OriginalArticle Organizational Culture, Team Climate, and

Original Article

Organizational Culture, Team Climate, andQuality Management in an Important PatientSafety Issue: Nosocomial Pressure Ulcers

Marije Bosch, PhD, Ruud J.G. Halfens, PhD, Trudy van der Weijden, MD, PhD, Michel Wensing, PhD, Reinier Akkermans, MSc,Richard Grol, PhD

ABSTRACTBackground: Increasingly, policy reform in health care is discussed in terms of changing organizational

culture, creating practice teams, and organizational quality management. Yet, the evidence for thesesuggested determinants of high-quality care is inconsistent.

Aims: To determine if the type of organizational culture (Competing Values Framework), team climate(Team Climate Inventory), and preventive pressure ulcer quality management at ward level were relatedto the prevalence of pressure ulcers. Also, we wanted to determine if the type of organizational culture,team climate, or the institutional quality management related to preventive quality management at theward level.

Methods: In this cross-sectional observational study multivariate (logistic) regression analyses wereperformed, adjusting for potential confounders and institution-level clustering. Data from 1,274 patientsand 460 health care professionals in 37 general hospital wards and 67 nursing home wards in theNetherlands were analyzed. The main outcome measures were nosocomial pressure ulcers in patients atrisk for pressure ulcers (Braden score ≤ 18) and preventive quality management at ward level.

Results: No associations were found between organizational culture, team climate, or preventivequality management at the ward level and the prevalence of nosocomial pressure ulcers. Institutionalquality management was positively correlated with preventive quality management at ward level (adj. β

0.32; p < 0.001).Conclusions and Implications: Although the prevalence of nosocomial pressure ulcers varied con-

siderably across wards, it did not relate to organizational culture, team climate, or preventive qualitymanagement at the ward level. These results would therefore not subscribe the widely suggested impor-tance of these factors in improving health care. However, different designs and research methods (that gobeyond the cross-sectional design) may be more informative in studying relations between such complexfactors and outcomes in a more meaningful way.

KEYWORDS organizational culture, pressure ulcer, prevalence, quality of health care, institutional policies,team climate

Worldviews on Evidence-Based Nursing 2010; x(x):1–11. Copyright ©2010 Sigma Theta Tau International

Marije Bosch, Researcher, Scientific Institute for Quality of Healthcare, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; Ruud J.G. Halfens, AssociateProfessor, Department of Health Care and Nursing Science, Maastricht University, Maastricht, The Netherlands; Trudy van der Weijden, Professor, Implementation of Clinical PracticeGuidelines, School for Public Health and Primary Care (CAPHRI), Maastricht University, Maastricht, The Netherlands; Michel Wensing, Associate Professor, Scientific Institute forQuality of Healthcare, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; Reinier Akkermans, Statistician and Doctoral candidate, Scientific Institute forQuality of Healthcare, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; Richard Grol, Professor Quality of Care, Scientific Institute for Quality of Healthcare,Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.

This research was funded by the European Commission, Fifth Framework, Rebeqi project, contract no. QLRT-2001–00657.

Address correspondence to Marije Bosch, Scientific Institute for Quality of Healthcare, Radboud University Nijmegen Medical Centre, PO. Box 9101, 6500 HB, Nijmegen, TheNetherlands; [email protected]

Accepted 18 January 2010Copyright ©2010 Sigma Theta Tau International1545-102X1/10

Worldviews on Evidence-Based Nursing �xxxx 2010 1

Organizational Factors in Pressure Ulcer Prevention

INTRODUCTION

There is a growing belief that one of the keys to im-proved health care practice and better patient safety

is structural support at different organizational levels (Sol-berg 2000; Ferlie & Shortell 2001). Such support enablesthe implementation of systems to facilitate clinical deci-sion making (Moss et al. 1998; Solberg 2000). To securethe gains in quality, it has been suggested that changes inculture are needed alongside the structural changes (Scottet al. 2003a). Different types of organizational cultureshave been found to relate to health care performance out-comes including quality improvement activities (Shortellet al. 1995), team functioning (Strasser et al. 2002), careprocess indicators (Shortell et al. 2000), perceived orga-nizational performance (Gerowitz 1998), and evidence-based practices (Shortell et al. 2001). Few studies havedemonstrated a link to such patient outcomes as patientsatisfaction (Meterko et al. 2004) or physical and men-tal health outcomes (Shortell et al. 2000). In addition,the importance of effective health care teams in attainingsafe and high-quality care has been stressed (Institute ofMedicine 2001). Teamwork provides a means to reducecare fragmentation, widen professional skills, and han-dle the complexity of modern care (Firth-Cozens 1998;Friedman & Berger 2004). Higher levels of teamworkhave been found to produce improved outcomes in termsof overall clinical performance, (Haward et al. 2003) ab-sence of hospital physicians due to sickness (Kivimaki et al.2001), and patient outcomes (Campbell et al. 2001; Boweret al. 2003; Wheelan et al. 2003). For instance, a study inAmerican intensive care units found that staff members ofunits with mortality rates that were lower than predictedperceived their teams as functioning at higher stages ofgroup development. They perceived their team membersas less dependent and more trusting than did staff mem-bers of units with mortality rates that were higher thanpredicted (Wheelan et al. 2003).

