 2 3 4 5


1 2 3 4 5

67. Have a job or career that will bring you prestige and recognition from others?

1 2 3 4 5

COMMENTS: 65. How can the effectiveness of operating theatre teams be increased?



66. How can the job satisfaction of operating theatre teams be increased? BACKGROUND INFORMATION Gender (M or F): _____ How much experience do you have in this specialty? (years) _____ Position (please circle or underline):

Consultant anaesthetist Anaesthetic SpR Anaesthetic technician Consultant surgeon

Surgical SpR Senior House Officer Scrub nurse Perfusionist Other (please state) …………………………………………………………….. Thank you for your time.



APPENDIX 12: Rewording of the OTTMAQ for analysis purposes An asterisk indicates a reworded item. Theme 1: Perspectives on individual safety. 2. It makes no difference to me with whom I work. 4. Even when fatigued, I perform effectively during critical phases of operations. *5. I do not work better when I operate according to set guidelines. 8. During periods of low work activity, I would rather relax than keep busy with small

tasks. 10. I let other team members know when my workload is becoming (or about to

become) excessive. 12. My decision making ability is as good in emergencies as in routine situations. *21. I am less likely to make errors in tense or hostile situations. 26. If I perceive a problem with the management of a patient I will speak out, regardless

of who might be affected. 27. I am ashamed when I make a mistake in front of my other team members. *28. In critical situations, I do not rely on my superiors to tell me what to do. *33. I am equally effective when stressed or fatigued. 35. My performance is not adversely affected by working with an inexperienced or less

capable team member. *38. I do not become irritated when I have to work with inexperienced medical staff. *45. Personal problems cannot adversely affect my performance. Theme 2: Leadership and hierarchy issues 1. The senior person, if available, should take over and make all decisions in life-

threatening emergencies. *3. Senior staff should not encourage questions from junior medical and nursing staff

during operations, even if appropriate to do so. 9. Senior staff deserve extra benefits and privileges. 11. Doctors who encourage suggestions from operating team members are weak

leaders. 13. A regular debriefing of procedures and decisions after an operating session or shift

is an important part of developing and maintaining effective crew co-ordination. 14. Team members in charge should verbalise plans for procedures or actions and

should make sure that the information is understood and acknowledged by the others.

34. Leadership of the operating theatre team should rest with the medical staff. 41. There are no circumstances where a junior team member should assume control of

patient management. 40. A truly professional operating theatre team member can leave personal problems

behind when working in the operating theatre. 43. Good communication and team co-ordination are as important as technical

proficiency for operational efficiency and patient safety.



Theme 3: Expressing concerns *15. Junior operating team members should not usually be afraid to question the

decisions made by senior personnel. 16. Anaesthetists should express concerns about actions by surgeons when they have

doubts. 17. Surgeons should express concerns about actions by anaesthetists when they have

doubts. 18. Scrub nurses should express concerns about actions by anaesthetists when they

have doubts. *19. It is not better to agree with other operating theatre team members than to voice a

different opinion. *29. Operating team members should not be afraid question the decisions or actions of

senior staff, even when they do not threaten the safety of the operation. *30. Anaesthetists should feel able to question the actions of other anaesthetists in front

of team members. *31. Surgeons should feel able to question the actions of other surgeons in front of team

members. *32. Scrub nurses should feel able to question the actions of surgeons. 36. To resolve conflicts, team members should openly discuss their differences with

each other. Theme 4: Attitude to the Trust 23. Working for this hospital is like being part of a large family. 39. I am proud to work for this hospital. 47. I like my job. Theme 5: Teamworking attitude 6. We should be aware of and sensitive to the personal problems of other operating

team members. 7. Morale and productivity would be improved if surgeons, anaesthetists, anaesthetic

assistants and scrub nurses consider themselves part of one team. 20. The pre-session Joint Case Conference is important for safety and for effective

team management. 22. The doctor’s responsibilities include co-ordination of his/her work team and other

support areas. 24. Operating team members share responsibility for prioritising activities in high

workload situations. 25. Successful operating theatre management is primarily a function of the doctor’s

medical and technical proficiency. 37. Operating theatre team members should monitor each other for signs of stress or

fatigue. 42. Written procedures are needed for operating theatre situations. *44. The concept of all operating theatre personnel working as a team does work in our

hospital. 46. Effective operating theatre team co-ordination requires members to take into

account the personalities of other team members. 48. The work of each team member is sufficiently appreciated by the other members of

the team.



