Session 2 Quality-Based Procedures: Improving Quality and Consistency ... · Session 2 –Quality-Based Procedures: Improving Quality and Consistency in the Health System Speakers:

Session 2 – Quality-Based Procedures: Improving

Quality and Consistency in the Health System

Speakers: Laura Park-Wyllie, Erik Hellsten,

Stacey Brener, Dr. David Alter, Dr. Doug Lee

Moderator: Laura Park-Wyllie

Presenter Disclosure

2

• Session Name: Quality-Based Procedures: Improving Quality

and Consistency in the Health System

• Presenters: Laura Park-Wyllie (moderator), Stacey Brener, Erik

Hellsten, Dr. David Alter, Dr. Douglas Lee

• Relationships with commercial interests:

– Not Applicable

Disclosure of Commercial Support

• This session has received no commercial support

3

Mitigating Potential Bias

• Not applicable

4

Session Objectives

1. Learn about Health Quality Ontario’s approach to developing

evidence-informed, quality-based episodes of care

2. Learn about a high-level implementation strategy that leverages

stakeholder relationships to encourage the uptake of evidence-

based practices across the health system

5

Data Meets Clinical Intuition:

Developing the QBP Patient Cohorts and

Stratification Approach

Erik Hellsten

7

Excellent Care for All Act, 2010

(c) to promote health care that is supported by the best available scientific evidence by,

(i) making recommendations to health care organizations and other entities on standards of care in the health system, based on or respecting clinical practice guidelines and protocols, and

(ii) making recommendations, based on evidence and with consideration of the recommendations in subclause (i), to the Minister concerning the Government of Ontario’s provision of funding for health care services and medical devices

Why Us? HQO’s Legislated Mandate with Respect to

Funding

Why Quality Based Procedures? (QBP) Context for this Work

The Ministry asked HQO to work with expert panels to develop analysis and recommendations to inform the new Quality-Based Procedures episode-based hospital funding policy for the following clinical areas:

Congestive Heart Failure (Clinical Handbook published)

Chronic Obstructive Pulmonary Disease (Clinical Handbook published)

Stroke (Clinical Handbook published)

Hip Fracture (Clinical Handbook finalized)

Primary Hip and Knee Replacement (In progress)

Pneumonia (In progress)

Key tasks:• Define patient cohort(s), scope of the episode of care, subgroups, risk adjustment approach

• Identification of evidence-based recommended practices, key performance indicators and implementation considerations

Х Out of scope: Unit costing analysis, pricing, payment design

HQO tasked with completing all the above for each area in 5 months

8

HQO’s Quality-Based Procedure Process

9

Expert Panel

Comprised of

clinical,

administrative,

community,

and ministry

experts

Patient cohort and

stratification

approach developed

using administrative

data in conjunction with

the expert panel

Recommended

Practices developed

with evidentiary

support and expert

panel consensus

Implementation

levers and barriers to

the recommended

practices

Expert Panel

Comprised of

clinical,

administrative,

community,

and ministry

experts

OH

TA

C

Indicators to identify

outcome measures of

successful

implementation and

ongoing utilization of

good clinical practices

identified in the QBP

Nodal Network

knowledge dissemination

plan developed jointly with

key health system

partners

Phase II

DevelopmentPhase III

Measurement

& KTNN

Phase I

Preparation


10

Patient cohort and

stratification

approach developed


data in conjunction

with the expert panel

Where We Started:

Mapping the COPD Patient Journey Through an Acute Exacerbation

11

Not

responding

Assess

Severity of

Exacerbation

Patient

presents

at ED with

COPD acute

exacerbation

ED

DIAGNOSTICS- Spirometry

- X-ray

Mild

Severe

Acute

respiratory

failure

TREATMENT

IN ED- Antibiotics

- Cortisosteroids

DISCHARGE PLANNING- Refer to smoking cessation counselling

- Vaccinations (influenza & pneumococcal)

