lable at ScienceDirect

Social Science & Medicine 101 (2014) 107e112

Contents lists avai

Social Science & Medicine

journal homepage: www.elsevier .com/locate/socscimed

Short report

Teaching hospital performance: Towards a community of sharedvalues?

Marianna Mauro a,*, Emma Cardamone b, Giusy Cavallaro b, Etienne Minvielle c,Francesco Rania b, Claude Sicotte d, Annarita Trotta b

aDepartment of Clinical and Experimental Medicine, Magna Græcia University of Catanzaro, Campus of Germaneto, Viale Europa, 88100, Catanzaro, ItalybDepartment of Legal, Historical, Economic and Social Sciences, Magna Græcia University of Catanzaro, Campus of Germaneto, Viale Europa, 88100,Catanzaro, ItalycCentre for Medical Research, Medicine, Health and Society (INSERM U750/CNRS 8169), Institut Gustave Roussy, 114 rue Edouard Vaillant, 94895,Villejuif Cedex, FrancedDepartment of Health Administration, University of Montreal, CP 6128 Succursale Centreville, Montreal, Quebec, H3C 3J7, Canada

a r t i c l e i n f o

Article history:Available online 22 November 2013

Keywords:Hospital managementPerformance measurementItalyTeaching hospitalHospital stakeholdersShared values

* Corresponding author.E-mail address: mauro@unicz.it (M. Mauro).

0277-9536/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.socscimed.2013.11.027

a b s t r a c t

This paper explores the performance dimensions of Italian teaching hospitals (THs) by considering themultiple constituent model approach, using measures that are subjective and based on individual idealsand preferences. Our research replicates a study of a French TH and deepens it by adjusting it to thecontext of an Italian TH. The purposes of this research were as follows: to identify emerging views on theperformance of teaching hospitals and to analyze how these views vary among hospital stakeholders. Weconducted an in-depth case study of a TH using a quantitative survey method. The survey uses aquestionnaire based on Parsons’ social system action theory, which embraces the major models oforganizational performance and covers three groups of internal stakeholders: physicians, caregivers andadministrative staff. The questionnaires were distributed between April and September 2011. The resultsconfirm that hospital performance is multifaceted and includes the dimensions of efficiency, effective-ness and quality of care, as well as organizational and human features. There is a high degree ofconsensus among all observed stakeholder groups about these values, and a shared view of performanceis emerging. Our research provides useful information for defining management priorities to improve theperformance of THs.

� 2013 Elsevier Ltd. All rights reserved.

Introduction

Research on performance measurement has attracted signifi-cant interest from both practitioners and researchers, particularlywithin the healthcare industry (van der Geer, van Tuijl, & Rutte,2009). Recent healthcare system reforms, seeking to increaseaccountability, cost effectiveness, sustainability and quality of care,as well as to update strategies in several countries, interest manystakeholders, including health professionals, patients and thecommunities that surround hospitals (Alonzo & Simon, 2008).

Stakeholders have reacted to the redefinition of hospital per-formance towards a more market-oriented view of health services(Koelewijn, Ehrenhard, Groen, & van Harten, 2012). Many hospitalperformance models have been developed on the basis of this newanalytical perspective (Bravi et al., 2013; Klassen et al., 2010). In

All rights reserved.

particular, “High-PerformanceWork Systems” (HPWSs) are capableof improving outcomes in healthcare (McAlearney et al., 2011).However, few researchers have investigated these frameworks inthe context of organization theory, in which organizations arebattlegrounds for stakeholders who seek to influence the criteriafor effectiveness to advance their own differing interests (Guisset,Sicotte, Leclercq, & D’Hoore, 2002). According to the multipleconstituentmodel of organizational performance, the stakeholders’perspectives must be considered in an integrated way for thepurposes of evaluation. Thus, as shown in Table 1, it is possible torefer to Parsons’ social system action theory (Parsons, 1951) tocombine four major frameworks for the evaluation of organiza-tional performance (Rationale, Human relations, Open systems andInternal processes) into a single comprehensive, theoreticallygrounded model that embodies the currently fragmented ap-proaches to performance management in healthcare organizations(HCOs) (Sicotte et al., 1998).

Using the model in Table 1, which was applied successfully tothe French Bicêtre TH (Minvielle et al., 2008), this paper explores

Table 1Analytical framework for a multiple constituent model of the organizational per-formance of HCOs: definition of questionnaire dimensions.