APPENDIX 13: Responses to the OTTMAQ

Responses to section 1 of the OTTMAQ questionnaire: Peadiatric Cardiac Surgery

19 Total responses unless indicated otherwise No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


1 (5.3%)

4 (21.0%)

1 (5.3%)

8 (42.1%)

5 (26.3%)

2. It makes no difference to me with whom I work (18 total).

7 (38.9%)

4 (22.2%)

0 4 (22.2%)

3 (16.7%)


1 (5.3%)

0 1 (5.3%)

5 (26.3%)

12 (63.2%)


6 (31.6%)

6 (31.6%)

1 (5.3%)

4 (21%)

2 (10.5%)


4 (21.0%)

5 (26.3)

3 (15.8%)

4 (21.0%)

3 (15.8%)


0 1 (5.3%)

2 (10.5%)

7 (36.8%)

9 (47.4%)


0 0 0 2 (10.5%)

17 (89.5%)

8. During periods of low work activity, I would rather relax than keep busy with small tasks (18 total).

4 (22.2%)

6 (33.3%)

3 (16.7%)

5 (27.8%)

0


31 (15.83%)

4 (21%)

1 (5.3%)

6 (31.6%)

5 (26.3%)


0 3 (15.8%)

2 (10.5%)

8 (42.1%)

6 (31.6%)


16 (84.2%)

1 (5.3%)

0 1 (5.3%)

1 (5.3%)


1 (5.3%)

2 (10.5%)

3 (15.8%)

6 (31.6%)

7 (36.8%)


0 0 1 (5.3%)

9 (47.4%)

9 (47.4%)



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


0 0 1 (5.3%)

4 (21%)

14 (73.7%)


6 (31.6%)

11 (57.9%)

0 2 (10.5%)

0


0 0 1 (5.3%)

5 (26.3%)

13 (68.4%)


0 1 (5.3%)

0 5 (26.3%)

13 (68.4%)

18. Scrub nurses should express concerns about actions by anaesthetists when they have doubts (18 total). correct

1 (5.6%)

0 2 (11.1%)

6 (33.3%)

9 (50%)


10 (52.6%)

7 (36.8%)

1 (5.3%)

0 1 (5.3%)


0 0 2 (10.5%)

5 (26.3%)

12 (63.2%)


1 (5.3%)

1 (5.3%)

2 (10.5%)

8 (42.1%)

7 (36.8%)


0 2 (10.5%)

1 (5.3%)

6 (31.6%)

9 (47.4%)


4 (21%)

2 (10.5%)

2 (10.5%)

7 (36.8%)

4 (21%)

24. Operating team members share responsibility for prioritising activities in high workload situations (18 total).

1 (5.6%)

2 (11.1%)

3 (16.7%)

5 (27.8%)

7 (38.9%)

25. Successful operating theatre

management is primarily a function of the doctor’s medical and technical proficiency.

2 (10.5%)

6 (31.6%)

3 (15.8%)

5 (26.3%)

3 (15.8%)


1 (5.3%)

2 (10.5%)

4 (21.0%)

6 (31.6%)

6 (31.6%)


5 (26.3%)

5 (26.3%)

1 (5.3%)

5 (26.3%)

3 (15.8%)


5 (26.3%)

4 (21.0%)

3 (15.8%)

5 (26.3%)

2 (10.5%)



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


3 (15.8%)

8 (42.1%)

4 (21%)

2 (10.5%)

2 (10.5%)


4 (21.0%)

6 (31.6%)

4 (21.0%)

4 (21.0%)

1 (5.3%)

31. Surgeons should never question the actions of other surgeons in front of team members.

4 (21.0%)

8 (42.1%)

3 (15.8%)

3 (15.8%)

1 (5.3%)


5 (26.3%)

10 (52.6%)

3 (15.8%)

0 1 (5.3%)


1 (5.3%)

1 (5.3%)

2 (10.5%)

12 (63.2%)

3 (15.8%)

34. Leadership of the operating theatre team should rest with the medical staff (18 total).

2 (11.1%)

2 (11.1%)

3 (16.7%)

5 (27.8%)

4 (22.2%)


1 (5.3%)

10 (52.6%)

4 (21%)

0 4 (21.0%)


2 (10.5%)

0 5 (26.3%)

7 (36.8%)

5 (26.3%)

37. Operating theatre team members should monitor each other for signs of stress or fatigue (18 total).

3 (16.7%)

3 (16.7%)

1 (5.6%)

7 (38.9%)

4 (22.2%)

38. I become irritated when I have to work with inexperienced medical staff (18 total).

5 (27.8%)

2 (11.1%)

5 (27.8%)

6 (33.3%)

0

39. I am proud to work for this hospital.

1 (5.3%)

0 6 (31.6%)

4 (21%)

8 (42.1%)


1 (5.3%)

2 (10.5%)

2 (10.5%)

5 (26.3%)

9 (47.4%)

41. There are no circumstances where a junior team member should assume control of patient management.

7 (36.8%)

8 (42.1%)

3 (15.8%)