- Refer to pulmonary rehab

- Refer to GP for follow-up visit in 2 weeks

INPATIENT

TREATMENT- Antibiotics

- Steroids

Admit to

medical /

respiratory

ward

HOME- Pulmonary rehab

- GP follow-up

- Outpatient care

Death

Death

Recovery

GO TO

INPATIENT

TREATMENT

Recovery

NPPV + USUAL CARE

FOR RESPIRATORY

FAILURE

Admit to medical /

respiratory ward

Recovery

GO TO

DISCHARGE

PLANNING

Death

IMV + USUAL CARE

FOR RESPIRATORY

FAILURE

Recovery

Death

Elicit patient

preference for

ventilation

Pass

GO TO

DISCHARGE

PLANNING

Fail

SPONTANEOUS

BREATHING

TEST

WEAN

FROM

IMV

GO TO

IMV

Episode trigger event

Episode endpoint

Health state / response

Intervention module

‘Go to’ specified module

Assessment point

Not

responding /

acute R.F.

Not responding

GO TO

NPPV

VAP

Recovery

VAP

GO TO ACUTE

RESPIRATORY

FAILURE

GO TO

DISCHARGE

PLANNING

12

Where We Finished: The Episode of Care Model for Acute Exacerbations of COPD

Patient presents with

suspected exacerbation

of COPD

Usual medical

care (in ED /

outpatient)

NPPV

IMV

Go to usual

medical care

(inpatient)

Go to ventilation

(NPPV or IMV)

Severe Level of care

Usual medical

care (inpatient)

Go to IMV

End of life care

Wean

from IMVDecision on

ventilation

modality or

palliative care

Treatment fails

Recovers

Treatment fails

Assess recovery

ModerateLevel of care

MildLevel of care

Assess recovery

Assess recovery

Assess recovery

Discharge planning

& full clinical

assessment

Assess

level of care

required

Home

Home

Home

Home

Recovers

Recovers

Recovers

Treatment fails

Treatment fails

Discharge planning

& full clinical

assessment

Discharge planning

& full clinical

assessment

Usual medical

care (inpatient)

Discharge planning

& full clinical

assessment

N = 43,215Pr = 1.0

N = 19,337Pr = 0.447

N = 22,054Pr = 0.511

N = 1,824P = .042

N = 773P = .018

N = 1051Pr = .024

Legend

Care module

Assessment node

Episode endpoint

Death

Usual medical

care (inpatient)

13

-10.9%

-2.5%

15.2%

-11.4%

-2.6% -1.6%

17.8%13.8%

-4.8%-7.7%

-35.8%

5.9%

47.2%

-3.4%

3.5%

-3.8%-6.4%

11.0%

-40.0%

-20.0%

0.0%

20.0%

40.0%

60.0%

% D

iffe

ren

ce

in

Re

so

urc

e In

ten

sit

y W

eig

ht

(RIW

)

Patient Characteristics

Comorbidity Index = 2

Age ≥75 yrsActivities of Daily Living > 2

(Severely Impaired)

Comorbidity

Index = 0

Patient Characteristics Driving Variation in Hip Fracture Utilization

14

Stratifying the Hip Fracture Population:

Drawing on Clinical Experience

0

100

200

300

400

500

600

13

5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

# Cases (2010 / 11)

Total Acute IP LOS (Days)

Admit from Community - 'Complex'

Admit from LTC

Admit from Community - 'Healthy'

Median LOS: 10 days

Median LOS: 6 days

Median LOS: 8 days

15

The Hip Fracture Episode of Care: Presentation to 90 Days Post-Admission

Hip Fracture Inpatient

Orthogeriatric Care ProgramPatient presents

with suspected

hip fracture

Assess and

medically

stabilize

No surgery

Home with rehab / follow-up

N = 12,860

Pr = 1.0

Counts and proportions from Discharge

Abstract Database (2011/12) and Hip

Fracture Scorecard (Q1Q2 FY2011-12)