The development of pressure ulcers constitutes a preva-lent and serious health care problem. Patients remainlonger in hospital than otherwise might be needed and of-ten require intensive treatment. Studies show a wide rangeof prevalence rates from less than 5% (Lyder 2001) to over40% (Thomson & Brooks 1999). The total cost of pressureulcers have been estimated to be around £1.4–£2.1 bil-lion (4% of total NHS expenditure) annually in the UnitedKingdom (Bennett et al. 2004), around US$11 billion in2006 in the United States (Russo et al. 2008), and the mostconservative estimate of a Dutch study was 1% of the totalhealth care budget (Severens et al. 2002). However, mostpressure ulcers can be prevented with high-quality care,and they therefore constitute an important quality-of-care

indicator (Lyder 2003). For example, in 2003, pressure ul-cers were introduced as a quality indicator by the Nether-lands Health Care Inspectorate.

The prevention of pressure ulcers requires concertedaction on the part of health care professionals. Effectivecommunication and careful coordination of tasks and re-sponsibilities are thus necessary. This means that the team-work and organizational culture within an institution maycertainly influence the way the pressure ulcer preventivecare is organized and structured at ward level and, perhaps,the prevalence of nosocomial pressure ulcers directly. Withthe increasing demand for transparency in patient safetyissues, it is of utter importance for policy makers to identifyfactors that may contribute to low prevalence figures.

In this study, we aimed to determine if the type of or-ganizational culture, team climate, and preventive qual-ity management at ward level related to the prevalence ofnosocomial pressure ulcers in patients at risk for pressureulcers. Also, we wanted to determine if the type of organi-zational culture, team climate, or the institutional qualitymanagement related to preventive quality management atthe ward level.

METHODS

Design and PopulationSince 1998, the prevalence of pressure ulcers has been an-nually measured in the Netherlands via standardized pointprevalence screening (Bours et al. 2002) on a prespecifiedday. In 2005, each Dutch health care organization was in-vited to participate on April 4. In participating wards, allpatients that handed in informed consent were includedin the measures. Adult patients at risk for pressure ulcers(Braden score ≤ 18; age ≥ 18 years) were included in theanalyses. Besides patient outcomes, various structural fac-tors and indicators to measure pressure ulcer preventivequality management at ward and organizational level werescored.

In this study, a secondary analysis of data collected dur-ing the 2005 national prevalence study for those patientsat risk of pressure ulcers was performed. All of the 61 gen-eral hospitals and 92 nursing homes, which participated inApril 2005, were invited to complete an additional ques-tionnaire on team climate and organizational culture. Thequestionnaires were sent to the managers of two randomlyselected wards per institution in May and June of 2005for distribution to one doctor and four nurses or nursingassistants from the same patient care team on a ward.

Four hundred ninety-four respondents from 117 wardsin 25 hospitals and 36 nursing homes returned our ques-tionnaire for team climate and organizational culture.

2 xxxx 2010 �Worldviews on Evidence-Based Nursing


Response rates were 41% for the hospital wards and 39%for the nursing home wards; 25% of the nonresponse wascaused because the surveys could not be allocated to aspecific ward. Of 2,686 patients, who handed in informedconsent, 1,322 patients were excluded because they didnot meet our patient inclusion criteria (Braden score ≤ 18;age ≥ 18 years). Furthermore, when less than three careproviders of a ward returned our team climate and organi-zational culture questionnaires (8 wards, 15 respondents),the patients from this particular ward were excluded aswell (n = 90). Therefore, the analyses included a total of1,274 patients on 104 wards with measures derived from460 health care providers. The prevalence screening wasapproved by the Maastricht University Medical Centreethics committee, and all included patients handed in writ-ten informed consent.

MeasuresOur clinical outcome was the case-mix-adjusted pointprevalence of nosocomial pressure ulcers (grade II orworse) in risk patients per ward. Patients’ risk for pressureulcer development was assessed using the Braden Scale(Bergstrom et al. 1987), which encompasses six subscalesthat conceptually reflect degrees of sensory perception,moisture, activity, mobility, nutrition, and friction andshear. All subscales are rated from 1 (worst) to 4 (best),apart from friction and shear, which is rated from 1 to3. The subscales were summed for a total score, rangingfrom 6 to 23. In this study, a cut-off point of 18 was cho-sen, since this is recommended for all settings (Bergstromet al. 1998). Pressure ulcers were assessed according to theEuropean Pressure Ulcer Advisory Panel (EPUAP) gradingsystem (EPUAP 2008). Grade I ulcers were not countedas it is difficult to accurately diagnose nonblanchable dis-coloration (Bergstrom et al. 1996). And only nosocomialpressure ulcers were counted as this means that the ul-cers developed in the current institution and thereforerepresent quality of care. In order to standardize mea-surements, prior to measurement, health care personnelinvolved were trained by the coordinator in each institu-tion and the research project team. Each patient was scoredby a health caregiver working on the patients’ ward, andby a health care provider working on a different ward.Disagreements were discussed and solved with an inde-pendent third trained care provider (please refer to Bourset al. 2002 for more information).

The second outcome was the sum score of eight indi-cators for preventive quality management at ward level,which was measured by a checklist completed by the headof each participating ward (Bours et al. 2002; Halfens et al.2005) (see Figure 1; Cronbach’s alpha 0.69).

The first independent variable was the sum score for the11 indicators of quality management at the institutionallevel (Bours et al. 2002; Halfens et al. 2005) (see Figure 1;Cronbach’s alpha 0.62), measured by a checklist that wascompleted by the coordinator in each participating organi-zation. The quality indicators included in this study wereformulated by a team of Dutch experts in the pressure ulcerfield, and were based on (inter)national guidelines (EPUAP1998; Dutch Institute for Healthcare Improvement [CBO]2002), and expert opinion.

To measure organizational culture, we used the Com-peting Values Framework (CVF). The framework dis-tinguishes two dimensions: “internally oriented” versus“externally oriented,” and “stability” versus “flexibilityand change,” resulting in four ideal types of culture (seeFigure 2). The framework encompasses five questions re-ferring to particular aspects of organizational life: overallcharacter, leadership style, institutional bonding, strategicemphasis, and reward system. For each of these questions,respondents are asked to distribute a total of 100 pointsacross four sets of statements in such a manner that theorganizational nature of the institution is best represented(Quinn & Kimberly 1984). For each of the five questions,nonblank respondent errors were corrected for by propor-tionally adjusting the responses to add to 100. The numberof points allocated to each type of organizational culturewas then averaged across the five questions. Cronbach’salphas were 0.69 for group culture, 0.56 for developmen-tal culture, 0.51 for hierarchical culture, and 0.41 for therational culture statements. The scores reported for thegroup, developmental and hierarchical types of organiza-tional culture were somewhat lower than those reported inother studies in health care settings (Shortell et al. 2001;Meterko et al. 2004) but still considered acceptable in lightof the low number of items (5) within the concepts. Foreach ward, the mean scores for the four types of organi-zational culture were determined. The data were opera-tionalized in two manners, which have been described inprevious studies (Shortell et al. 2000; Meterko et al. 2004).First, the separate scores for the different types of organi-zational cultures were treated as a continuous variable.Second, for each ward the dominant type of culture (thetype that received the highest mean score) was determined,which led to a categorical variable.

Teamwork was measured with the short version ofthe Team Climate Inventory (TCI) (Anderson & West1994; Loo & Loewen 2002). The underlying theory as-sumes that effective team performance often results fromteam activities, which are characterized by: (1) a focuson clear and realistic objectives (vision), (2) a participa-tory and nonthreatening climate for team interaction (par-ticipative safety), (3) commitment to high standards of



Figure 1. Indicators of the pressure ulcer quality management at the level of the institution and at the level of the ward.

performance (task orientation), and (4) support for at-tempts at innovation (support for innovation). The in-strument uses 14 statements or questions reflecting theunderlying factors, and 5-point Likert scales ranging from1 (strongly disagree, indicating a “poor” team climate) to5 (strongly agree, indicating high levels of team climate).Overall Chronbach’s alpha for the 14 questions was 0.87.For each subscale, mean scores were calculated per indi-vidual and then averaged to ward-level scores. We thencombined these to one overall ward score (Campbell et al.2001).

Both the CVF and the TCI were translated into Dutch,following the guidelines for cross-cultural adaptation(Beaton et al. 2000). Items were translated from Englishto Dutch by a bilingual researcher, followed by back trans-lation from Dutch to English by a second bilingual re-searcher. Discrepancies between the original questionnaireitems and the back translation were identified and solvedwith a third bilingual researcher and all researchers were

familiar with the theoretical constructs. A pilot test wasperformed to measure acceptation, item response, and ter-minology, on the basis of which one small adaptation inwording was made. While the measures were administeredat the individual level, the climate and culture variables re-quired shared perceptions. Analyses of variance verifiedthat the individual responses for team climate and organi-zational culture could be validly aggregated at the level ofthe team. That is, the within-team variability was clearlyless than the between-team variability (F-values rangingfrom 1.9 to 2.3, p < 0.001).

The following—possibly confounding—variables(Bours et al. 2004) were also included in the analyses:age, derived from the medical records; and type of ward,derived from the ward checklist and malnutrition. Thedefinition of malnutrition was based on the literatureand consultation of experts in the field of malnutritionin Europe, and was operationalized by one of the threefollowing criteria: (1) a Body Mass Index (BMI) less than



Figure 2. Competing values framework.

18.5 with age <85, or a BMI of less than 24 with age≥85; (2) unintentional weight loss (6 kg during the last6 months or 3 kg during the last month); or (3) a BMIbetween 18.5 and 20 with age <85, and between 24and 26 with age ≥85 in combination with no nutritionalintake for 3 days or reduced intake for more than10 days (Meijers et al. 2009). If any one of these threecriteria could be confirmed, the patient was diagnosed asmalnourished (Halfens et al. 2006).

Statistical AnalysesDescriptive statistics were calculated for the patients andthe wards. Bivariate correlations were examined to checkfor particularly high intercorrelations. Univariate associa-tions between the different variables were examined usingSPSS 12.0.1 (SPSS Inc., Chicago, IL; ANOVAs, t-tests, Pear-son’s correlation, and Chi-square test). Multilevel logisticregression analyses were performed using the Glimmixprocedure in SAS for Windows V8.2 (SAS Institute, Cary,NC) with patients (level 1) nested within institutions (level2) to explore associations between organizational culture,team climate, and preventive quality management at wardlevel on the one hand and the prevalence of pressure ulcerson the other. It was hard to attain a good estimate of therandom ward effect due to the small number of wards perinstitution. The random effect was therefore distributed tothe level of the institution. Since we were interested in theeffect of each of our variables of interest separately (culture,team climate, and quality management), we used separatemodels with one of these variables at a time. The four types

of organizational culture could not be examined togetherwithin a single model as the scores were not independent.Seven separate models were thus created for the prevalenceof pressure ulcers: four models for the four organizationalcultures, one for the dominant organizational culture, onefor the team climate and, finally, one for preventive qual-ity management at ward level. In each model, the effectsof patient age, malnutrition, and type of ward were con-trolled for. Adjusted odds ratios (adj. OR) were calculatedto describe the associations between the determinantsand the outcome. An OR > 1 means a positive associa-tion with the outcome. Two-sided p-levels of <0.05 wereconsidered statistically significant. Finally, the influencesof organizational culture, team climate, and institutionalquality management on preventive quality management atward level were examined. General Linear Model regres-sion analysis was performed using SPSS 12.0.1, and thetype of ward was controlled for.

RESULTS

Patient and Ward CharacteristicsTable 1 shows patient and ward characteristics. The to-tal number of patients with nosocomial pressure ulcers(grade II or worse) was 129. The mean point preva-lence of nosocomial pressure ulcers (grade II or worse)in risk patients was higher in the general hospitals as com-pared to the nursing homes (14.2% and 9.7%, respectively,p = 0.02). We distinguished 7 different types of wards:in general hospitals 12 surgery wards, 15 internal wards,



TABLE 1Characteristics of patient population (N = 1,274) and wards (N = 104)

GENERAL HOSPITAL NURSING HOMEPATIENTS (N = 235) PATIENTS (N = 1,039)

Age (mean, SD) 73.5 (14.3) 81.0 (11.6)Male (mean %) 48.5% 27.5%Braden score (mean, SD) 14.8 (2.7) 14.5 (2.6)

GENERAL HOSPITAL WARDS (N = 37) NURSING HOME WARDS (N = 67)

Prevalence of nosocomial pressure ulcers(grade II or worse) (mean %, SD)

14.2 (0.0–25.0)a 9.7 (0.0–12.5)a

Malnutrition (mean %, SD) 31.2 (29.3) 18.6 (0.0–29.6)a

Preventive pressure ulcer quality managementat ward level (0–8) (mean, SD)

4.8 (1.9) 5.2 (1.8)

Organizational culture (mean, SD)Group (0–100) 28.6 (8.6) 32.3 (11.9)Developmental (0–100) 18.1 (6.4) 18.3 (5.8)Hierarchical (0–100) 31.9 (8.0) 29.2 (9.4)Rational (0–100) 21.4 (5.6) 20.2 (7.3)

Dominant culture (%)Group 37.8% 53.0%Developmental 5.4% 1.5%Hierarchical 51.4% 40.9%Rational 5.4% 4.5%

Team climate overall score (1–5) (mean, SD) 3.71 (0.24) 3.75 (0.34)Institutional quality management (0–11) (mean, SD) 8.5 (2.0) 6.3 (2.6)aDue to skewed distributions: 25th and 75th percentiles presented.

and 10 other wards were included, and in nursing homes36 psychogeriatic, 7 somatic reactivation, 18 somatic longstay, and 6 other wards. In 51.4% of the general hospitalwards, hierarchical culture scored highest (mean acrosswards of 31.9); in 53% of the nursing hospital wards, groupculture scored highest (mean across wards of 32.3). Theteam climate scores were 3.71 for general hospitals and3.75 for nursing homes. The general hospitals scored bet-ter with respect to the institutional quality managementthan the nursing homes (8.5 vs. 6.3).

Multivariate Associations with Prevalenceof Nosocomial Pressure UlcersTable 2 shows the associations between different typesof organizational cultures, team climate, and preventivequality management at ward level on the one hand andprevalence of nosocomial pressure ulcers on the other.Prevalence did not relate to any of the variables. Overall,the pooled dominant culture variable was not significantlyrelated to pressure ulcers (p = 0.17), although it showeda tendency in favor of a predominantly rational ward ascompared to predominantly group or hierarchical culture

TABLE 2Influences of different types of organizational cultures, teamclimate, and preventive pressure ulcer quality management atward level or dominant type of organizational culture onprevalence of nosocomial pressure ulcers adjusted for patient age,malnutrition, and type of ward

ADJ. OR 95% CI

Group culture (0–100) 1.00 0.98, 1.02Developmental

culture (0–100)1.02 0.98, 1.06

Hierarchical culture (0–100) 0.99 0.97, 1.02Rational culture (0–100) 0.99 0.96, 1.02Team climate (1–5) 1.00 0.50, 2.02Preventive pressure ulcer

quality managementat ward level (0–8)

0.96 0.88, 1.06

Dominant culture (categorical)∗

Group 4.21 1.20, 14.79Developmental 4.03 0.66, 24.83Hierarchical 4.00 1.13, 14.23Rational 1.00 –

∗p = 0.17



TABLE 3Influences of different types of organizational cultures, teamclimate, institutional quality management, or dominant type oforganizational culture on preventive pressure ulcer qualitymanagement at ward level adjusted for type of ward

ADJ. B 95% CI

Group culture (0–100) −0.01 −0.04, 0.03Developmental culture (0–100) −0.02 −0.08, 0.05Hierarchical culture (0–100) 0.01 −0.03, 0.06Rational culture (0–100) 0.01 −0.05, 0.06Team climate (1–5) 0.86 −0.32, 2.04Institutional quality management (0–11) 0.32 0.18, 0.47∗

Dominant culture typeGroup −0.04 −1.77, 1.69Developmental 1.03 −1.66, 3.72Hierarchical 0.27 −1.46, 1.99Rational – –

∗p < 0.001

wards (adj. OR 4.2; p = 0.03 and adj. OR 4.0; p = 0.03,respectively).

Multivariate Associations with Pressure UlcerPreventive Quality Management at Ward LevelThe different types of organizational cultures and teamclimate were not associated with preventive quality man-agement at ward level (Table 3). However, institutionalquality management was strongly associated with preven-tive quality management at ward level (p < 0.001) andaccounted for 17.2% of the 23.1% explained variance.

DISCUSSION

Although organizational culture, team climate, and qual-ity management have all been argued to potentially con-tribute to the improvement of patient care, the present re-sults show only the quality of pressure ulcer managementat institutional level to strongly contribute to preventivequality management at ward level. However, contrary toour expectations, a significant link to the prevalence ofnosocomial pressure ulcers was not detected.

Strengths and Weaknesses of the Study andComparison with Other StudiesThere are several possible explanations for the absence ofassociations with the outcome variables in this study. First,the relations between the organizational factors and out-comes may be much more complex than we suggested. It ispossible that other aspects of an institution or team medi-ate or moderate these relationships, such as the perceivedlevel of external accountability, the way a team learnsfrom mistakes, team characteristics such as size, compo-

sition and workload, and human recourse management.To identify important issues, it is important to go beyondthe cross-sectional quantitative studies. Further researchmight benefit from combining qualitative and quantitativemethods, and also study the interrelations between differ-ent factors in the system. For instance, a recent review onlinkages between organizational factors, medical errors,and patient safety (Hoff et al. 2004) suggests that stud-ies should take a “systems perspective” (Shortell & Singer2008). In addition to a variable such as “teamwork,” alarger, more interconnected web of organizational dynam-ics should be considered in the analyses based on severaldifferent theoretical models. However, the multilevel de-sign of this type of quantitative studies implies that onecan study only a limited number of factors at cluster level.So, other designs such as qualitative studies are neededto study a broad range of interconnected factors. A recentreview on teamwork within the psychological literaturesuggested that since teamwork is dynamic, its manifesta-tion can vary based on a vast number of variables, suchas team environment, type of task, individual difference,and perceived workload (Salas et al. 2005). Therefore, tofully understand such a construct, the authors argued thatit is insufficient to take a single snapshot of team perfor-mance. Instead, it should be sampled during a variety ofconditions and situations, including both laboratory andapplied research settings (Salas et al. 2005).

Also, we may have found relationships if we had selectedoutcomes that are more closely linked to culture and cli-mate such as job satisfaction of caregivers. Although theselection of a clinical outcome is important to distinguishbetween process and outcome variables, one might runthe risk with the selection of such a specific variable ofbecoming too narrow to reflect the complexity of modernpatient care (Horn 2006). Also, the prevention of pressureulcers requires an ongoing, continuous care delivery pro-cess by several caregivers, compliant with guideline rec-ommendations. It is likely that there is a considerable gapbetween what is recorded in medical records or in policydocuments and protocols, and what is actually done inpractice (Bates-Jensen et al. 2003b). Although, for healthcare management and policy makers, it is still very im-portant to know whether preventive quality managementat ward level indeed leads to better safety outcomes forpatients, it would have been informative if we also hadthe information regarding the actual preventive care pro-vided to each patient. Therefore, the results of this studywould benefit from additional research using preferablyother methods such as practice observations to register theactual implementation of recommendations at patient level(Bates-Jensen et al. 2003a) in addition to our sum score ofpreventive quality management at ward level.



A second possible explanation may be that the measure-ment instruments were unsuitable for finding the relation-ships we explored. We selected the instruments we usedin this study because they had been validated previously,and had been used in numerous other studies. By usingthe same instruments, one allows for better comparisonbetween the different studies, which is important to gaininsight into the relevance of these constructs. A review(Scott et al. 2003a) on the association between organiza-tional culture and health care performance concluded thatin 4 out of 10 studies that used the CVF, significant asso-ciations were found. However, these studies were mostlyconcerned with performance as perceived by managers.The observed associations were also found for only thoseaspects of performance, which were clearly valued by thedominant culture. Since the prevention of pressure ulcersrelies on a concerted action between different caregiverswith clear responsibilities, one could argue that the more“social” cultures, which tend to enhance team function-ing and creativity (the flexible group and developmentaltypes) show positive associations with preventive effortsand the prevalence of pressure ulcers. However, prevent-ing ulcers also requires a strict adherence to the evidence-based protocol. Therefore, on the other hand, one couldalso argue that the more control-orientated rational or hi-erarchical culture types with a focus on policies, proce-dures, and production might be just as useful in attainingsafety in patient care. However, our findings did not sub-scribe to either of these ideas, although an earlier studyon the relations between the CVF, the extent of qualityimprovement, and pressure ulcers in nursing homes didfind that quality improvement implementation was greaterin those nursing homes with an organizational culturethat emphasizes innovation and teamwork (Berlowitz et al.2003). However, like in our study, no significant associa-tion was found between quality improvement implemen-tation and either adherence to guideline recommendationsas reported in patient records or the rate of pressure ulcerdevelopment.

We also failed to find significant associations betweenteamwork and the outcomes. Previous studies that usedthe TCI explored associations between team climate andperformance showed mixed results. For instance, a studyin primary care identified team climate as a key part of pro-viding high-quality care across a range of areas (Campbellet al. 2001). Gibbon et al. used the TCI to examine whetherteam-coordinated approaches could improve staff attitudesto team working in stroke care. They concluded that at-titudes did not significantly improve; teams appeared totake a long time to establish cohesion and develop sharedvalues (Gibbon et al. 2002). The absence of associations

in our study thus contributes to the debate concerning themeaning of the concepts of organizational culture and teamclimate (Scott et al. 2003b), and the measurement of thesecomplex constructs (Colla et al. 2005). For instance, sinceclimate and culture are considered to be shared attributes,individual measures are aggregated to ward level. Yet, thisignores the fact that different subgroups (e.g., nurses anddoctors within a team) may have different opinions and re-duces variation that may exist between individuals (Scottet al. 2003b). Other studies have raised concerns about thesuitability of using survey-based scales to show the link be-tween such complex factors and clinical patient outcomes(Colla et al. 2005). Future research would probably bene-fit from qualitative research methods and the explorationof instruments that are more sensitive to measure thesecomplex constructs.

Third, other methodological limitations could haveplayed a role. For instance, in order to minimize the bur-den on the participating teams, the number of respondentsfor the organizational culture and team climate question-naires was small (between 3 and 5) while the number ofrespondents for a team or group variable ideally should beas high as possible. The Cronbach’s alphas for the orga-nizational culture subscales, for example, were quite low,which shows low internal consistency for the subscales,which may have diluted potential relations. Also, responsebias may have influenced the results. For instance, weasked the ward managers to distribute the questionnairesamong members of the same patient care team at theirward. It is possible that respondents who were asked tocomplete the questionnaires could have been more in-clined to assign higher scores on team climate as com-pared to others who were not asked to participate in thestudy. However, scores on team climate were in line withprevious studies (Dackert & Brenner 2002; Loo 2003).Mean culture scores tend to vary considerably betweenseveral studies so it is hard to compare our scores to scoresof studies conducted previously. Finally, response rateswere quite low, and therefore we may have lacked thepower to find associations. The wide confidence inter-vals in Table 3 for the categorical dominant culture vari-able show that this point may, at least partly, limit ourresults.

CONCLUSIONS

Despite the possible limitations on this study, to ourknowledge, it is the first Dutch study in hospitals andnursing homes to explore the associations between orga-nizational culture, teamwork, and quality management foran important patient safety issue measured in an objective



manner. The absence of significant associations for our fac-tors of interest and the prevalence of nosocomial pressureulcers contributes to the ongoing discussion of the poten-tial of these constructs to improve patient care. Additionalresearch is needed to determine if, under what conditions,and exactly how these variables can be meaningfully mea-sured and to learn about their potential contribution tosafer and higher quality patient care. Mixed methods aswell as methods to generate knowledge within a context ofapplication may contribute to future knowledge develop-ing.

ReferencesAnderson N.R. & West M.A. (1994). The Team Climate

Inventory. Windsor, UK: NFER-Nelson.Bates-Jensen B.M., Cadogan M., Jorge J. & Schnelle J.E.

(2003a). Standardized quality-assessment ulcer care inthe nursing home. Journal of the American GeriatricsSociety, 51(9), 1195–1202.

Bates-Jensen B.M., Cadogan M., Osterweil D., Levy-StormsL., Jorge J., Al-Samarrai N., Grbic V. & Schnelle J.F.(2003b). The minimum data set pressure ulcer indica-tor: Does it reflect differences in care processes relatedto pressure ulcer prevention and treatment in nursinghomes? Journal of the American Geriatrics Society, 51(9),1203–1212.