0 1 (5.3%)


2 (10.5%)

3 (15.8%)

5 (26.3%)

7 (36.8%)

2 (10.5%)


0 0 1 (5.3%)

5 (26.3%)

13 (68.4%)


7 (36.8%)

5 (26.3%)

3 (15.8%)

4 (21%)

0



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


5 (26.3%)

3 (15.8%)

1 (5.3%)

8 (42.1%)

2 (10.5%)


0 1 (5.3%)

1 (5.3%)

11 (57.9%)

6 (31.6%)

47. I like my job.

0 0 0 7 (36.8%)

12 (63.2%)

48. The work of each team member is sufficiently appreciated by the other members of the team.

2 (10.5%)

5 (26.3%)

3 (15.8%)

5 (26.3%)

4 (21%)

Table A13.1: Numbers of responses given to each question for the f irst section of the OTTMAQ survey.

Responses to section 1 of the OTTMAQ questionnaire: Orthopaedic Surgery

The percentages quoted are rounded to the nearest whole number and hence may not total 100% for every question item. 19 total responses unless indicated otherwise. No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


0 2 (10%)

5 (26%)

4 (21%)

8 (42%)

2. It makes no difference to me with whom I work.

11 (58%)

3 (16%)

1 (5%)

1 (5%)

3 (16%)


0 1 (5%)

0 2 (10%)

16 (84%)


5 (26%)

5 (26%)

0 2 (10%)

7 (37%)


2 (10%)

3 (16%)

5 (26%)

6 (32%)

3 (16%)


0 2 (10%)

5 (26%)

6 (32%)

6 (32%)


0 0 0 2 (10%)

17 (89%)

8. During periods of low work activity, I would rather relax than keep busy with small tasks.

3 (16%)

1 (5%)

4 (21%)

4 (21%)

7 (37%)



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly


2 (10%)

4 (21%)

4 (21%)

3 (16%)

6 (32%)


0 3 (16%)

2 (10%)

9 (47%)

5 (26%)


16 (84%)

1 (5%)

1 (5%)

0 1 (5%)


0 6 (32%)

2 (10%)

5 (26%)

6 (32%)


1 (5%)

0 2 (10%)

6 (32%)

10 (53%)


0 1 (5%)

1 (5%)

3 (16%)

14 (74%)


10 (53%)

4 (21%)

1 (5%)

4 (21%)

0


0 0 2 (10%)

3 (16%)

14 (74%)


0 0 2 (10%)

4 (21%)

13 (68%)

18. Scrub nurses should express concerns about actions by anaesthetists when they have doubts.

2 (10%)

0 3 (16%)

5 (26%)

9 (47%)


1 (5%)

2 (10%)

2 (10%)

5 (26%)

9 (47%)


1 (5%)

1 (5%)

6 (32%)

5 (26%)

6 (32%)


8 (42%)

6 (32%)

1 (5%)

3 (16%)

1 (5%)


1 (5%)

2 (10%)

0 5 (26%)

11 (58%)


5 (26%)

3 (16%)

2 (10%)

5 (26%)

4 (21%)

24. Operating team members share responsibility for prioritising activities in high workload situations.

1 (5%)

1 (5%)

3 (16%)

5 (26%)

9 (47%)



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

25. Successful operating theatre management is primarily a function of the doctor’s medical and technical proficiency.

8 (42%)

1 (5%)

4 (21%)

4 (21%)

2 (10%)


0 5 (26%)

2 (10%)

7 (37%)

5 (26%)


2 (10%)

4 (21%)

3 (16%)

9 (47%)

1 (5%)


0

3 (16%)

3 (16%)

7 (37%)

6 (32%)


5 (26%)

3 (16%)

4 (21%)

6 (32%)

1 (5%)


3 (16%)

10 (53%)

1 (5%)

5 (26%)

0

31. Surgeons should never question the actions of other surgeons in front of team members.

3 (16%)

9 (47%)

2 (10%)

5 (26%)

0


0 2 (10%)

2 (10%)

7 (37%)

8 (42%)


0 4 (21%)

0 11 (58%)

4 (21%)

34. Leadership of the operating theatre team should rest with the medical staff.

2 (10%)

7 (37%)

3 (16%)

2 (10%)

5 (26%)


5 (26%)

4 (21%)

3 (16%)

5 (26%)

2 (10%)


0 0 2 (10%)

0 17 (89%)

37. Operating theatre team members should monitor each other for signs of stress or fatigue.

1 (5%)

2 (10%)

1 (5%)

8 (42%)

7 (37%)

38. I become irritated when I have to work with inexperienced medical staff.

0 3 (16%)

6 (32%)

6 (32%)

4 (21%)

39. I am proud to work for this hospital. 1 (5%)

2 (10%)

5 (26%)

5 (26%)

6 (32%)


2 (10%)

2 (10%)

1 (5%)

8 (42%)

6 (32%)

41. There are no circumstances where a 3 7 1 6 2



No.