Most responsible diagnosis or comorbidity

diagnosis of S72.0*, S72.1* or S72.2*,

excluding S72.00*

Legend

Care module

Assessment node

Pathway endpoint

Decision to treat /type of surgery /anesthesiaon treatment

Conservative

treatment

Surgery

Decision on post-acute care path

Post-op

stabilization

& early

mobilization

Home-based

rehabilitation

Home withfollow-up

Pr = 0.18

Long-termcare (with rehab)

Long-termcare

Inpatient

rehabilitation

Pr = 0.42 Pr = 0.09 Pr = 0.21

Transfer in

/ out of

hospital for

surgery

Repatriation to

index hospital

CCC / slow

stream rehab

Patient’s pre-fracture level of care

LTCCommunity

‘Healthy’

Community

‘Complex’

N = 7,066

Pr = 0.548N = 3,557

Pr = 0.276

N = 2,275

Pr = 0.176

Post-acute care to 90 days

following index hospitalization

Pre-op

careSurgery

Episode of Care Evidence Synthesis for

Recommended Practices

Stacey Brener


17

Expert Panel

Comprised of

clinical,

administrative,

community,

and ministry

experts

Patient cohort and

stratification

approach developed


data in conjunction with

the expert panel

Recommended

Practices developed

with evidentiary

support and expert

panel consensus

Implementation

levers and barriers to

the recommended

practices

Expert Panel

Comprised of

clinical,

administrative,

community,

and ministry

experts

OH

TA

C

Indicators to identify

outcome measures of

successful

implementation and

ongoing utilization of

good clinical practices

identified in the QBP

Nodal Network

knowledge dissemination

plan developed jointly with

key health system

partners

Phase II

DevelopmentPhase III

Measurement

& KTNN

Phase I

Preparation


18

Recommended

Practices developed

with evidentiary

support and expert

panel consensus

Sample Care Pathway

19

Patient

Groups

Inpatient care

Module 1

Module 2

Module 3 Module 5Module

6

Module

7Module 4

Decision

Module

8Module 9

Module

10

Decision

End

Post-acute care

Legend

Care module

Assessment node

Episode endpoint

Index

Event

Index event

1 to 20 recommendations for

each of the modules

20

Approach to Applying Evidence to Modules

• Identify guidelines covering entire pathway with guidance from medical librarians, and confirmed with Expert Panel

• Use AGREE II instrument to rate and identify 3-4 best clinical guidelines developed with most methodological rigour, including at least 1 contextually relevant (Canadian) guideline.

Patient

Groups

Inpatient care

Module 1

Module 2

Module 3 Module 5 Module 6Module

7Module 4

Decision

Module 8 Module 9 Module 10

Decision

End

Post-acute care

Index Event

Appraisal of Guidelines for Research & Evaluation II

6 domains

1) Scope and Purpose

2) Stakeholder Involvement

3) Rigour of Development

4) Clarity of Presentation

5) Applicability

6) Editorial Independence

Evidence-based Care Module

21




• Begin to populate the relevant modules with Canadian guidelines, while flagging controversy between the guidelines

• Identify related previously conducted HQO evidence based analyses and OHTAC recommendations

Patient

Groups

Inpatient care

Module 1

Module 2


7Module 4

Decision


Decision

End

Post-acute care

Index Event

Decision Determinants Framework which is considered for all

OHTAC recommendations:

o Overall clinical benefit

o Value for money

o Consistency with societal and ethical values

o Feasibility of adoption into the health care system


22







• A Rapid Review may be conducted for areas of conflict or controversy or where uncertainty around the evidence exists

• In some cases, it may be appropriate for HQO to proceed to a full Evidence based analysis (EBA) and revise the episode of care recommendations accordingly.

Patient

Groups

Inpatient care

Module 1

Module 2


7Module 4

Decision


Decision

End

Post-acute care

Index Event

Evidence Products Comparison

23

Rapid Review

Vs.