Dimensions Definition Sub-dimensions

Rationale This dimension focuseson the effective realizationof an organization’s objectives.

Goal attainmentEfficacy/EffectivenessPatient satisfaction

Open systems(Relation withenvironment)

This dimension evaluatesan organization’s capacityto acquire resources and itsrelationship with itsenvironment.

Cost control/EfficiencyAttractiveness/Capacityto acquire resourcesOpenness

Internal processes This dimension classifiesan organization as high-performing when it conductsits operations without problemsor undue internal strains.

Productivity Internalorganization

Human relations(shared values/organizationalclimate)

Performance is defined in termsof an organization’s internalhealth, which includes sharedvalues and the organizationalclimate.

Public service valuesProfessional valuesOrganizational valuesWork climatePersonal achievementvalues

Source: Minvielle et al., 2008

M. Mauro et al. / Social Science & Medicine 101 (2014) 107e112108

these performance dimensions in the Mater Domini teaching hos-pital (MD TH) of Catanzaro, located in the Calabria region (insouthern Italy): one of the eight Italian regions involved in a re-covery plan aiming to restore economic balance. Within thehealthcare industry, THs that combine clinical care, teaching andresearch play an important economic role in the attainment ofeffectiveness as a social tool. The international literature highlightsfactors that threaten the survival of THs worldwide such as theirgovernance complexity, coordination issues and current lack ofefficiency (Liu, Forgione, & Younis, 2012).

Traditional performance measurement systems that focus onlyon accounting and financial measures appear to be inadequate.Each stakeholder group has different preferences, purposes andvalues. Thus, a multidimensional system of performance mea-surement is appropriate for teaching hospitals.

Italy represents a remarkable opportunity for the study of per-formance measurements for THs: each Italian TH is an interestingcase study because it has specific characteristics in terms of costs,revenues and activities, and although performance managementhas increased in the Italian national health service (INHS) over thelast twenty years, it can still be considered a work in progress,particularly in southern Italy, where there is an urgent need for awide-scale evaluation of the system.

This research aims to contribute to the debate about hospitalperformance in organizational theory from the stakeholders’ per-spectives. The purposes of our research were as follows: 1) todescribe emerging views on hospital performance and 2) to analyzehow such views vary among selected groups of stakeholders,including a comparison of the Italian case with the French one. Weconducted an in-depth explorative case study of a TH using aquantitative survey method.

The originality of this study lies in its application of amodel that,adjusted to the Italian context, allows for the assessment of THperformance from the perspective of internal stakeholders, indi-cating keymanagement priorities in financial distress contexts. Themodel can also analyze important performance dimensions that aresuggested by the HPWSs literature but usually neglected.

Methods

Our tool for data collection was a questionnaire that was elab-orated on the basis of a theoretical framework developed by Sicotte

et al. (1998). Consistent with the objectives of our study, thequestionnaire, compared with qualitative interviews, ensures theanonymity of respondents and guarantees the absence of any dis-tortions originating from the interviewer. High response rates andsystematic controls on questionnaire compilation, as well as shortand unambiguous items, were ensured. Our questionnaire isorganized into 3 levels: the 4 main dimensions are divided into 13sub-dimensions that were measured by 66 individual items. Eachitem, addressing a particular aspect of the hospital’s performance,was measured on a Likert scale of 10 levels of accord/discord. Anindividual responder used a discrete numerical scale of values from0 (total discord 5 insignificant item) to 10 (absolute accord 5

main item) to score each response.To adapt the questionnaire to the Italian context, we inserted

questions about the responders: their seniority, affiliation withfixed operating units (OUs) and responsibilities. We adjusted theitems in the Human relations dimension. First, the items Staff pre-serves patient dignity and Staff preserves patient confidentiality werecombined in accordance with Italian laws that protect the dignityand privacy of the patient (art. 1 of Law n. 833/1978 and Decree n.196/2003). Second, we eliminated the item Staff management dif-fuses information about the hospital’s history because the TH selectedfor our study (MD TH) was established in 1995. We replaced thisitem with Staff management listens to and discusses personnel’ssuggestions to emphasize the working team. Third, in accordancewith the national law on continuous staff training (Decree n. 502/92 and subsequent additions), we added the item Invests in thevocational training of personnel. Such adjustments improved thequestionnaire’s reliability (Cronbach’s alpha > 0.975) and permit usto investigate Human relations in-depth, as suggested in the Frenchstudy. The items were presented in random order to avoid condi-tioning bias. This survey was approved by the hospital ethics re-view board prior to initiation.

Population and sample

Our research methodology is based on a survey of a populationcomposed of three internal stakeholder groups (physicians, care-givers and administrative staff) at one Italian TH. Beginning withthe existing studies (Blake, Massey, Bala, Cummings, & Zotos, 2010),we identified the core internal stakeholders that affect the processof care in Italian THs: this is also justified considering that the re-sults of the analysis, focussing on inter-stakeholder relationships,are useful in defining priorities in management choices. This optionensures consistency with the French study.

According to Yin (2003), an exploratory analysis can be based ona single case study to confirm a theory and to represent a uniquesituation providing in-depth analysis and multiple sources of in-formation. However, limitations related to the findings’ general-ization can be identified, depending on the peculiar characteristicsof the organizational and national socio-economic contextinvestigated.

We focused on the MD TH because of its specific character-istics and its regional context. The MD hospital effectively rep-resents the context of southern Italy because it is the only THlocated in Calabria, which is recognized internationally as anunderdeveloped area. The financial constraints and the delayassociated with the professional management of the healthsystem influence health services supply in Calabria. Spending onhealth services contributed to the region’s budget deficit andresulted in the signing of the recovery plan, which stipulates thatthe MD TH is involved the Italian government in eliminating thedeficit. These institutional characteristics and local dynamicsmake the MD TH particularly interesting for performance mea-surement studies.

M. Mauro et al. / Social Science & Medicine 101 (2014) 107e112 109

In 2011, the MD had 31 OUs and 162 beds and discharged 9329patients. A 2-year observation period established that the MD THpopulation in 2011 was n ¼ 633, which included 95 administrativestaff, 297 physicians and 241 caregivers. We considered a randomsample of 300 employees of the total personnel in 2011 and ratedeach employee in terms of the 3 stakeholder classes. The non-responders composed only 9% of the sample. This loss of informa-tion prevented the justification of a reasonable analysis of possiblesystematic errors in the findings because the sample representedapproximately 50% of the total population.

The data set

The questionnaire was distributed between April andSeptember 2011. A questionnaire was considered valid if it con-tained nomore than 5 unanswered items, and the total score was inthe inter-5-percentile range. Of the 300 questionnaires distributed,273 (91%) were returned and 217 (72%) met the validity criteria.These 217 questionnaires were completed by 38 administrativestaff, 111 physicians and 68 caregivers, mainly nurses (Table 2).

Regarding personnel professionalism, physicians were dividedequally between basic skills (52.3%) for non-executives andadvanced or pro skills (47.8%) for management categories. Thecaregivers and the administrative staff had high percentages ofbasic skills: 88.2% and 97.4%, respectively. We conducted our ana-lyses using the SPSS software ver. 19.0.0.

From each questionnaire, we obtained a sub-questionnaire of 13pseudo items unifying all items of each sub-dimension and a sub-questionnaire of 4 pseudo items unifying all items of each dimen-sion. All questionnaires and sub-questionnaires were subjected toseveral methods of computing internal consistency reliability(Cronbach’s alpha, Split-half, SpearmaneBrown), factors for factoranalysis (recoding variables, Kaiser Meyer Olkin Test, R-factor) andpredictors of the total mean score (ANOVA). The alpha (0.987 forthe 66 items, 0.984 for the 13 pseudo items, 0.975 for the 4 pseudoitems) indicates a scale with reasonable internal consistency reli-ability. The SpearmaneBrown coefficient (0.968, Split-half method)indicates an excellent balance. The QeQ Plot, KolmogornoveSmirnov and ShapiroeWilk tests permitted verification of theappropriateness of the variables; the KaisereMeyereOlkin Test(0.840) established whether the correlation matrix could be fac-torized; the ANOVA explained the total mean score for 4dimensions.

Results

We obtained the following results for the four dimensions. TheOpen systems was the dimension with the lowest mean (7.4879)and the highest standard deviation (1.63224). For the Human re-lations dimension, the statistics indicated the opposite relationship(mean 8.2379, standard deviation 1.58637).

All of the results confirmed that the mean distributions of thefour dimensions were not normal and that the data were clusteredaround higher values and concentrated at the middles of the

Table 2Response to the survey questionnaire.