Statement

Disagree Strongly

Disagree Slightly

Neutral

Agree slightly

Agree strongly

junior team member should assume control of patient management.

(16%) (37%) (5%) (32%) (10%)


0 2 (10%)

5 (26%)

4 (21%)

8 (42%)


3 (16%)

0 1 (5%)

0 15 (79%)


2 (10%)

8 (42%)

5 (26%)

3 (16%)

1 (5%)


2 (10%)

5 (26%)

3 (16%)

6 (32%)

3 (16%)


1 (5%)

4 (21%)

2 (10%)

7 (37%)

5 (26%)

47. I like my job. 0 1 (5%)

2 (10%)

6 (32%)

10 (53)%

48. The work of each team member is sufficiently appreciated by the other members of the team.

1 (5%)

7 (37%)

4 (21%)

7 (37%)

0

Table A13.2: Numbers of responses given to each question for the f irst section of the OTTMAQ survey



APPENDIX 14: Processes at each stage of the patient journey Pre-admission procedures Pre-admission procedures include introducing the parents and the child to the ward, meeting the cardiac liaison sister, having standard blood tests (usually performed by the cardiology SHO) – these include full blood count, urea and electrolytes, blood for cross-matching and any other test that may be appropriate. The child will have an echocardiogram so that their current heart anatomy and function may be better understood, and they may also have a chest x-ray to determine the anatomy of the lungs, to check for the presence of any lung infection and also to visualise the heart in relation to other thoracic structures. The parents and child may meet the play therapist. A history of the child’s presenting condition will be taken by the cardiologist, and the patient may also be seen by members of other specialist teams and anaesthetists if there are co-mordibities or other pre-existing conditions. The pre-operative ward Patients are admitted to the ward shortly before the day they are due to have surgery. Normally, this will be the day before. However, if there are other co-existing medical conditions, or if the patient has to be admitted to hospital in order to be prepared for surgery, they will be admitted further in advance of their scheduled operation to allow such preparation. During their stay on the ward further blood tests may be taken to assess their overall medical condition and their fitness for surgery and they will be reviewed by the cardiology, surgery and anaesthetic teams, as well as teams from other medical and surgical specialties as required. They may also undergo either a diagnostic or interventional catheterisation if this is appropriate. Further pre-surgical tests There are several other possible tests that may be carried out while the patient is upon the ward. As well as the chest x-ray and blood tests mentioned above, the patient will also have an electrocardiogram (ECG) taken to investigate the electrical activity of their heart. They may also have a trans-oesophageal echocardiogram taken, which gives a picture of much greater clarity then a standard external echo. In certain cases, the patient may have a stress echo performed. This is an echocardiogram taken while the patient is given a carefully administered set of drugs that affect cardiac function. Patients may have their cardiac function assessed via an exercise tolerance test, whereby they exercise under controlled conditions on a treadmill while the electrical activity of their heart is recorded. In some instances valuable information for the pending surgery may be obtained via Magnetic Resonance Imaging (MRI). However, in neonates this has the disadvantage that the patient undergoes a general anaesthetic for the duration of the test. As a further diagnostic aid, a Computed Tomography (CT) scan may also be undertaken. In addition to any of the above tests, the ward nursing staff will record baseline observations of blood pressure, pulse and temperature at regular intervals during the patient's stay on the ward. The patient will be prepared for surgery the night before by being fasted and undergoing a series of pre-operative checks. On the day of surgery, they will be taken to the induction room by either their parents or a porter and accompanied by a staff nurse. Cardiac catheteristaion If it is decided that the patient will undergo a catheterisation, either diagnostic or interventional, the patient is prepared for the procedure on the ward by being starved the evening before. Prior to the patient being transferred to the catheter laboratory, ward nursing staff complete a pre-procedure checklist to document that adequate patient identification is available and that fasting, medication and consent procedures have been observed. The patient is accompanied to the catheter laboratory by a staff nurse, and is then anaesthetised and transferred to the lab itself. Interventional procedures are



more complex than diagnostic procedures and carry a greater procedural risk. This means that for many interventional procedures, an ICU or HDU bed needs to be available for the patient. Otherwise the patient is extubated and remains in a recovery area outside the catheter laboratory for a short period of time until the anaesthetist is satisfied that they may return to the ward. Transfer to the operating theatre When the patient is transferred from the ward to the operating theatre, they are initially taken into the induction room. The anaesthetist will administer appropriate anaesthetic and, assisted by an operating department practitioner, they will intubate the patient and then insert arterial and central venous lines, for invasive blood pressure monitoring and delivery of fluids and drugs respectively. Once the patient has been stabilised on the ventilator, they will be moved into the operating theatre and the theatre staff will finalise the patient ready for the operation to commence. Transfer to intensive care Once the surgical procedure has been completed the patient is transferred to a portable ventilator and monitoring system, attached to an ICU bed, by the anaesthetics team. The patient is stabilised and taken immediately to ICU. A detailed handover between the theatre team and the ICU team is then conducted, while the patient is connected to the ICU equipment and all of the instrumentation is re-calibrated. A plan is agreed for weaning the patient from ventilation and supporting his or her other physiological systems. Transfer to a post-operative ward Once the patient’s medical and surgical teams are satisfied that his or her physiological systems are functioning sufficiently well, the patient will be transferred to a ward (via a high dependency unit in cases where extended monitoring is required). Assuming that the patient’s condition does not deteriorate and that they do not require further surgery, the patient is reviewed at regular intervals until they are considered fit for discharge. During this time, any other actions required to ensure adequate post-discharge support (such as parental education, appointments for follow-up clinics, letters to GP’s etc.) are taken, ready for the day of discharge.



APPENDIX 15: Information flow at each stage of the patient journey Information exchange prior to admission A member of the booking office is present at the Joint Case Conference where all patient referrals are discussed, and a plan for treatment of the patient is made (see section 2.C.C. for details). The booking office then sets a date for pre-admission and agrees a date for surgery with the secretary of the consultant surgeon. A list of patients that are due to be admitted on a specific day are entered into one of the booking databases and are made available to the sister or charge nurse designated as a bed manager for that day. The nurse in charge of the ward that the patient will be admitted to is also informed of any admissions for that day via the booking system. Ward admission Once the patient has been admitted to the ward, the patient’s bedside nurse will be the primary source of up-to-date information. This will be circulated through the other nursing staff on the ward, informally and via formal handovers, and also to the nurse in charge of the ward, as well as doctors and other members of medical staff (for example, specialist nurses) that may be involved in the patient’s care. Exchange between the patient’s surgical team, medical team and the nursing staff occurs during ward rounds. The patient may also be taken from the ward to a different location in the hospital for some types of test, for example MRI, and this would be arranged in consultation with the nursing staff on the patient’s ward. If there are clinical features or complications that arise during the patient’s stay on the ward, these may be discussed at the JCC, or within individual team briefings. Any change in the date for surgery will be agreed between the consultant’s secretary and the booking office. At the time of the study there seemed to be no clear and consistent strategy for disseminating such information to other members of staff. The day before the patient is due to have their surgery, they will be prepared for their procedure by the nursing staff on the ward. The patient will also be reviewed by an anaesthetist and any potential anaesthetic difficulties will be noted. Once the patient's position on the following day’s operating list has been confirmed, the patient is fasted overnight. Once theatres are prepared and ready for the patient, a theatre nurse will telephone the ward to tell them to bring the patient down, and a porter will also be contacted if needed. The surgical procedure Once the patient arrives in theatre, they are taken to the induction room. The theatre manager will have access to the operating schedule and will be made aware of any cancellations or delays. During the patient’s operation, the anaesthetist will liaise with ICU to keep them up-to-date on the status of the patient, the stage of the operation and any potential delays. The staff nurse in ICU, and the intensive care doctors, will prepare for the arrival of the patient. Transfer to ICU Once the patient has been transferred to ICU, the information associated with them will be circulated around the ICU staff, during the process of hand-overs and ward rounds, as well as reviews by the intensive care unit doctors. Other support staff, both associated with ICU (such as perfusionists) or staff from departments outside ICU (for example, nuclear medicine) will be involved with the patient’s ongoing treatment if appropriate. At this time, other support staff, such as the cardiac liaison team, may also be involved, and will follow the patient’s care as they are transferred from ICU to either a standard ward or a high-dependency unit.



Transfer to post-ICU ward Once the patient has recovered sufficiently, they will be transferred to either a standard ward or an HDU. The information associated with the patient will be handed over to the nursing team by a member of ICU staff. Should the patient be appropriate for imminent discharge, the discharge coordinator will be made aware by the nursing team on the ward, and also the consultant’s secretary will be informed by the booking office so that a follow-up review can be arranged.



APPENDIX 16: Details of sub-simulations Bed management on ward In order to allow for one diagnostic procedure, the sub-simulation consists of multiple pairs of work centres, organised as groups. A group is a type of simulation entity, and has particular properties associated with it. Each bed is represented by a single group, which in turn contains work centres representing the patient before they have left for their diagnostic procedure and the patient after they have returned from their procedure. Multiple diagnostic procedures, as implemented in the main simulation, require groups to be composed of additional work centres, one for each diagnostic test plus the pre-test and post-test work centres. Each group then forms a chain of work centres, which the patient advances along after each procedure. Note that this is where the concept of a group is very useful, since the patient remains in the same group, i.e. the same bed, even though they may leave the ward several times to have different diagnostic procedures, and thus will progress through several different work centres in their particular group. Assigning the work centres to a group, which represents a single bed, is the key to how the sub-simulation works. Groups can have properties which affect all of the work centres in the group. In the case of the sub-simulation, we restrict each group to only having one patient associated with it at any one time. In this way, the bed is automatically ‘saved’, as another patient cannot enter the group until the patient associated with that group has moved to another part of the simulation. As each patient is admitted into the simulation, a ‘surgical preparation’ label is generated relating to the number of pre-operative diagnostic procedures that the patient requires, and is described in more detail below. There is visual logic code in the ‘treatment’ work centre of each group which blocks the patient moving to the ‘finished’ work centre until their surgical preparation has reached 5. At this point, the patient leaves the ward, and proceeds to the induction room. The group is then ready to admit a new patient. When the patient leaves the ward to go to the appropriate diagnostic centre, a new work item is created, which ‘becomes’ the patient as they undergo the CT scan, for example. The original patient remains as a dummy work item in the ‘treatment’ work centre of the correct bed. This blocks the bed from having another patient admitted into it. Note that all of the patient labels are updated as each work item returns from each of the diagnostic work centres. Once the appropriate diagnostic process has been completed, the visual logic code in the relevant diagnostic work centre unblocks the next work centre in the group, and the patient moves long the chain of work centres in the group itself. Once the patient has reached the final bed state, namely the final work centre in the group, they are ready to leave the ward and proceed to the induction room for their operation. Diagnostic procedures There are many different types of operative procedure for congenital cardiac surgery. A selection of these are modelled in the simulation. Each patient arriving via the elective pathway is assigned labels named ‘surgical preparation’ and ‘true cardiac diagnosis’. The surgical preparation label represents the number of diagnostic procedures required for an appropriate diagnosis to be reached. The sub-simulation works by using SIMUL8’s ‘route on label value’ dynamic. Each of the diagnostic work centres is associated with a number from 1 to 4. Incoming work items are routed to the work centre which matches their ‘surgical preparation’ label value. Once the patient has left the work centre, their ‘surgical preparation’ label is incremented by 1. In this way, they are gradually routed through each necessary diagnostic procedure in turn, assuming that the equipment does not fail or another error occurs. Once the diagnostics are complete, the patient’s surgical preparation value reaches the required value to open a ‘gate’ in the admission ward work centre which allows the patient to join the queue for the induction room.



APPENDIX 17: Calibration of model parameters In this section, we discuss the various simulation parameters and describe how they may be estimated or measured. We group these according to the following categories: available from existing sources, observable, available via expert opinion or inestimable. Label Source of estimate Number of patients typically admitted in a week for cardiac surgery

Data available

Number and type of cardiac operations performed in a week

Data available

Type of diagnostic procedures Data available Frequency of errors occurring Data available, expert opinion Types and natures of errors occurring Data available, expert opinion Number of beds per ward area Data available, observation Number of units of transfused blood requested out-of-hours

Data available, expert opinion

Length of cardiac surgical procedures Data available, expert opinion Numbers of staff working in various parts of the patient pathway

Observation, expert opinion

Frequency of equipment failures Expert opinion Number of emergency admissions Data available Number of weekend and evening operations performed

Data available

Modeling of the patient pathway Observation, expert opinion Frequency of missing notes and x-rays Expert opinion Staff allocation and shift patterns Observation, expert opinion



APPENDIX 18: Visual Logic code for the simulation model “Visual logic” is the name given to SIMUL8’s internal programming language. The language relates to SIMUL8’s event-driven structure, namely that each time an event occurs in the simulation, the SIMUL8 program pauses to decide which action to take next. Using visual logic, the programmer can take strategic advantage of these pauses to control the processing of events. Visual logic is written into numerous sections, each of which is triggered in response to certain events in the simulation model. Example events include a new work item arriving at a work entry point, a work centre locating a work item and pulling it into the work centre, the user clicking the reset button and the model completing its simulation run. In this appendix, we present the visual logic code used in the simulation. The code is presented in the following form: the work centre or event where the code occurs is in bold type, with the criteria for executing the code in italics. The body of the code itself is shown in regular type. VL SECTION: Bed 1 treat Work Complete Logic1 IF Surgical preparation < 5 SET Route 1 = 0 Add Work To Queue Main Work Item Type , Dummy storage unit SET Bed label 1 = 1 VL SECTION: Bed 2 treat Work Complete Logic1 IF Surgical preparation < 5 SET Route 2 = 0 Add Work To Queue Main Work Item Type , Dummy storage unit SET Bed label 2 = 1 VL SECTION: Bed 3 treat Work Complete Logic1 IF Surgical preparation < 5 SET Route 3 = 0 Add Work To Queue Main Work Item Type , Dummy storage unit SET Bed label 3 = 1 VL SECTION: CT Scanner Route In After Logic1 SET Label_change_DIAG[Hospital number,5] = 1 VL SECTION: Dummy for label copying Route In After Logic1 SET Initial surg prep = Surgical preparation VL SECTION: Dummy WC for label processing 1 Route In After Logic1 'This is for the DIAGNOSTIC pathway SET var_batch_id = var_batch_id+1 SET labl_batch_id = var_batch_id SET Diag_path_labels[Hospital number,labl_batch_id] = Test label A initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+1] = Test label B initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+2] = Test label C initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+3] = Test label D initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+4] = Test label E initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+5] = Test label F initially set to 0 SET Diag_path_labels[Hospital number,labl_batch_id+6] = Surgical preparation SET Diag_path_labels[Hospital number,labl_batch_id+7] = Cross match status SET var_batch_id = 0



VL SECTION: Dummy WC for label processing 2 Route In After Logic1 'This is for the TRANSFUSION pathway SET var_batch_id = var_batch_id+1 SET labl_batch_id = var_batch_id SET Trans_path_labels[Hospital number,labl_batch_id] = Test label A initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+1] = Test label B initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+2] = Test label C initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+3] = Test label D initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+4] = Test label E initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+5] = Test label F initially set to 0 SET Trans_path_labels[Hospital number,labl_batch_id+6] = Surgical preparation SET Trans_path_labels[Hospital number,labl_batch_id+7] = Cross match status SET var_batch_id = 0 VL SECTION: Echo department Route In After Logic1 SET Label_change_DIAG[Hospital number,1] = 1 SET Label_change_DIAG[Hospital number,2] = 1 VL SECTION: Induction (out of hours) Route In Before Logic1 IF Queue for CICU.Count Contents = 8 Block Current Routing VL SECTION: Induction room Route In After Logic1 'Check the flag in "Label_ch", read from the diagnostic pathway and write the corresponding values to the labels. LOOP 1 >>> loop_position >>> 6 IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label A initially set to 0 = Diag_path_labels[Hospital number,1] IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label B initially set to 0 = Diag_path_labels[Hospital number,2] IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label C initially set to 0 = Diag_path_labels[Hospital number,3] IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label D initially set to 0 = Diag_path_labels[Hospital number,4] IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label E initially set to 0 = Diag_path_labels[Hospital number,5] IF Label_change_DIAG[Hospital number,loop_position] = 1 SET Test label F initially set to 0 = Diag_path_labels[Hospital number,6] 'Now check the flags from the transfusion pathway and write the appropriate label values from the n+1 column LOOP 1 >>> loop_position >>> 6 IF Label_change_TRANS[Hospital number,loop_position] = 1 SET Test label A initially set to 0 = Trans_path_labels[Hospital number,1] IF Label_change_TRANS[Hospital number,loop_position] = 1 SET Test label B initially set to 0 = Trans_path_labels[Hospital number,2] IF Label_change_TRANS[Hospital number,loop_position] = 1 SET Test label C initially set to 0 = Trans_path_labels[Hospital number,3] IF Label_change_TRANS[Hospital number,loop_position] = 1 SET Test label D initially set to 0 = Trans_path_labels[Hospital number,4] IF Label_change_TRANS[Hospital number,loop_position] = 1 SET Test label E initially set to 0 = Trans_path_labels[Hospital number,5] IF Label_change_TRANS[Hospital number,loop_position] = 1



SET Test label F initially set to 0 = Trans_path_labels[Hospital number,6] VL SECTION: Induction room Route In Before Logic1 IF Queue for CICU.Count Contents = 8 Block Current Routing VL SECTION: MRI Scanner Route In After Logic1 SET Label_change_DIAG[Hospital number,3] = 1 SET Label_change_DIAG[Hospital number,4] = 1 VL SECTION: Phlebotomy Route In After Logic1 SET Label_change_DIAG[Hospital number,6] = 1 VL SECTION: Reset Logic1 Clear Sheet Data Only Diag_path_labels[1,1] Clear Sheet Data Only Trans_path_labels[1,1] Clear Sheet Data Only Label_change_DIAG[1,1] Clear Sheet Data Only Label_change_TRANS[1,1] VL SECTION: Transfusion department Route In After Logic1 SET Label_change_TRANS[Hospital number,3] = 1 VL SECTION: Phlebotomy Route In After Logic1 ‘VL code to increase the probability of an error occurring ‘when the queue for the phlebotomoy work centre ‘gets to certain sizes SET Label_change_DIAG[Hospital number,6] = 1 'Read the status of the queue and acquire a value from the appropriate distribution 'Comment IF Phlebotomy queue.Count Contents > 20 Display Text "Phleb queue has more than 20 patients" , "" , 0 , 60 SET RND_phleb_error = 55 patients error distribution IF RND_phleb_error = 1 Display Text "55 occurred" , "" , 600 , 0 'Note that in this case, it is easier to just flag an error as being = 1 and resolution of the error = 0 (for now). SET EC_phleb_overworked = 1 'Comment IF Phlebotomy.Count Contents > 0 Select Current Work Item Phlebotomy , 1 Move Work Item To Error path queue , -1 IF Phlebotomy queue.Count Contents > 15 Display Text "Phleb queue has more than 15 patients" , "" , 0 , 40 SET RND_phleb_error = 50 patients error distribution IF RND_phleb_error = 1 Display Text "50 occurred" , "" , 600 , 20 SET EC_phleb_overworked = 1 'Comment IF Phlebotomy.Count Contents > 0 Select Current Work Item Phlebotomy , 1 Move Work Item To Error path queue , -1 IF Phlebotomy queue.Count Contents > 10 Display Text "Phleb queue has more than 10 patients" , "" , 0 , 20



SET RND_phleb_error = 45 patients error distribution IF RND_phleb_error = 1 Display Text "45 occurred" , "" , 600 , 40 SET EC_phleb_overworked = 1 'Comment IF Phlebotomy.Count Contents > 0 Select Current Work Item Phlebotomy , 1 Move Work Item To Error path queue , -1 IF Phlebotomy queue.Count Contents > 5 Display Text "Phleb queue has more than 5 patients" , "" , 0 , 0 VL SECTION: CT Scanner ‘First of a set of VL routines to simulate missing notes SET CT Scanner.Operation Time = 150 'First, check to see if the notes are missing IF Notes missing = 1 SET Label_change_DIAG[Hospital number,5] = 1 'If the notes are missing, call the distribution to see if they are recovered SET RND_1 = PJGDistribution 2 IF RND_1 = 0 SET Notes missing = 0 SET EC_patient's_notes = EC_patient's_notes+1 'If the notes aren't missing, check to see if they get mislaid en route to the scanner ELSE SET RND_1 = PJGDistribution 2 IF RND_1 = 0 CALL ASUB,notes,CT SET CT Scanner.Operation Time = 500 SET Label_change_DIAG[Hospital number,5] = 1 VL SECTION: ASUB,notes,CT 'Comment Display Text "Notes mislaid en route to CT" , "" , 0 , 0 'Notes have been mislaid, so update the appropriate digit in the error code to 1 SET EC_patient's_notes = EC_patient's_notes+10 SET Notes missing = 1 'Change the operation time to represent the time required to find the notes and hence document the test results SET CT Scanner.Operation Time = 500 'Note that the operating time MUST be reset to the original value once the notes have been found VL SECTION: ASUB,notes,echo 'Comment SET EC_patient's_notes = EC_patient's_notes+10 SET Notes missing = 1 SET Echo department.Operation Time = 500 VL SECTION: ASUB,notes,MRI 'Comment SET EC_patient's_notes = EC_patient's_notes+10 SET Notes missing = 1



SET MRI Scanner.Operation Time = 500 VL SECTION: ASUB,notes,phleb 'Comment SET EC_patient's_notes = EC_patient's_notes+10 SET Notes missing = 1 SET Phlebotomy.Operation Time = 500 VL SECTION: Echo department Route In After Logic1 'Check the distribution to see if the notes are present with the patient SET RND_1 = PJGDistribution 2 IF RND_1 = 0 CALL ASUB,notes,CT 'Call "ASUB One" to generate appropriate error code SET Label_change_DIAG[Hospital number,1] = 1 SET Label_change_DIAG[Hospital number,2] = 1 VL SECTION: MRI Scanner Route In After Logic1 SET RND_1 = PJGDistribution 2 'Check to see if the notes have been mislaid IF RND_1 = 0 CALL ASUB,notes,CT 'If the notes are not present, call "ASUB One" to generate the error routine SET Label_change_DIAG[Hospital number,3] = 1 SET Label_change_DIAG[Hospital number,4] = 1 VL SECTION: Phlebotomy Route In After Logic1 SET RND_1 = PJGDistribution 2 IF RND_1 = 0 CALL ASUB,notes,CT SET Label_change_DIAG[Hospital number,6] =

Documents

IDENTIFYING AND REDUCING ERRORS IN THE ...Identifying and Reducing Errors in the Operating Theatre 26 September 2005 Version 1.0 8 15.3 Characterising the Paediatric Heart Surgery