Evidence Based Analysis

Question 1 Specific Question Potentially Multiple Questions

Time Frame 2 Weeks 16 Weeks

Literature Search 5 to 10 years Comprehensive

Types of Studies Systematic reviews/ Meta-

analyses

Comprehensive

Outcomes 2 (up to 4) No limit

Type of Analysis Summary of a synthesis report

- Summarize as reported in

SR

Original Synthesis Report

- Meta-analysis + Qualitative Analysis

- Selection of appropriate studies, subgroups

Quality

Assessment

Use SR assessment or

GRADE

GRADE all outcomes comprehensively

Economics None Full Economic Analysis

Contextualization Limited expert panel feedback • Multiple Expert panel meetings on a specific

topic, contact primary authors and additional

experts in field,

• OHTAC review and recommendation

• Decision Determinants

Inferences Very Low/Cautious Interpretation

of Findings

Moderate-High/Evidence Based Conclusions

24






• A Rapid Review may be conducted for areas of conflict or controversy or where uncertainty around the evidence exists

• In some cases, it may be appropriate for HQO to proceed to a full Evidence based analysis (EBA) and revise the episode of care recommendations accordingly.

• Utilize expert consensus where evidence is limited, not contextually relevant or nonexistent

Patient

Groups

Inpatient care

Module 1

Module 2


7Module 4

Decision


Decision

End

Post-acute care

Index Event


2525

Patient

Groups

Inpatient care

Module 1

Module 2

Module 3 Module 5Module

6

Module

7Module 4

Decision

Module

8Module 9

Module

10

Decision

End

Post-acute care

Legend

Care module

Assessment node

Episode endpoint

Index

Event

Index event

1 to 20 recommendations for

each of the modules

Sample Care Pathway

QBP – Physician Perspective

Dr. David Alter

Perspectives

• Co-Chair of the Quality Based Procedures, Congestive Heart

Failure

• Health Services Research

• Physician

27

RationaleAligning system-expenditures with care-quality in hopes of

improving efficiency, accountability, and outcomes of care

28

Evidence

29

Empirical Data

30

1. Prevalence (i.e., proportion of patients in different pathways)

2. Interferences on quality indicators

Empirical Data

Independent predictors of 30-day death or re-admission

among patients hospitalized with congestive heart failure

31

Consensus – Pathway Development

32

Consensus – Pathway Development

33

Implementation

34

Health Quality Ontario (HQO) Knowledge

Translation Nodal Network

Laura Park-Wyllie

HQO Knowledge Translation Approach

1. Engagement with experts and stakeholders is integrated

throughout evidence development process.

2. Knowledge dissemination plan and implementation

considerations are developed jointly with key health system

partners.

36

Integrated Knowledge Translation Nodal Network

Framework

37

OHA Implementation and KT

OMA Implementation and KT

CAHO Implementation and KT

CCAC Implementation and KT

OCFP Implementation and KT

CCN, OSN, Implementation and KT

LHINs Implementation and KT

RNAO Implementation and KT

Macro Node

Collaboration with Experts &

Stakeholders

Secondary Node [

Key Strategic Partners Activate

Knowledge Translation Networks to Promulgate

and Implement Episode of Care

HQOMonitor Key Performance Indicators for

Episode of care

Review and

Monitor KPIs for

Episode of Care

HQO

HQO

Development of Episode of Care and Indicators

Key Stakeholders

Experts

EDS Hybrid Model

Multi-Stakeholder Integrated Knowledge Translation

Nodal Network Process From Evidence Development to Knowledge Translation/Implementation Support for Best Practice Implementation

KTNN Phases KTNN Process

HQO Expert panel

Clinical Evidence-based

Best Practice

KT Strategy and

Implementation Planning

Development of

Implementation Tools

Pilot Test

Delivery and Dissemination

Ongoing Implementation

Support

Consideration of Feedback

Engagement and Input

Identify Clinical

Champions

Develop Collaborative

KT Strategy

Identify Implementation

Tools

Conduct Early

Evaluation

Lead the Dissemination

Utilize Indicators,

Support Networks

Bring Feedback via

Loop

• KTNN partners nominate experts to panels.

• KTNN may participate in panel meeting, if appropriate

• HQO Chairs and expert panel members become

clinical champions and provide leadership for

adoption.

• Develop strategy for knowledge translation and

implementation support.

• Identify tools and levels that could be developed.