Groups N Questionnaires

Sent Returned Invalid Valid Responserate

(a) (b)

Administrative staff 95 45 41 1 2 38 84%Caregivers 241 105 92 7 18 68 65%Physicians 297 150 140 2 26 111 74%Total 633 300 273 10 46 217 72%

distributions. Thus, we concentrated on the influence of eachdimension on the total mean score (8.0087 � 1.53766). Althougheach dimension, which was considered a block of all associateditems, had a different weight (Rationale, 24.2%; Open systems, 21.2%;Internal processes, 21.2%; Human relations, 33%), the dimensionsinfluenced only 3.3% of the total mean score because the analysis ofvariance (ANOVA) demonstrated that the remaining principalpercentage was due to the internal data within each dimension. Inour case, the Human relations dimension (8.2379 � 1.58637) rep-resented the maximum influence and confirmed the marker ofanalysis. Ordering the confidence intervals (CIs) of the mean scoresof each sub-dimension, we identified five categories (Fig. 1, Table 3)around the CI of the total mean score [7.8; 8.2].

Important differences between our study and the French onecan be highlighted. First, our results for work climate (CI [7.85;8.31]) were near the middle of the range for the 13 sub-dimensionsand confirmed that on the questionnaire, this element (7/66 itemsvs. 6/66 items) represented the general preferences of all stake-holders. Although the ANOVA revealed significant differencesamong administrative staff, caregivers and physicians, whenapplied to caregivers and physicians, the same test produced pos-itive results, excluding all of the different professions. Second, ourresults for the goal attainment sub-dimension (CI [7.66; 8.09]) werethe fourth highest in region 2, indicating the particular corporateinterests of all stakeholders. Finally, the classification of the orga-nizational values (CI [8.02; 8.47]) in region 5 supported the domi-nance of Human relations.

Considering all of the items, 84.7% of the respondents’ prefer-ences were included in the value, ranging from 7 to 10. Aftercalculating the mean value of each item, we found that the fiveitems with the highest mean scores belonged to Human relations(Staff preserves patient dignity, 8.76; Staff is empathetic to patients,8.58; Invests in personnel training, 8.55), with the exception of twoRationale items (Assesses the impact of the services/care it provides,8.57; Sets up goals and strives to reach them, 8.55). In contrast, thefive items with the lowest mean scores belonged to Open systems(Care unit managers are widely renowned, 6.37; Attracts the mostrenowned hospital directors, 6.48; Develops strong ties with thecommunity (associations, cultural centres), 7.04;Well regarded by themedia, 7.11; Large job supplier, 7.16). Although all of these resultsunderscore the importance of the humanitarian spirit that is typicalof the medical and sanitary deontology and the effects of corporatevisions on teamwork, two particular findings should be noted: thehigh scores that were attributed to the Human relations andRationale dimensions. The question on investments in personneltraining, which was not used on the French questionnaire, showsthat stakeholders (Admin. Staff, 7; Caregivers, 8.94; Physicians, 8.85)believe that a hospital is performing if it invests in competency. Thisitem is also included in personal achievement, which is ranked in thetop five sub-dimensions (Table 3). The emphasis on Rationale couldbe explained by the financial constraints of a regional health systemthat required greater expenditure rationalization. The items withthe lowest mean scores belong to the Open systems. A possible in-crease in workload, the presence of important people on the staffand the relationship of the hospital to other communities were notconsidered important to achieving this objective.

Finally, the administrative staff members appeared to be lesssensitive than the caregivers and physicians to the performancemeasurement topic. The organizational TH complexities, strongindividualism and a lack of professionalism among administrativestaff produce a poor sense of belonging to the hospital. Additionally,the distribution of economic incentives is based on an evaluationsystem that does not consider the contribution of the individual tooverall performance. Both physicians and caregivers provided amaximum mean score for the item Staff preserves patient dignity

Fig. 1. Ranking of the 13 sub-dimensions into five groups according to the confidence intervals of the mean scores.

M. Mauro et al. / Social Science & Medicine 101 (2014) 107e112110

and confidentiality, whereas administrative staff provided amaximummean score for the item Staff is empathetic to patients. Allgroups provided the lowest scores for the item Care unit managersare widely renowned. Administrative staff and physicians adoptedopposing positions on the item Staff is devoted, which caregiversplaced in the middle of the range, a unique, distinctive result at thislevel. For 13 sub-dimensions, the physicians and the caregivers hadsimilar preferences (Dx ¼ 0.3 � 0.161), and the administrative staffconfirmed the results of the caregivers with a mean score of lessthan 1.31 � 0.225. Every group clearly preferred the Human re-lations dimension (Admin. Staff, 6.98; Caregivers, 8.40; Physicians,8.69); the Rationale dimension (Admin. Staff, 6.92; Caregivers, 8.28;Physicians, 8.66) and the Internal processes dimension (Admin. Staff,6.85; Caregivers, 8.11; Physicians, 8.46) over the Open systemsdimension (Admin. Staff, 6.50; Caregivers, 7.50; Physicians, 8.03).

Discussion

In this study, we first analysed emerging views on hospitalperformance and then examined whether these views differed for

Table 3Ranking of sub-dimensions by case studies.

Sub-dimensions N Mean � Std. dev Ran

Italy France Italy France Ital

Adm

Personal achievement values 217 402 8.35 � 1.67 8.50 � 1.30 5Efficacy/Effectivness 217 402 8 34 � 1.74 8.43 � 1.13 6Public service values 217 402 8.32 � 1.65 3.43 � 1.11 1Patient satisfaction 217 402 8.28 � 1.62 8.27�1.09 2Cost control/Efficiency 217 402 8.25 � 1.80 8.04 � 1.38 7Professional values 217 402 8.25 � 1.65 8.51 � 1.09 4Organizational values 217 402 8.25 � 1.65 7.84 � 1.58 3Internal organization 217 402 8.17 � 1.71 8.25 � 1.05 9Work climate 217 402 8.08 � 1.77 8.38 � 1.41 10Goal attainment 217 402 7.87 � 1.60 6.75 � 1.84 11Productivity 217 402 7.83 � 1.56 7.20 � 1.30 8Openness 217 402 7.60 � 1.63 7.40 � 1.36 12Attractiveness/Capacity to

acquire resources217 402 740 � 1.72 6.77 � 1.51 13

three groups of internal stakeholders (caregivers, physicians,administrative staff), thereby broadening the results of a surveypreviously conducted in a French context (Minvielle et al., 2008).

Regarding the first objective, our findings confirm that hospitalperformance is a multidimensional concept that extends beyondconventional aspects such as efficacy, effectiveness and quality ofcare. Other important dimensions must be considered, includingorganisational and human values, such as quality of work life.

As in the French case, the three stakeholder groups assigned agreater importance to the Human relations model that is notconsidered by traditional measurement tools. In both the Frenchand Italian cases, regarding the Human relations sub-dimensions,the emphasis is on personal achievement concerning the quality ofworking life as well as public service and professional values thatrelate to shared and organizational ideals. Physicians pay particularattention to public service, whereas caregivers focus on professionalvalues, mainly because HCOs are professional bureaucracies inwhich the connections between individuals and health pro-fessionals are closed. An emphasis was made on staff competencies(personal achievement), particularly systems of evaluation and

king

y France

in. Staff Caregivers Phisicians Admin. Staff Caregivers Phisicians

3 1 1 2 12 3 4 3 59 2 5 5 34 7 6 6 47 4 9 8 71 9 3 1 25 5 8 10 96 8 7 7 68 6 2 4 8

10 10 11 13 1311 11 12 11 1012 12 10 9 1113 13 13 12 12

M. Mauro et al. / Social Science & Medicine 101 (2014) 107e112 111

praise, as well as personnel training, by caregivers in theMD TH andphysicians and administrative staff at Bicêtre TH. Moreover, ethicalissues, the ability to empathize with patients and the pride asso-ciated with belonging to the organization (organizational values)are more important in the Italian case than in the French case, inwhich solidarity among team members and respect for compe-tencies and experiences (work climate) were dominant.

The Rationale model also appears to be relevant. In the Italianstudy, in particular, the five items with the highest mean scoresincluded two elements of Rationale, whereas in the French case,only one item was included. The new shared concern for theRationale aspects could be explained by the significant financialconstraints in Italy. Since the early 1990s, healthcare reforms havebeen transforming the national health service to secure globalspending limits while simultaneously promoting efficiency in caredelivery and enhancing responsiveness to consumers. Although theRationale elements are more significant in the Italian case than inthe French case, a similar perception of performance can beobserved in these two contexts. Indeed, although the focus on costcontrol/efficiency is relevant, unlike in previous research on the MDTH, care efficacy and patient satisfaction are still two of the mostimportant factors in hospital performance, and less attention isalways assigned to the productivity of health services.

However, compared with the French study, we gained someimportant insights. First, we broadened the analysis by adding in-vestment in personnel training to the Human relations issues. Thisitem, with the fourth highest mean score, confirms the importanceof organizational investments in staff training in ensuring thequality of services provided. It is particularly important to guar-antee a variety of business missions for THs and to offer the highestquality of care because academia recognizes them as ‘peaks ofexcellence’ within the healthcare system. Furthermore, to under-stand how work experiences influence stakeholder perspectives,we inserted questions about professional skills; it appeared thatmanagement categories were focused primarily on personalachievement values.

The second purpose of this study was to examine whether andhow the concept of hospital performance varied among stake-holder groups. Two interesting trends emerged: 1) the adminis-trative staff appeared to be less sensitive to performancemeasurement topics in the Italian hospital than in the Frenchhospital and 2) all stakeholder groups were in overall agreementregarding hospital performance, conferring greater importance tothe Human relations and Rationale dimensions and devoting lessattention to the relationships with the environment (Open systems).

The latter concern may also be explained by examining theHealth Care System’s features. Although in Italy, the Health CareSystem is an NHS financed primarily by earmarked nationalcorporate and value-added taxes, healthcare in France is providedby a Statutory Health Insurance system, which is funded primarilyby employer/employee-earmarked income and payroll taxes.Regarding providers, although primary care is paid on a fee-for-service basis (and capitation in Italy or pay for performance inFrance), hospitals (chiefly public entities) are reimbursed primarilythrough case-based payments in both systems. The containment ofhealthcare costs is a key issue in both cases.

In summary, our results confirm most of the findings obtainedwith the French research. Beyond the differences mentioned above,stakeholders’ prevailing attention to the human development andRationale dimensions within organizational performance suggeststhe importance of adopting integrated and balanced approaches toevaluations of hospitals. It appears that modern hospitals aremoving towards a community with more shared values whereperformance is a common concern among all observed stakeholdergroups.

Our study may have additional value for scholars in that itcontributes to the ongoing debate regarding the performance ofTHs from the perspective of stakeholders, broadening knowledgeand offering evidence from a financial distress context: althoughthere is a strong focus on economic balance, Human relations arestill relevant.

Opportunities for managers can be addressed by providing aguide for identifying emerging views on hospital performance andfor obtaining strategic information useful in defining priorities inmanagerial choices in crisis contexts. The consensus of stake-holders regarding specific Human relations topics provides aninnovative way for hospital managers to establish managementpriorities to improve the performance of THs. Indeed, according tothe literature on HPWSs, leaders should focus on ensuring highinvolvement practices that concern staff skills and competences,such as employee training, participation and decision making(personal achievement), to obtain higher levels of people satisfactionand subsequently increase the quality of care; team working andshared ideals (public service, professional values) may also positivelyimpact care efficacy. While still allowing for stakeholder sugges-tions, balanced scorecard development should consider manage-ment priorities, particularly to support ethical relationships withpatients (organizational values).

By focussing attention on the lack of interest among adminis-trative staff (who comprise 97.4% of technical personnel at a basiclevel of professionalism, not in executive positions) in performancetopics, managers can develop specific strategies to increase theirinvolvement and invest in personnel training. This study providesuseful findings for managers relating to performance monitoring interms of prioritizing human resource management systems insupporting HPWSs. When assessing the performance of THs,managers should use appropriate balanced scorecard systemsbased on work climate and employee motivation indicators (e.g.,employee satisfaction scores, incentive plan, number of meetingsfor planning activities) as well as expertise and skills indices (e.g.,training hours per employee, percentage of employees involved indevelopment plans) to consider important dimensions, such as theorganizational and human ideals related to workplace quality. Priorstudies reported links between HPWSs and employee and organi-zational outcomes (e.g., higher satisfaction/engagement, improvedrecruitment, lower turnover) that also show that performancemonitoring focused on HPWSs, providing essential feedback ondevelopmental needs, represent an important strategy forimproving the quality of care and patient safety (McAlearney et al.,2011).

Because the Italian context is characterized by important reor-ganization processes of regional health services, an interest inperformance topics, from the perspective of hospital stakeholders,is important to the achievement of healthcare objectives. Thisimportance is particularly true with regard to THs, which have anurgent need to manage their complex dynamics and inefficiencies(Liu et al., 2012).

In this light, our findings have potential for practice and futureresearch: key internal stakeholders are important vehicles forpromoting performance implementation. Our results show thatmanagers need to focus e more than in the French context e bothon organizational values and on dimensions connected to theeffectiveness, efficiency and quality of care processes. This focus isparticularly relevant in countries e such as Italy ewhere for a longtime, the financing system has not been connected to health goalattainment. Additionally, to strongly promote a culture of perfor-mance evaluation, it is necessary to include administrative staffwith professional skills.

We acknowledge that this study has limitations. Although thequestionnaire was anonymous, there was a risk of biased responses

M. Mauro et al. / Social Science & Medicine 101 (2014) 107e112112

because of the phenomenon of social desirability. The main limi-tation of this statistical analysis was its limited scale and frame-work, which did not permit the inclusion of additionalquestionnaire items that could balance the four sections and inte-grate hospital performance knowledge questions. Future researchshould focus on enlarging the sample of case studies and evaluatingopportunities for comparison with other THs to assess hospitalnetwork performance. Additionally, consideration of the impor-tance conferred to the rational perspective could be useful inintegrating information from internal stakeholders with that of themain external stakeholders, patients. The final result will be theproposal of a framework that, considering the priorities defined bythe key stakeholders e internal and external e could supportmanagers in their choices.

References

Alonzo, A., & Simon, A. (2008). Have stethoscope, will travel: contingent employ-ment among physician health care providers in the United States. Work,Employment and Society, 22, 635e654.

Blake, R. T., Massey, A. P., Bala, H., Cummings, J., & Zotos, A. (2010). Driving health ITimplementation success: insights from The Christ Hospital. Business Horizons,53, 131e138.

Bravi, F., Gibertoni, D., Marcon, A., Sicotte, C., Minvielle, E., Rucci, P., et al. (2013).Hospital network performance: a survey of hospital stakeholders’ perspectives.Health Policy, 109, 150e157.

Guisset, A. L., Sicotte, C., Leclercq, P., & D’Hoore, W. (2002). Defining hospital per-formance: a survey among key stakeholders in hospitals. Sciences Sociales etSanté, 20(2), 65e102.

Klassen, A., Miller, A., Anderson, N., Shen, J., Schiariti, V., & O’ Donnell, M. (2010).Performance measurement and improvement frameworks in health, educationand social services systems: a systematic review. International Journal forQuality in Health Care, 22(1), 44e69.

Koelewijn, W. T., Ehrenhard, M. L., Groen, A. J., & van Harten, W. H. (2012). Intra-organizational dynamics as drivers of entrepreneurship among physicians andmanagers in hospitals of western countries. Social Science & Medicine, 75(5),795e800.

Liu, L. L., Forgione, D., & Younis, M. Z. (2012). A comparative analysis of the CVPstructure of nonprofit teaching and for-profit non-teaching hospitals. Journal ofHealth Care Finance, 39(1), 12e38.

McAlearney, A. S., Garman, A. N., Song, P. H., McHugh, M., Robbins, J., &Harrison, M. I. (2011). High-performance work systems in health care man-agement, part 2: qualitative evidence from five case studies. Health Care Man-agement Review, 36(3), 214e226.

Minvielle, E., Sicotte, C., Champagne, F., Contandriopoulos, A. P., Jeantet, M.,Préaubert, N., et al. (2008). Hospital performance: competing or shared values?Health Policy, 87, 8e19.

Parsons, T. (1951). The social system. New York: Free Press.Sicotte, C., Champagne, F., Contandriopoulos, A. P., Barnsley, J., Béland, F., &

Leggat, S. G. (1998). A conceptual framework for analysis of health care orga-nizations’ performance. Health Services Management Research, 11, 24e48.

van der Geer, E., van Tuijl, H. F., & Rutte, C. G. (2009). Performance management inhealthcare: performance indicator development, task uncertainty, and types ofperformance indicators. Social Science & Medicine, 69(10), 1523e1530.

Yin, R. (2003). Case study research: design and methods (3rd ed.). Beverly Hills: SagePublications.