Beaton D.E., Bombardier C., Guillemin F. & Ferraz M.B.(2000). Guidelines for the process of cross-culturaladaptation of self-report measures. Spine, 25(24), 3186–3191.

Bennett G., Dealey C. & Posnett J. (2004). The cost ofpressure ulcers in the UK. Age and Ageing, 33(3), 230–235.

Bergstrom N., Braden B., Kemp M., Champagne M. & RubyE. (1996). Multi-site study of incidence of pressureulcers and the relationship between risk level, demo-graphic characteristics, diagnoses, and prescription ofpreventive interventions. Journal of the American Geri-atrics Society, 44(1), 22–30.

Bergstrom N., Braden B., Kemp M., Champagne M. & RubyE. (1998). Predicting pressure ulcer risk—A multisitestudy of the predictive validity of the Braden Scale. Nurs-ing Research, 47(5), 261–269.

Bergstrom N., Demuth P.J. & Braden B.J. (1987). Aclinical-trial of the Braden scale for predicting pressuresore risk. Nursing Clinics of North America, 22(2), 417–428.

Berlowitz D.R., Young G.J., Hickey E.C., Saliba D.,Mittman B.S., Czarnowski E., Simon B., Anderson J.J.,Ash A.S., Rubenstein L.V. & Moskowitz M.A. (2003).Quality improvement implementation in the nursing

home 12. Health Services Research, 38(1 Pt 1), 65–83.

Bours G.J.J.W., Halfens R.J.G., Berger M.P.F., Abu SaadH.H. & Grol R.T.P.M. (2004). Development of a modelfor case mix adjustment of pressure ulcer prevalencerates. Medical Care, 41(1), 45–55.

Bours G.J.J.W., Halfens R.J.G, Huijer Abu-Saad, H. & GrolR.T.P.M. (2002). Prevalence, prevention, and treatmentof pressure ulcers: Descriptive study in 89 institutionsin the Netherlands. Research in Nursing & Health, 25,99–110.

Bower P., Campbell S., Bojke C. & Sibbald B. (2003). Teamstructure, team climate and the quality of care in primarycare: An observational study. Quality and Safety in HealthCare, 12, 273–279.

Campbell S.M., Hann M., Hacker J., Burns C., Oliver D.,Thapar A., Mead N., Safran D.G. & Roland M.O. (2001).Identifying predictors of high quality care in Englishgeneral practice: Observational study. British MedicalJournal, 323(7316), 784–787.

The Dutch Institute for Healthcare Improvement (CBO).(2002). Richtlijn decubitus. Utrecht, The Netherlands:Author.

Colla J.B., Bracken A.C., Kinney L.M. & Weeks W.B.(2005). Measuring patient safety climate: A review ofsurveys. Quality & Safety in Health Care, 14(5), 364–366.

Dackert I. & Brenner S. (2002). Team climate inventorywith merged organization. Psychological Reports, 91,651–656.

European Pressure Ulcer Advisory Panel (EPUAP).(1998). Pressure ulcer prevention guidelines. Oxford, UK:Author.

European Pressure Ulcer Advisory Panel (EPUAP). (2008).Pressure ulcer prevention guidelines. Oxford, UK: Author.

Ferlie E.B. & Shortell S.M. (2001). Improving the quality ofhealth care in the United Kingdom and the United States:A framework for change. Milbank Quarterly, 79(2), 281–315.

Firth-Cozens J. (1998). Celebrating teamwork. Quality inHealth Care, 7(Suppl.), S3–S7.

Friedman D.M. & Berger D.L. (2004). Improving teamstructure and communication—A key to hospital effi-ciency. Archives of Surgery, 139, 1194–1198.

Gerowitz M.B. (1998). Do TQM interventions change man-agement culture? Findings and applications. QualityManagement in Health Care, 6(3), 1–11.

Gibbon B., Watkins C., Barer D., Waters K., Davies S.,Lightbody L. & Leathley M. (2002). Can staff attitudesto team working in stroke care be improved? Journal ofAdvanced Nursing, 40(1), 105–111.



Halfens R.J.G., Janssen M.A.P. & Meijers J.M.M. (2006).Rapportage resultaten: Landelijke prevalentiemeting Zorg-problemen. Maastricht, The Netherlands: UniversiteitMaastricht.

Halfens R.J.G., Janssen M.A.P., Meijers J.M.M. &Mistiaen P. (2005). Rapportage resultaten LandelijkePrevalentiemeting Zorgproblemen 2005. Maastricht, TheNetherlands: Universiteit Maastricht, sectie Verpleg-ingswetenschap.

Haward R., Amir Z., Borrill C., Dawson J., Scully J., WestM. & Sainsbury R. (2003). Breast cancer teams: The im-pact of constitution, new cancer workload, and meth-ods of operation on their effectiveness. British Journal ofCancer, 89(1), 15–22.

Hoff T., Jameson L., Hannan E. & Flink E. (2004). A reviewof the literature examining linkages between organiza-tional factors, medical errors, and patient safety. MedicalCare Research and Review, 61(1), 3–37.

Horn S.D. (2006). Performance measures and clinicaloutcomes. Journal of the American Medical Association,296(22), 2731–2732.

Institute of Medicine. (2001). Crossing the quality chasm: Anew health care system for the 21st century. Washington,DC: National Academy Press.

Kivimaki M., Sutinen R., Elovainio M., Vahtera J.,Rasanen K., Tovry S., Ferrie J.E. & Firth-CozensJ. (2001). Sickness absence in hospital physicians:2 year follow up on determinants. Journal of Oc-cupational and Environmental Medicine, 58, 361–366.

Loo R. (2003). Assessing “team climate” in project teams.International Journal of Project Management, 21, 511–517.

Loo R. & Loewen P. (2002). A confirmatory factor-analyticand psychometric examination of the team climateinventory—Full and short versions. Small Group Re-search, 33(2), 254–265.

Lyder C.H. (2001). Quality of care for hospitalized Medi-care patients at risk for pressure ulcers. Archives of In-ternal Medicine, 161, 1549–1554.

Lyder C.H. (2003). Pressure ulcer prevention and man-agement. Journal of the American Medical Association,289(2), 223–226.

Meijers J.M.M., Halfens R.J.G., Van Bokhorst-de Van DerSchueren M.A.E., Dassen T. & Schols J.M.G.A. (2009).Malnutrition in Dutch healthcare: Prevalence, preven-tion, treatment and quality indicators. Nutrition, 25(5),512–519.

Meterko M., Mohr D.C. & Young G.J. (2004). Teamworkculture and patient satisfaction in hospitals. MedicalCare, 4(5), 492–498.

Moss F., Garside P. & Dawson S. (1998). Organizationalchange: The key to quality improvement. Quality inHealth Care, 7(Suppl.), S1–S2.

Quinn R.E. & Kimberly J.R. (1984). Paradox, planning,and perseverance: Guidelines for managerial practice.In J.R. Kimberly & R.E. Quinn (Eds.), Managing orga-nization transitions (pp. 295–313). Homewood, IL: DowJones-Irwin.

Russo C.A., Steiner C. & Spector W. (2008, De-cember). Hospitalizations related to pressure ulcersamong adults 18 years and older, 2006. HCUP Sta-tistical Brief no. 64. Rockville, MD: Agency forHealthcare Research and Quality. Retrieved March17, 2010, from http://www.hcup-us.ahrq.gov/reports/statbriefs/sb64.pdf.

Salas E., Sims D.E. & Burke C.S. (2005). Is there a “big five”in teamwork? Small Group Research, 36(5), 555–599.

Scott T., Mannion R., Marshall M. & Davies H.(2003a). Does organisational culture influence healthcare performance? A review of the evidence. Jour-nal of Health Services Research & Policy, 8(2), 105–117.

Scott T., Mannion R., Davies H.T. & Marshall M.N.(2003b). Implementing culture change in health care:Theory and practice. International Journal for Quality inHealth Care, 15(2), 111–118.

Severens J.L., Habraken J.M., Duivenvoorden S. & Fred-eriks C.M. (2002). The cost of illness of pressure ulcersin The Netherlands. Advances in Skin and Wound Care,15(2), 72–77.

Shortell S.M., Jones R.H., Rademaker A.W., Gillies R.R.,Dranove D.S., Hughes E.F., Budetti P.P., Reynolds K.S.& Huang C.F. (2000). Assessing the impact of total qual-ity management and organizational culture on multipleoutcomes of care for coronary artery bypass graft surgerypatients. Medical Care, 38(2), 207–217.

Shortell S.M., O’Brien J.L., Carman J.M., Foster R.W.,Hughes E.F., Boerstler H. & O’Connor E.J. (1995).Assessing the impact of continuous quality improve-ment/total quality management: Concept versus im-plementation. Health Services Research, 30(2), 377–401.

Shortell S.M. & Singer S.J. (2008). Improving patient safetyby taking systems seriously. Journal of the American Med-ical Association, 299(4), 445–447.

Shortell S.M., Zazzali J.L., Burns L.R., Alexander J.A.,Gillies R.R., Budetti P.P., Waters T.M. & ZuckermanH.S. (2001). Implementing evidence-based medicine:The role of market pressures, compensation incentives,and culture in physician organizations. Medical Care,39(7 Suppl. 1), I62–I78.



Solberg L.I. (2000). Guideline implementation: What theliterature doesn’t tell us. Joint Commission Journal ofQuality Improvement, 26(9), 525–537.

Strasser D.C., Smits S.J., Falconer J.A., Herrin J.S. & BowenS.E. (2002). The influence of hospital culture on re-habilitation team functioning in VA hospitals. Journalof Rehabilitation Research and Development, 39(1), 115–125.

Thomson J.S. & Brooks R.G. (1999). The economicsof preventing and treating pressure ulcers: A pi-lot study. Journal of Wound Care, 8(6), 312–316.

Wheelan S., Burchill C. & Tilin F. (2003). The link be-tween teamwork and patients’ outcomes in intensivecare units. American Journal of Critical Care, 12, 527–534.


Documents

OriginalArticle Organizational Culture, Team Climate, and