• HQO and KTNN partners develop tools as

relevant to their constituencies.

• If appropriate, KT partners may evaluate

implementation approach.

• Provincial and regional meetings

• Target stakeholder briefings, Educational Sessions,

Training Workshops, Newsletters, Toolkits

• Episode of Care Indicators

• Regional Support Networks

• Community of Practice Networks

• KTNN partners provide feedback from field to HQO

to ensure products are useful to team.

Moving Beyond the QBP to Evaluation and

Implementation

Dr. Douglas Lee

HF Recommendations – Acute Phase

40

Mechanical ventilation PA monitoring

BIPAP IABP, assistive devices

Oxygen Monitor electrolytes, renal

function, troponins, CXR

Lasix IV or PO Record fluid input/output

IV vasoactive agents Record weight

Telemetry Other therapies (ASA, IV

heparin, statins)

1:1 nurse-to-patient ratio ECG

ACE inhibitors/ARBs Assessment of precipitating

factors (e.g., infection,

ischemia)

Beta-blockers Discuss advanced directives

Ultrafiltration Vital signs

HF Recommendations – Subacute Phase

41

Daily weights Renal function assessment

6-hr input/output Assessment for ischemia:

Coronary angiography

Non-invasive risk

stratification

Revascularization

procedure

Salt restriction Assessment of valvular heart

disease

Evaluation for valve surgery

or repair

Fluid restriction Screen for complications (e.g.,

arrhythmia, urosepsis, COPD,

renal failure, pneumonia)

Electrolytes

HF Recommendations – Discharge Planning

42

Diuretic monitoring and

management

Predischarge functional capacity

and mobility assessment

Evidence-based

pharmacotherapy

Predischarge cognitive and

social support assessment

Counselling

Medication

Lifestyle (alcohol, smoking)

Daily weight and self-

monitoring

Diet

Physical activity

Advanced care directives

Physician appointments:

GP/FP, Internal Medicine,

Cardiology

Timely documentation

Discharge notes dictated &

sent to PCP within 1 week

Moving Beyond the QBP: HF Indicators

43

44

Acute Heart Failure Risk Stratification

• Respiratory distress

• Hypoxemia

• Severity of pulmonary edema

• Poorly responsive to furosemide

• Hemodynamic compromise

• Significant arrhythmias

• Positive troponin

• Concomitant acute life-threatening disorders

45

46

Moving Beyond the QBP:

Improving Quality of HF Care Decisions in the ED

47

0100200300

400500600700800900

1,000

0.0

0

0.0

4

0.0

8

0.1

2

0.1

6

0.2

0

0.2

4

0.2

8

0.3

2

0.3

6

0.4

0

0.4

4

Nu

mb

er

of

Pati

en

ts

Admitted

Discharged

Overlap

Region

For Death

Analysis

(30 day)

Predicted Probability of Death

Overlap Predicted Prob of Death:

4.8-6.5% 30-d mortality

1189 Discharged

3704 Admitted

Lee DS, et al. Circulation: Heart Failure 2010; 3:228-35

Mortality: Discharged vs. Admitted

48

0

2

4

6

8

10

12

14

0 10 20 30 40 50 60 70 80 90 100

% M

ort

ality

Days to Death

Lee DS, et al. Circulation: Heart Failure 2010; 3:228-35

Early Follow-up of HF: Improved Survival

49

0.80

0.82

0.84

0.86

0.88

0.90

0.92

0.94

0.96

0.98

1.00

100 150 200 250 300 350 400

Survival Time (days)

% S

urv

iva

l Card+PC

Card

PC

No MD

A: Card+PC vs. PC:

HR 0.79 (0.63-1.00) p=0.045

B: PC vs. NoMD:

HR 0.75 (0.64-0.87) p<0.001

A

B

Lee DS, et al. Circulation 2010; 122:1806-1814

Rationale: In Ontario Emergency Departments

• Inefficiency: Some low risk HF patients are unnecessarily

admitted to hospital instead of having effective community based

follow-up care

• Safety: Some high risk HF patients are inappropriately

discharged – will die at home

50

Aim Statement

• To reduce admission rates of low-risk heart failure (HF) patients

presenting to the emergency department by 25% while reducing

the discharge of high risk HF patients.

51

Quality Improvement TeamChristopher Sulway, PT, TCLHIN

Douglas S. Lee, MD, UHN

Shanas Mohamed, RN, UHN

Medical Staff Nursing Allied Health /

Admin

H. Ross, MD – PMCC HF Lead

S. Sabah, MD – ED Assoc Head

H. Abrams, MD – Chief, GIM

B. Coke, MD – GIM

A. Woo, MD – Head, Echo Lab

H. Amad, MD – UHN Cardiology

R. Iwanochko, MD – TWHCardiology

Site Lead

S. McIntaggart – VP Clinical

L. Flockhart – PMCC Clinical

Director

K. Partridge – Amb. Clinics

P. Neilsen – Cardiology Ward

L. Belford – ACNP, PMCC HF

P. Lui – Pharmacy

O. Fernandes – Pharmacy

S. Miguel – Clinics booking

L. Biclar – Echo booking

CCAC

Summary of Findings

• Reasons for high number of low risk HF being admitted

• No criteria and poor practices to assess risk in HF patients

• No process in ED to monitor low and medium risk patients to decide if admission is needed

• No reliable follow up in community

– Too many phone calls to ensure appropriate follow-up

– Concern of poor transition (slip through crack)

• No easy way to make a referral 24-7

53

54Date of download:

6/15/2012

Copyright © The American College of Physicians.

All rights reserved.

From: Prediction of Heart Failure Mortality in Emergent Care: A Cohort Study

Emergency

Heart failure

Mortality

Risk

Grade

Ann Intern Med. 2012;156(11):767-775. doi:10.7326/0003-4819-156-11-201206050-00003

EHMRG HF Risk Stratification in the ED

55

Intervention – Quality Improvement in CHF Care

(QUICC) Initiative

1. Risk stratification: EHMRG decision support algorithm

2. Checklist to assist in deciding safety of discharge

3. Rapid 24-hr follow-up clinic

4. Automatic referral to rapid home care visit

5. ED virtual observation unit

56

57

58

0

10

20

30

40

50

60

Low Med High

0

10

20

30

40

50

60

Low Med High

Potential Impact in Ontario

Risk Profiles (Fiscal 2007 data)

Hospital Admitted Discharged% %

34%

42%

24%

51%

37%

12%

Ontario Statistics: Follow-up (Fiscal 2007 data)

59

4852

50 51

58

42

0

10

20

30

40

50

60

70

80

90

100

Low Med-

High

Low LowMed-

High

Med-

High

% Follow-up

by FP w/in

2 days

Family MD or C

w/in 2 days

% Follow-up

by C w/in

7 days

% o

f P

tsD

isc

ha

rge

d f

rom

ED

Reflections on Current and Future State

• HQO began it’s QBP program just over 1 year ago.

• To date, HQO has developed 6 evidence-based, best practice, clinical handbooks to inform quality-based funding policy for Ontario.

• An additional 5 handbooks are actively in-development with provincial expert advisory panels (community-based focus).

• The QBP program of work within HQO has led to an active and productive period of developing customized evidence synthesis, analytic, and engagement methods to support the development of QBP evidence-based best practices.

60

Reflections on Current and Future State

• New innovative research (risk stratification) and proof-of-concept

programs (specialized heart failure clinic models) have been

associated with HQO’s QBP work.

• A recent focus in the evolution of HQO’s QBP program has been

on collaborating with key strategic health system partners to

facilitate the knowledge translation and uptake of the QBP best

practices.

• The Ministry is using QBP clinical best practices to develop the

funding policies (episode of care pricing) under a separate

timeline.

• Looking forward 2013-2014: Community-based QBPs

61

Thank You

62

Documents

Session 2 Quality-Based Procedures: Improving Quality and Consistency ... · Session 2 –Quality-Based Procedures: Improving Quality and Consistency in the Health System Speakers: