Researchers all around the globe have not yet come to an end as regards the supposed positive impact of traditional performance management systems in healthcare, and some research has shown that, paradoxically, performance management policies do not always lead to improved hospital performance. Despite the extensive research identifying the “pitfalls” of the NPM reforms around Europe and the unintended consequences for hospital staff and patients, little is known about the mechanisms that caused those negative effects, which essentially creates a research gap worth investigating. This PhD study tries to address this gap and show why do traditional PM Systems in healthcare not always lead to improved performance, by outlining the unintended consequences of the Greek healthcare reform in a public hospital. By conducting empirical research using a case-study, and by adopting a systemic perspective, this research addresses this gap and sheds light on how hospital performance is perceived by stakeholders of a Greek public hospital and what mechanisms drive its dynamic behaviour. Following a systemic approach, the selected case study - which is a real hospital in the Greek Healthcare system - allowed us to investigate the causing mechanisms of the negative consequences of the Greek healthcare reform on the performance of the case hospital. In doing so, we framed our analysis using the Dynamic Performance Management methodology. Recently, researchers have started to see those negative outcomes as “system pitfalls”, occurring from the non-linear interconnection and the dynamic interaction of the different elements and factors that comprise the health system and the healthcare institutions, i.e., their structure, the policies implemented, the behaviour and the decisions of healthcare workers and patients inside this system. The implementation of a systemic performance assessment methodology in Healthcare is sponsored by many recent scholarly contributions in the field (Arnaboldi et al., 2015; Costanza et al., 2014; Bivona, 2010, 2015; Bivona & Cosenz, 2017a, 2017b; Bivona & Noto, 2020; Davahli et al., 2020; Franco-Santos & Otley, 2018; Fryer et al., 2009; Helal, 2016; Renmans et al., 2017; Mwita, 2000; Noto et al., 2020; Vainieri, Ferrè, et al., 2019; Vainieri, Noto, et al., 2020; Wang et al., 2020). Adopting a systemic perspective means taking as a unit of analysis the organisation as a whole, and not one unit or department; acknowledging its internal and external environment and culture in which health care is performed; and considering the concurrent existence of the pitfalls documented as inherent to the structure of the system and the policies implemented. Studies using such a methodology would be necessary in order to address the gap in existing knowledge, as well as to support policy-makers in designing better, more quality-oriented healthcare policies, interventions and reforms in the future. The purpose of this study was to empirically conceptualise a qualitative model of hospital performance as perceived by stakeholders of a Greek public hospital and use the DPM analysis in order to help policymakers in Greece re-design performance management policies and foster hospital performance. We adopted a systemic, participatory, inductive and dynamic approach by combining the Group Model Building and System Dynamics methodologies into the Dynamic Performance Management approach (Bianchi, 2016). Other research traditions identified in our study are the Stakeholders Theory and Participation. All those approaches stand in the constructivist side of the continuum as research approaches, because they all consider realities as subjective, complex and multi-layered, actively shaped by perceptions and opinions of stakeholders (De Gooyert, 2019; Lane & Schwaninger, 2008). Mixed methods were used to facilitate our approach, combining primary qualitative data from two Group Model Building sessions; four open, unstructured preliminary interviews; and seven semi-structured, disconfirmatory interviews; with secondary, qualitative and quantitative data from a scoping literature review and from a critical literature review; as well as from official, open-access, online text-documents and closed-access, internal text-documents of the hospital’s interdepartmental communication. An open call for participation in the research was sent by email to around 70 different hospitals in the cities of Athens and Thessaloniki in Greece, and the gatekeeper was identified. Starting from the gatekeeper, snowball sampling was used to select 10 participants in the case hospital for the Group Model Building (GMB) sessions, including at least one person from each main key-stakeholder category that our extensive stakeholder analysis identified (i.e., managers, doctors, nurses, paramedics and patients), with the purpose of “eliciting” their mental models and “capturing” them in a qualitative system dynamics model (causal loop diagram). Four of the participants were also interviewed before the GMB sessions (face-to-face, one-to-one preliminary interviews). Convenient sampling was used in order to identify seven more public hospital stakeholders from other public hospitals in Greece for the disconfirmatory interviews. The data analysis included a Scoping Review of the International Literature of Performance Management in the Health Sector; a Critical Review of the Literature on the Greek Healthcare Reform; a Stakeholder Analysis; a Narrative Analysis of Preliminary Interviews and Documents; a Qualitative System Dynamics Analysis (Causal Loop Diagram) of the Simplified version of the Conceptual Model of Hospital Performance created during the GMB sessions; and, finally, the Dynamic Performance Management (DPM) analysis. The GMB sessions helped hospital stakeholders gain a better understanding of what hospital performance is in a more systematic way; define it; show its trend (dynamic behaviour) in the hospital during the last decade in a diagram; and conceptualise it as a system, depicted as a qualitative system dynamics model of hospital performance (CLD - Causal Loop Diagram). The two final versions of this CLD Model (i.e., the Conceptual and the Policy Models of Hospital Performance, available in Appendixes 21 and 22 respectively and thoroughly described in terms of the variables and links they contain in Appendix 24) are the main outputs of the GMB sessions, and formed the basis of our analysis and research findings. The Conceptual Model of Hospital Performance is a CLD model that depicts the actual structure of hospital performance and can be used to explain its currently low levels, whereas the Policy Model of Hospital Performance is extended to incorporate the policy structure, i.e., the changes in the system structure which are necessary, according to our participant stakeholders, in order to improve hospital performance. Hospital performance was defined by the participant stakeholders as the provision of patient-centred care to the patient, with safety (for the patients and the staff); responsibility (adherence to protocols, proportions and procedures) and dignity (nice and clean facilities, reduced waiting times and no informal payments). The historical trend of the Hospital performance in the case hospital was also depicted in a diagram over time called Reference Mode (available in Appendix 19). The Reference Mode created and agreed upon by the participants showed that, despite the counterintuitive negative outcomes documented, the level of the overall performance in the case hospital has been slightly increasing after the healthcare reform and is now stabilizing. Our research showed that the Performance Management policies introduced during the Greek healthcare reform had a negative impact on many aspects of hospital performance in general, and in our case hospital in particular. The new policies undeniably contributed to the reduction of hospital spending, but they simultaneously contributed to the deterioration of hospital service quality. Goal-setting, the main PM strategy followed by Greek public hospitals according to Law N4369/16, is until today not properly implemented in the case hospital and managers seem to treat performance objectives as completely separated from performance and quality, and to consider them totally outside of their everyday tasks. Those findings of the preliminary interviews and documents analysis were validated from the findings of the pretests, conducted before the GMB sessions. Four of the goals that were set by the division managers of the case hospital came up during the GMB sessions and were integrated in the CLD model that the participants built: Standardization of the nursing forms of the nursing departments and units; Standardization of clinical procedures; Use of an Information System in the Interdepartmental Communication; and Application of digital signature and electronic document management. We combined our findings from the documents’ analysis with the descriptions of those goals, as set by the division managers, and we informed them with the findings from our DPM instrumental and objective analysis, which allowed us identify the activities and the resources that are needed for the achievement of each of those four goals. In that respect, we found that apart from the “tangible” strategic resources identified by the managers of the case hospital (e.g., financial and human resources) as essential in the achievement of each of those four goals, Management Capacity - which is an intermediate, administrative product of the hospital, built by the public workers - was equally necessary. Out of all the unintended negative outcomes of the Greek healthcare reform documented in the literature, we found the following seven negative outcomes to be present at the case hospital: (1) Low Quality and Safety of Services perceived by health workers and patients; (2) Low Patient Satisfaction; (3) Informal Payments; (4) High Mortality Rates; (5) Numerous Medical Errors; (6) High Nosocomial & Multidrug-Resistant Bacteria Infections Rates; (7) Low adherence to Clinical Guidelines and Treatment Protocols. Regarding those seven negative outcomes, the analysis of the simplified version of the Conceptual Model of Hospital Quality which the participant stakeholders created during the GMB sessions at the case hospital, showed that: (1) Low Quality and Safety are mostly associated with the variables Survival Rate / Patients' Health Status & Quality of Life and Complications of our model, and can be explained by the dominance of the balancing loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause those two variables to decrease as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment (Dynamic Hypothesis 1). (2) Low Patient Satisfaction can be explained by the dominance of the loops B1 – Word of Mouth & Waiting Times, B2 – Patient Satisfaction & Attendance to Patients’ Needs, B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, all of which lead to a gradual decrease and stabilisation of Patient Satisfaction and of Hospital Reputation in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Informal Payments for early Surgery/Admission longer Waiting List for Surgery or Admission, longer Waiting Time in ER & Outpatient Services and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment. (Dynamic Hypothesis 2). (3) The existence of Informal Payments can be explained by the Loop R2 – Informal Payments & Corruption, which leads to a perpetual increase of private spending and to the outspread of corruption between the case hospital doctors, given the good reputation of the case hospital and the long waiting lists that are already in place. This phenomenon is sustained by the current policies in place, which favour the creation of long waiting lists. However, this phenomenon is also sustained by factors external to the case hospital and to our model, such the relative tolerance of the Ministry of Health and of the authorities, and the widespread idea between patients in Greece that informal payments are necessary for a timely and proper treatment. (Dynamic Hypothesis 3). (4) High Mortality Rates can be explained by the Loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which lead to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to higher Failure & Mortality Rates. (Dynamic Hypothesis 4). (5) Numerous Medical Errors can be explained by the Loop B3 - Actual Time Available & Errors, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at higher Difficulty of Shift Schedule for nurses and doctors, less Proper Communication & Attendance to Patients’ Needs and, finally, to more medical, nursing and patients’ Errors (Dynamic Hypothesis 5). (6) High Nosocomial & Multidrug-resistant bacteria Infections Rates can be explained by the loops R5 – Multidrug Resistance in the General Population and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause Nosocomial Infections to increase in the long run, resulting at more Complications and higher Multidrug Resistance in the General Population (Dynamic Hypothesis 6). (7) Low Adheremce to Clinical Guidelines and Treatment Protocols can be explained by the loop B4 - Actual Time Available and Adherence to Guidelines & Protocols, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient in the long run, as in the Limits to Success archetype, resulting at increased Difficulty of Shift Schedule for nurses and doctors, low Availability of Equipment, ICT, Standard Procedures & Digital Forms and, finally, to low Adherence to Guidelines & Protocols. In order to test those seven hypotheses, a quantified SD model (a stock-flow diagram) would be needed, as that would enable us to run simulations and test our hypothesis in different scenarios to analyse the loop dominance. Such a model is out of the scope and purposes of the present, qualitative study and is not included, but is recommended for future research. However, we used the Dynamic Performance Management analysis as an alternative method, in order to: (1) identify Strategic Resources, Performance Drivers and End Results of hospital performance and show their role in the hospital performance management and measurement; (2) show how the time factor influences the overall hospital performance; (3) understand the contribution of each one of the four hospital divisions (the Medical, the Nursing, the Administrative & Financial and the Technical division) on the End Results (i.e., the final hospital services produced); (4) allow the division managers to start concentrating on the core intermediate, administrative products that divisions are required to deliver on the process that leads to the final end-results; (5) map the ultimate and intermediate services value chain provided to both external and internal users of the case hospital; (6) make performance measures (i.e., the drivers and end-results associated with the delivery of products) explicit and then link them to the goals and objectives of the division managers of the case hospital; (7) discuss the insights that the DPM analysis offers us for a sustainable Performance Management in Greek public hospitals in general, and in the case hospital in particular. The identification of Strategic Resources, Performance Drivers and intermediate End Results, as well as the different views that our DPM analysis offered (i.e., instrumental, dynamic, subjective, objective) provided the hospital decision-makers with signs of potential future shift in End Results, and can help public hospital managers in Greece interpret and calculate the consequences of an incident or the implications of a policy; show possible discrepancies on performance; and try to mitigate it. The performance measures we identified could be helpful to foresee possible changes in the financial and clinical results of public hospitals in Greece. When framed in a wider sense than budgetary control, transaction cost drivers can provide hospital managers and policy makers in Greece with valuable information for strategic planning, such as the opportunity to identify trade-offs in space and in time (e.g., higher costs for investments and for managerial capacity building in the short-run, versus investments in equipment, ICT, and facilities that would increase performance in the long run). Thus, the performance management policies adopted at the case hospital during the healthcare reform ( i.e., structure and process reforms undertaken) and their overall impact for Greek public hospitals’ outputs and outcomes, can now be examined through a different “lenses” by the hospital managers; lenses that will allow them overcome the seven counterintuitive, negative outcomes documented, and align the hospital’s and the different division’s and departments’ goals and actions to achieve improved efficiency and effectiveness, along with better hospital service quality for patients.

Researchers all around the globe have not yet come to an end as regards the supposed positive impact of traditional performance management systems in healthcare, and some research has shown that, paradoxically, performance management policies do not always lead to improved hospital performance. Despite the extensive research identifying the “pitfalls” of the NPM reforms around Europe and the unintended consequences for hospital staff and patients, little is known about the mechanisms that caused those negative effects, which essentially creates a research gap worth investigating. This PhD study tries to address this gap and show why do traditional PM Systems in healthcare not always lead to improved performance, by outlining the unintended consequences of the Greek healthcare reform in a public hospital. By conducting empirical research using a case-study, and by adopting a systemic perspective, this research addresses this gap and sheds light on how hospital performance is perceived by stakeholders of a Greek public hospital and what mechanisms drive its dynamic behaviour. Following a systemic approach, the selected case study - which is a real hospital in the Greek Healthcare system - allowed us to investigate the causing mechanisms of the negative consequences of the Greek healthcare reform on the performance of the case hospital. In doing so, we framed our analysis using the Dynamic Performance Management methodology. Recently, researchers have started to see those negative outcomes as “system pitfalls”, occurring from the non-linear interconnection and the dynamic interaction of the different elements and factors that comprise the health system and the healthcare institutions, i.e., their structure, the policies implemented, the behaviour and the decisions of healthcare workers and patients inside this system. The implementation of a systemic performance assessment methodology in Healthcare is sponsored by many recent scholarly contributions in the field (Arnaboldi et al., 2015; Costanza et al., 2014; Bivona, 2010, 2015; Bivona & Cosenz, 2017a, 2017b; Bivona & Noto, 2020; Davahli et al., 2020; Franco-Santos & Otley, 2018; Fryer et al., 2009; Helal, 2016; Renmans et al., 2017; Mwita, 2000; Noto et al., 2020; Vainieri, Ferrè, et al., 2019; Vainieri, Noto, et al., 2020; Wang et al., 2020). Adopting a systemic perspective means taking as a unit of analysis the organisation as a whole, and not one unit or department; acknowledging its internal and external environment and culture in which health care is performed; and considering the concurrent existence of the pitfalls documented as inherent to the structure of the system and the policies implemented. Studies using such a methodology would be necessary in order to address the gap in existing knowledge, as well as to support policy-makers in designing better, more quality-oriented healthcare policies, interventions and reforms in the future. The purpose of this study was to empirically conceptualise a qualitative model of hospital performance as perceived by stakeholders of a Greek public hospital and use the DPM analysis in order to help policymakers in Greece re-design performance management policies and foster hospital performance. We adopted a systemic, participatory, inductive and dynamic approach by combining the Group Model Building and System Dynamics methodologies into the Dynamic Performance Management approach (Bianchi, 2016). Other research traditions identified in our study are the Stakeholders Theory and Participation. All those approaches stand in the constructivist side of the continuum as research approaches, because they all consider realities as subjective, complex and multi-layered, actively shaped by perceptions and opinions of stakeholders (De Gooyert, 2019; Lane & Schwaninger, 2008). Mixed methods were used to facilitate our approach, combining primary qualitative data from two Group Model Building sessions; four open, unstructured preliminary interviews; and seven semi-structured, disconfirmatory interviews; with secondary, qualitative and quantitative data from a scoping literature review and from a critical literature review; as well as from official, open-access, online text-documents and closed-access, internal text-documents of the hospital’s interdepartmental communication. An open call for participation in the research was sent by email to around 70 different hospitals in the cities of Athens and Thessaloniki in Greece, and the gatekeeper was identified. Starting from the gatekeeper, snowball sampling was used to select 10 participants in the case hospital for the Group Model Building (GMB) sessions, including at least one person from each main key-stakeholder category that our extensive stakeholder analysis identified (i.e., managers, doctors, nurses, paramedics and patients), with the purpose of “eliciting” their mental models and “capturing” them in a qualitative system dynamics model (causal loop diagram). Four of the participants were also interviewed before the GMB sessions (face-to-face, one-to-one preliminary interviews). Convenient sampling was used in order to identify seven more public hospital stakeholders from other public hospitals in Greece for the disconfirmatory interviews. The data analysis included a Scoping Review of the International Literature of Performance Management in the Health Sector; a Critical Review of the Literature on the Greek Healthcare Reform; a Stakeholder Analysis; a Narrative Analysis of Preliminary Interviews and Documents; a Qualitative System Dynamics Analysis (Causal Loop Diagram) of the Simplified version of the Conceptual Model of Hospital Performance created during the GMB sessions; and, finally, the Dynamic Performance Management (DPM) analysis. The GMB sessions helped hospital stakeholders gain a better understanding of what hospital performance is in a more systematic way; define it; show its trend (dynamic behaviour) in the hospital during the last decade in a diagram; and conceptualise it as a system, depicted as a qualitative system dynamics model of hospital performance (CLD - Causal Loop Diagram). The two final versions of this CLD Model (i.e., the Conceptual and the Policy Models of Hospital Performance, available in Appendixes 21 and 22 respectively and thoroughly described in terms of the variables and links they contain in Appendix 24) are the main outputs of the GMB sessions, and formed the basis of our analysis and research findings. The Conceptual Model of Hospital Performance is a CLD model that depicts the actual structure of hospital performance and can be used to explain its currently low levels, whereas the Policy Model of Hospital Performance is extended to incorporate the policy structure, i.e., the changes in the system structure which are necessary, according to our participant stakeholders, in order to improve hospital performance. Hospital performance was defined by the participant stakeholders as the provision of patient-centred care to the patient, with safety (for the patients and the staff); responsibility (adherence to protocols, proportions and procedures) and dignity (nice and clean facilities, reduced waiting times and no informal payments). The historical trend of the Hospital performance in the case hospital was also depicted in a diagram over time called Reference Mode (available in Appendix 19). The Reference Mode created and agreed upon by the participants showed that, despite the counterintuitive negative outcomes documented, the level of the overall performance in the case hospital has been slightly increasing after the healthcare reform and is now stabilizing. Our research showed that the Performance Management policies introduced during the Greek healthcare reform had a negative impact on many aspects of hospital performance in general, and in our case hospital in particular. The new policies undeniably contributed to the reduction of hospital spending, but they simultaneously contributed to the deterioration of hospital service quality. Goal-setting, the main PM strategy followed by Greek public hospitals according to Law N4369/16, is until today not properly implemented in the case hospital and managers seem to treat performance objectives as completely separated from performance and quality, and to consider them totally outside of their everyday tasks. Those findings of the preliminary interviews and documents analysis were validated from the findings of the pretests, conducted before the GMB sessions. Four of the goals that were set by the division managers of the case hospital came up during the GMB sessions and were integrated in the CLD model that the participants built: Standardization of the nursing forms of the nursing departments and units; Standardization of clinical procedures; Use of an Information System in the Interdepartmental Communication; and Application of digital signature and electronic document management. We combined our findings from the documents’ analysis with the descriptions of those goals, as set by the division managers, and we informed them with the findings from our DPM instrumental and objective analysis, which allowed us identify the activities and the resources that are needed for the achievement of each of those four goals. In that respect, we found that apart from the “tangible” strategic resources identified by the managers of the case hospital (e.g., financial and human resources) as essential in the achievement of each of those four goals, Management Capacity - which is an intermediate, administrative product of the hospital, built by the public workers - was equally necessary. Out of all the unintended negative outcomes of the Greek healthcare reform documented in the literature, we found the following seven negative outcomes to be present at the case hospital: (1) Low Quality and Safety of Services perceived by health workers and patients; (2) Low Patient Satisfaction; (3) Informal Payments; (4) High Mortality Rates; (5) Numerous Medical Errors; (6) High Nosocomial & Multidrug-Resistant Bacteria Infections Rates; (7) Low adherence to Clinical Guidelines and Treatment Protocols. Regarding those seven negative outcomes, the analysis of the simplified version of the Conceptual Model of Hospital Quality which the participant stakeholders created during the GMB sessions at the case hospital, showed that: (1) Low Quality and Safety are mostly associated with the variables Survival Rate / Patients' Health Status & Quality of Life and Complications of our model, and can be explained by the dominance of the balancing loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause those two variables to decrease as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment (Dynamic Hypothesis 1). (2) Low Patient Satisfaction can be explained by the dominance of the loops B1 – Word of Mouth & Waiting Times, B2 – Patient Satisfaction & Attendance to Patients’ Needs, B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, all of which lead to a gradual decrease and stabilisation of Patient Satisfaction and of Hospital Reputation in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Informal Payments for early Surgery/Admission longer Waiting List for Surgery or Admission, longer Waiting Time in ER & Outpatient Services and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment. (Dynamic Hypothesis 2). (3) The existence of Informal Payments can be explained by the Loop R2 – Informal Payments & Corruption, which leads to a perpetual increase of private spending and to the outspread of corruption between the case hospital doctors, given the good reputation of the case hospital and the long waiting lists that are already in place. This phenomenon is sustained by the current policies in place, which favour the creation of long waiting lists. However, this phenomenon is also sustained by factors external to the case hospital and to our model, such the relative tolerance of the Ministry of Health and of the authorities, and the widespread idea between patients in Greece that informal payments are necessary for a timely and proper treatment. (Dynamic Hypothesis 3). (4) High Mortality Rates can be explained by the Loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which lead to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to higher Failure & Mortality Rates. (Dynamic Hypothesis 4). (5) Numerous Medical Errors can be explained by the Loop B3 - Actual Time Available & Errors, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at higher Difficulty of Shift Schedule for nurses and doctors, less Proper Communication & Attendance to Patients’ Needs and, finally, to more medical, nursing and patients’ Errors (Dynamic Hypothesis 5). (6) High Nosocomial & Multidrug-resistant bacteria Infections Rates can be explained by the loops R5 – Multidrug Resistance in the General Population and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause Nosocomial Infections to increase in the long run, resulting at more Complications and higher Multidrug Resistance in the General Population (Dynamic Hypothesis 6). (7) Low Adheremce to Clinical Guidelines and Treatment Protocols can be explained by the loop B4 - Actual Time Available and Adherence to Guidelines & Protocols, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient in the long run, as in the Limits to Success archetype, resulting at increased Difficulty of Shift Schedule for nurses and doctors, low Availability of Equipment, ICT, Standard Procedures & Digital Forms and, finally, to low Adherence to Guidelines & Protocols. In order to test those seven hypotheses, a quantified SD model (a stock-flow diagram) would be needed, as that would enable us to run simulations and test our hypothesis in different scenarios to analyse the loop dominance. Such a model is out of the scope and purposes of the present, qualitative study and is not included, but is recommended for future research. However, we used the Dynamic Performance Management analysis as an alternative method, in order to: (1) identify Strategic Resources, Performance Drivers and End Results of hospital performance and show their role in the hospital performance management and measurement; (2) show how the time factor influences the overall hospital performance; (3) understand the contribution of each one of the four hospital divisions (the Medical, the Nursing, the Administrative & Financial and the Technical division) on the End Results (i.e., the final hospital services produced); (4) allow the division managers to start concentrating on the core intermediate, administrative products that divisions are required to deliver on the process that leads to the final end-results; (5) map the ultimate and intermediate services value chain provided to both external and internal users of the case hospital; (6) make performance measures (i.e., the drivers and end-results associated with the delivery of products) explicit and then link them to the goals and objectives of the division managers of the case hospital; (7) discuss the insights that the DPM analysis offers us for a sustainable Performance Management in Greek public hospitals in general, and in the case hospital in particular. The identification of Strategic Resources, Performance Drivers and intermediate End Results, as well as the different views that our DPM analysis offered (i.e., instrumental, dynamic, subjective, objective) provided the hospital decision-makers with signs of potential future shift in End Results, and can help public hospital managers in Greece interpret and calculate the consequences of an incident or the implications of a policy; show possible discrepancies on performance; and try to mitigate it. The performance measures we identified could be helpful to foresee possible changes in the financial and clinical results of public hospitals in Greece. When framed in a wider sense than budgetary control, transaction cost drivers can provide hospital managers and policy makers in Greece with valuable information for strategic planning, such as the opportunity to identify trade-offs in space and in time (e.g., higher costs for investments and for managerial capacity building in the short-run, versus investments in equipment, ICT, and facilities that would increase performance in the long run). Thus, the performance management policies adopted at the case hospital during the healthcare reform ( i.e., structure and process reforms undertaken) and their overall impact for Greek public hospitals’ outputs and outcomes, can now be examined through a different “lenses” by the hospital managers; lenses that will allow them overcome the seven counterintuitive, negative outcomes documented, and align the hospital’s and the different division’s and departments’ goals and actions to achieve improved efficiency and effectiveness, along with better hospital service quality for patients.

(2021). Why do Traditional Performance Management Systems in Healthcare not always lead to Improved Performance? Outlining the Unintended Consequences of the Greek Healthcare Reform in a Public Hospital through a Dynamic Performance Management Approach..

Why do Traditional Performance Management Systems in Healthcare not always lead to Improved Performance? Outlining the Unintended Consequences of the Greek Healthcare Reform in a Public Hospital through a Dynamic Performance Management Approach.

LENAKAKI, Angeliki
2021-01-01

Abstract

Researchers all around the globe have not yet come to an end as regards the supposed positive impact of traditional performance management systems in healthcare, and some research has shown that, paradoxically, performance management policies do not always lead to improved hospital performance. Despite the extensive research identifying the “pitfalls” of the NPM reforms around Europe and the unintended consequences for hospital staff and patients, little is known about the mechanisms that caused those negative effects, which essentially creates a research gap worth investigating. This PhD study tries to address this gap and show why do traditional PM Systems in healthcare not always lead to improved performance, by outlining the unintended consequences of the Greek healthcare reform in a public hospital. By conducting empirical research using a case-study, and by adopting a systemic perspective, this research addresses this gap and sheds light on how hospital performance is perceived by stakeholders of a Greek public hospital and what mechanisms drive its dynamic behaviour. Following a systemic approach, the selected case study - which is a real hospital in the Greek Healthcare system - allowed us to investigate the causing mechanisms of the negative consequences of the Greek healthcare reform on the performance of the case hospital. In doing so, we framed our analysis using the Dynamic Performance Management methodology. Recently, researchers have started to see those negative outcomes as “system pitfalls”, occurring from the non-linear interconnection and the dynamic interaction of the different elements and factors that comprise the health system and the healthcare institutions, i.e., their structure, the policies implemented, the behaviour and the decisions of healthcare workers and patients inside this system. The implementation of a systemic performance assessment methodology in Healthcare is sponsored by many recent scholarly contributions in the field (Arnaboldi et al., 2015; Costanza et al., 2014; Bivona, 2010, 2015; Bivona & Cosenz, 2017a, 2017b; Bivona & Noto, 2020; Davahli et al., 2020; Franco-Santos & Otley, 2018; Fryer et al., 2009; Helal, 2016; Renmans et al., 2017; Mwita, 2000; Noto et al., 2020; Vainieri, Ferrè, et al., 2019; Vainieri, Noto, et al., 2020; Wang et al., 2020). Adopting a systemic perspective means taking as a unit of analysis the organisation as a whole, and not one unit or department; acknowledging its internal and external environment and culture in which health care is performed; and considering the concurrent existence of the pitfalls documented as inherent to the structure of the system and the policies implemented. Studies using such a methodology would be necessary in order to address the gap in existing knowledge, as well as to support policy-makers in designing better, more quality-oriented healthcare policies, interventions and reforms in the future. The purpose of this study was to empirically conceptualise a qualitative model of hospital performance as perceived by stakeholders of a Greek public hospital and use the DPM analysis in order to help policymakers in Greece re-design performance management policies and foster hospital performance. We adopted a systemic, participatory, inductive and dynamic approach by combining the Group Model Building and System Dynamics methodologies into the Dynamic Performance Management approach (Bianchi, 2016). Other research traditions identified in our study are the Stakeholders Theory and Participation. All those approaches stand in the constructivist side of the continuum as research approaches, because they all consider realities as subjective, complex and multi-layered, actively shaped by perceptions and opinions of stakeholders (De Gooyert, 2019; Lane & Schwaninger, 2008). Mixed methods were used to facilitate our approach, combining primary qualitative data from two Group Model Building sessions; four open, unstructured preliminary interviews; and seven semi-structured, disconfirmatory interviews; with secondary, qualitative and quantitative data from a scoping literature review and from a critical literature review; as well as from official, open-access, online text-documents and closed-access, internal text-documents of the hospital’s interdepartmental communication. An open call for participation in the research was sent by email to around 70 different hospitals in the cities of Athens and Thessaloniki in Greece, and the gatekeeper was identified. Starting from the gatekeeper, snowball sampling was used to select 10 participants in the case hospital for the Group Model Building (GMB) sessions, including at least one person from each main key-stakeholder category that our extensive stakeholder analysis identified (i.e., managers, doctors, nurses, paramedics and patients), with the purpose of “eliciting” their mental models and “capturing” them in a qualitative system dynamics model (causal loop diagram). Four of the participants were also interviewed before the GMB sessions (face-to-face, one-to-one preliminary interviews). Convenient sampling was used in order to identify seven more public hospital stakeholders from other public hospitals in Greece for the disconfirmatory interviews. The data analysis included a Scoping Review of the International Literature of Performance Management in the Health Sector; a Critical Review of the Literature on the Greek Healthcare Reform; a Stakeholder Analysis; a Narrative Analysis of Preliminary Interviews and Documents; a Qualitative System Dynamics Analysis (Causal Loop Diagram) of the Simplified version of the Conceptual Model of Hospital Performance created during the GMB sessions; and, finally, the Dynamic Performance Management (DPM) analysis. The GMB sessions helped hospital stakeholders gain a better understanding of what hospital performance is in a more systematic way; define it; show its trend (dynamic behaviour) in the hospital during the last decade in a diagram; and conceptualise it as a system, depicted as a qualitative system dynamics model of hospital performance (CLD - Causal Loop Diagram). The two final versions of this CLD Model (i.e., the Conceptual and the Policy Models of Hospital Performance, available in Appendixes 21 and 22 respectively and thoroughly described in terms of the variables and links they contain in Appendix 24) are the main outputs of the GMB sessions, and formed the basis of our analysis and research findings. The Conceptual Model of Hospital Performance is a CLD model that depicts the actual structure of hospital performance and can be used to explain its currently low levels, whereas the Policy Model of Hospital Performance is extended to incorporate the policy structure, i.e., the changes in the system structure which are necessary, according to our participant stakeholders, in order to improve hospital performance. Hospital performance was defined by the participant stakeholders as the provision of patient-centred care to the patient, with safety (for the patients and the staff); responsibility (adherence to protocols, proportions and procedures) and dignity (nice and clean facilities, reduced waiting times and no informal payments). The historical trend of the Hospital performance in the case hospital was also depicted in a diagram over time called Reference Mode (available in Appendix 19). The Reference Mode created and agreed upon by the participants showed that, despite the counterintuitive negative outcomes documented, the level of the overall performance in the case hospital has been slightly increasing after the healthcare reform and is now stabilizing. Our research showed that the Performance Management policies introduced during the Greek healthcare reform had a negative impact on many aspects of hospital performance in general, and in our case hospital in particular. The new policies undeniably contributed to the reduction of hospital spending, but they simultaneously contributed to the deterioration of hospital service quality. Goal-setting, the main PM strategy followed by Greek public hospitals according to Law N4369/16, is until today not properly implemented in the case hospital and managers seem to treat performance objectives as completely separated from performance and quality, and to consider them totally outside of their everyday tasks. Those findings of the preliminary interviews and documents analysis were validated from the findings of the pretests, conducted before the GMB sessions. Four of the goals that were set by the division managers of the case hospital came up during the GMB sessions and were integrated in the CLD model that the participants built: Standardization of the nursing forms of the nursing departments and units; Standardization of clinical procedures; Use of an Information System in the Interdepartmental Communication; and Application of digital signature and electronic document management. We combined our findings from the documents’ analysis with the descriptions of those goals, as set by the division managers, and we informed them with the findings from our DPM instrumental and objective analysis, which allowed us identify the activities and the resources that are needed for the achievement of each of those four goals. In that respect, we found that apart from the “tangible” strategic resources identified by the managers of the case hospital (e.g., financial and human resources) as essential in the achievement of each of those four goals, Management Capacity - which is an intermediate, administrative product of the hospital, built by the public workers - was equally necessary. Out of all the unintended negative outcomes of the Greek healthcare reform documented in the literature, we found the following seven negative outcomes to be present at the case hospital: (1) Low Quality and Safety of Services perceived by health workers and patients; (2) Low Patient Satisfaction; (3) Informal Payments; (4) High Mortality Rates; (5) Numerous Medical Errors; (6) High Nosocomial & Multidrug-Resistant Bacteria Infections Rates; (7) Low adherence to Clinical Guidelines and Treatment Protocols. Regarding those seven negative outcomes, the analysis of the simplified version of the Conceptual Model of Hospital Quality which the participant stakeholders created during the GMB sessions at the case hospital, showed that: (1) Low Quality and Safety are mostly associated with the variables Survival Rate / Patients' Health Status & Quality of Life and Complications of our model, and can be explained by the dominance of the balancing loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause those two variables to decrease as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment (Dynamic Hypothesis 1). (2) Low Patient Satisfaction can be explained by the dominance of the loops B1 – Word of Mouth & Waiting Times, B2 – Patient Satisfaction & Attendance to Patients’ Needs, B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, all of which lead to a gradual decrease and stabilisation of Patient Satisfaction and of Hospital Reputation in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Informal Payments for early Surgery/Admission longer Waiting List for Surgery or Admission, longer Waiting Time in ER & Outpatient Services and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment. (Dynamic Hypothesis 2). (3) The existence of Informal Payments can be explained by the Loop R2 – Informal Payments & Corruption, which leads to a perpetual increase of private spending and to the outspread of corruption between the case hospital doctors, given the good reputation of the case hospital and the long waiting lists that are already in place. This phenomenon is sustained by the current policies in place, which favour the creation of long waiting lists. However, this phenomenon is also sustained by factors external to the case hospital and to our model, such the relative tolerance of the Ministry of Health and of the authorities, and the widespread idea between patients in Greece that informal payments are necessary for a timely and proper treatment. (Dynamic Hypothesis 3). (4) High Mortality Rates can be explained by the Loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which lead to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to higher Failure & Mortality Rates. (Dynamic Hypothesis 4). (5) Numerous Medical Errors can be explained by the Loop B3 - Actual Time Available & Errors, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at higher Difficulty of Shift Schedule for nurses and doctors, less Proper Communication & Attendance to Patients’ Needs and, finally, to more medical, nursing and patients’ Errors (Dynamic Hypothesis 5). (6) High Nosocomial & Multidrug-resistant bacteria Infections Rates can be explained by the loops R5 – Multidrug Resistance in the General Population and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause Nosocomial Infections to increase in the long run, resulting at more Complications and higher Multidrug Resistance in the General Population (Dynamic Hypothesis 6). (7) Low Adheremce to Clinical Guidelines and Treatment Protocols can be explained by the loop B4 - Actual Time Available and Adherence to Guidelines & Protocols, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient in the long run, as in the Limits to Success archetype, resulting at increased Difficulty of Shift Schedule for nurses and doctors, low Availability of Equipment, ICT, Standard Procedures & Digital Forms and, finally, to low Adherence to Guidelines & Protocols. In order to test those seven hypotheses, a quantified SD model (a stock-flow diagram) would be needed, as that would enable us to run simulations and test our hypothesis in different scenarios to analyse the loop dominance. Such a model is out of the scope and purposes of the present, qualitative study and is not included, but is recommended for future research. However, we used the Dynamic Performance Management analysis as an alternative method, in order to: (1) identify Strategic Resources, Performance Drivers and End Results of hospital performance and show their role in the hospital performance management and measurement; (2) show how the time factor influences the overall hospital performance; (3) understand the contribution of each one of the four hospital divisions (the Medical, the Nursing, the Administrative & Financial and the Technical division) on the End Results (i.e., the final hospital services produced); (4) allow the division managers to start concentrating on the core intermediate, administrative products that divisions are required to deliver on the process that leads to the final end-results; (5) map the ultimate and intermediate services value chain provided to both external and internal users of the case hospital; (6) make performance measures (i.e., the drivers and end-results associated with the delivery of products) explicit and then link them to the goals and objectives of the division managers of the case hospital; (7) discuss the insights that the DPM analysis offers us for a sustainable Performance Management in Greek public hospitals in general, and in the case hospital in particular. The identification of Strategic Resources, Performance Drivers and intermediate End Results, as well as the different views that our DPM analysis offered (i.e., instrumental, dynamic, subjective, objective) provided the hospital decision-makers with signs of potential future shift in End Results, and can help public hospital managers in Greece interpret and calculate the consequences of an incident or the implications of a policy; show possible discrepancies on performance; and try to mitigate it. The performance measures we identified could be helpful to foresee possible changes in the financial and clinical results of public hospitals in Greece. When framed in a wider sense than budgetary control, transaction cost drivers can provide hospital managers and policy makers in Greece with valuable information for strategic planning, such as the opportunity to identify trade-offs in space and in time (e.g., higher costs for investments and for managerial capacity building in the short-run, versus investments in equipment, ICT, and facilities that would increase performance in the long run). Thus, the performance management policies adopted at the case hospital during the healthcare reform ( i.e., structure and process reforms undertaken) and their overall impact for Greek public hospitals’ outputs and outcomes, can now be examined through a different “lenses” by the hospital managers; lenses that will allow them overcome the seven counterintuitive, negative outcomes documented, and align the hospital’s and the different division’s and departments’ goals and actions to achieve improved efficiency and effectiveness, along with better hospital service quality for patients.
Why do Traditional Performance Management Systems in Healthcare not always lead to Improved Performance? Outlining the Unintended Consequences of the Greek Healthcare Reform in a Public Hospital through a Dynamic Performance Management Approach.
2021
Researchers all around the globe have not yet come to an end as regards the supposed positive impact of traditional performance management systems in healthcare, and some research has shown that, paradoxically, performance management policies do not always lead to improved hospital performance. Despite the extensive research identifying the “pitfalls” of the NPM reforms around Europe and the unintended consequences for hospital staff and patients, little is known about the mechanisms that caused those negative effects, which essentially creates a research gap worth investigating. This PhD study tries to address this gap and show why do traditional PM Systems in healthcare not always lead to improved performance, by outlining the unintended consequences of the Greek healthcare reform in a public hospital. By conducting empirical research using a case-study, and by adopting a systemic perspective, this research addresses this gap and sheds light on how hospital performance is perceived by stakeholders of a Greek public hospital and what mechanisms drive its dynamic behaviour. Following a systemic approach, the selected case study - which is a real hospital in the Greek Healthcare system - allowed us to investigate the causing mechanisms of the negative consequences of the Greek healthcare reform on the performance of the case hospital. In doing so, we framed our analysis using the Dynamic Performance Management methodology. Recently, researchers have started to see those negative outcomes as “system pitfalls”, occurring from the non-linear interconnection and the dynamic interaction of the different elements and factors that comprise the health system and the healthcare institutions, i.e., their structure, the policies implemented, the behaviour and the decisions of healthcare workers and patients inside this system. The implementation of a systemic performance assessment methodology in Healthcare is sponsored by many recent scholarly contributions in the field (Arnaboldi et al., 2015; Costanza et al., 2014; Bivona, 2010, 2015; Bivona & Cosenz, 2017a, 2017b; Bivona & Noto, 2020; Davahli et al., 2020; Franco-Santos & Otley, 2018; Fryer et al., 2009; Helal, 2016; Renmans et al., 2017; Mwita, 2000; Noto et al., 2020; Vainieri, Ferrè, et al., 2019; Vainieri, Noto, et al., 2020; Wang et al., 2020). Adopting a systemic perspective means taking as a unit of analysis the organisation as a whole, and not one unit or department; acknowledging its internal and external environment and culture in which health care is performed; and considering the concurrent existence of the pitfalls documented as inherent to the structure of the system and the policies implemented. Studies using such a methodology would be necessary in order to address the gap in existing knowledge, as well as to support policy-makers in designing better, more quality-oriented healthcare policies, interventions and reforms in the future. The purpose of this study was to empirically conceptualise a qualitative model of hospital performance as perceived by stakeholders of a Greek public hospital and use the DPM analysis in order to help policymakers in Greece re-design performance management policies and foster hospital performance. We adopted a systemic, participatory, inductive and dynamic approach by combining the Group Model Building and System Dynamics methodologies into the Dynamic Performance Management approach (Bianchi, 2016). Other research traditions identified in our study are the Stakeholders Theory and Participation. All those approaches stand in the constructivist side of the continuum as research approaches, because they all consider realities as subjective, complex and multi-layered, actively shaped by perceptions and opinions of stakeholders (De Gooyert, 2019; Lane & Schwaninger, 2008). Mixed methods were used to facilitate our approach, combining primary qualitative data from two Group Model Building sessions; four open, unstructured preliminary interviews; and seven semi-structured, disconfirmatory interviews; with secondary, qualitative and quantitative data from a scoping literature review and from a critical literature review; as well as from official, open-access, online text-documents and closed-access, internal text-documents of the hospital’s interdepartmental communication. An open call for participation in the research was sent by email to around 70 different hospitals in the cities of Athens and Thessaloniki in Greece, and the gatekeeper was identified. Starting from the gatekeeper, snowball sampling was used to select 10 participants in the case hospital for the Group Model Building (GMB) sessions, including at least one person from each main key-stakeholder category that our extensive stakeholder analysis identified (i.e., managers, doctors, nurses, paramedics and patients), with the purpose of “eliciting” their mental models and “capturing” them in a qualitative system dynamics model (causal loop diagram). Four of the participants were also interviewed before the GMB sessions (face-to-face, one-to-one preliminary interviews). Convenient sampling was used in order to identify seven more public hospital stakeholders from other public hospitals in Greece for the disconfirmatory interviews. The data analysis included a Scoping Review of the International Literature of Performance Management in the Health Sector; a Critical Review of the Literature on the Greek Healthcare Reform; a Stakeholder Analysis; a Narrative Analysis of Preliminary Interviews and Documents; a Qualitative System Dynamics Analysis (Causal Loop Diagram) of the Simplified version of the Conceptual Model of Hospital Performance created during the GMB sessions; and, finally, the Dynamic Performance Management (DPM) analysis. The GMB sessions helped hospital stakeholders gain a better understanding of what hospital performance is in a more systematic way; define it; show its trend (dynamic behaviour) in the hospital during the last decade in a diagram; and conceptualise it as a system, depicted as a qualitative system dynamics model of hospital performance (CLD - Causal Loop Diagram). The two final versions of this CLD Model (i.e., the Conceptual and the Policy Models of Hospital Performance, available in Appendixes 21 and 22 respectively and thoroughly described in terms of the variables and links they contain in Appendix 24) are the main outputs of the GMB sessions, and formed the basis of our analysis and research findings. The Conceptual Model of Hospital Performance is a CLD model that depicts the actual structure of hospital performance and can be used to explain its currently low levels, whereas the Policy Model of Hospital Performance is extended to incorporate the policy structure, i.e., the changes in the system structure which are necessary, according to our participant stakeholders, in order to improve hospital performance. Hospital performance was defined by the participant stakeholders as the provision of patient-centred care to the patient, with safety (for the patients and the staff); responsibility (adherence to protocols, proportions and procedures) and dignity (nice and clean facilities, reduced waiting times and no informal payments). The historical trend of the Hospital performance in the case hospital was also depicted in a diagram over time called Reference Mode (available in Appendix 19). The Reference Mode created and agreed upon by the participants showed that, despite the counterintuitive negative outcomes documented, the level of the overall performance in the case hospital has been slightly increasing after the healthcare reform and is now stabilizing. Our research showed that the Performance Management policies introduced during the Greek healthcare reform had a negative impact on many aspects of hospital performance in general, and in our case hospital in particular. The new policies undeniably contributed to the reduction of hospital spending, but they simultaneously contributed to the deterioration of hospital service quality. Goal-setting, the main PM strategy followed by Greek public hospitals according to Law N4369/16, is until today not properly implemented in the case hospital and managers seem to treat performance objectives as completely separated from performance and quality, and to consider them totally outside of their everyday tasks. Those findings of the preliminary interviews and documents analysis were validated from the findings of the pretests, conducted before the GMB sessions. Four of the goals that were set by the division managers of the case hospital came up during the GMB sessions and were integrated in the CLD model that the participants built: Standardization of the nursing forms of the nursing departments and units; Standardization of clinical procedures; Use of an Information System in the Interdepartmental Communication; and Application of digital signature and electronic document management. We combined our findings from the documents’ analysis with the descriptions of those goals, as set by the division managers, and we informed them with the findings from our DPM instrumental and objective analysis, which allowed us identify the activities and the resources that are needed for the achievement of each of those four goals. In that respect, we found that apart from the “tangible” strategic resources identified by the managers of the case hospital (e.g., financial and human resources) as essential in the achievement of each of those four goals, Management Capacity - which is an intermediate, administrative product of the hospital, built by the public workers - was equally necessary. Out of all the unintended negative outcomes of the Greek healthcare reform documented in the literature, we found the following seven negative outcomes to be present at the case hospital: (1) Low Quality and Safety of Services perceived by health workers and patients; (2) Low Patient Satisfaction; (3) Informal Payments; (4) High Mortality Rates; (5) Numerous Medical Errors; (6) High Nosocomial & Multidrug-Resistant Bacteria Infections Rates; (7) Low adherence to Clinical Guidelines and Treatment Protocols. Regarding those seven negative outcomes, the analysis of the simplified version of the Conceptual Model of Hospital Quality which the participant stakeholders created during the GMB sessions at the case hospital, showed that: (1) Low Quality and Safety are mostly associated with the variables Survival Rate / Patients' Health Status & Quality of Life and Complications of our model, and can be explained by the dominance of the balancing loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause those two variables to decrease as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment (Dynamic Hypothesis 1). (2) Low Patient Satisfaction can be explained by the dominance of the loops B1 – Word of Mouth & Waiting Times, B2 – Patient Satisfaction & Attendance to Patients’ Needs, B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, all of which lead to a gradual decrease and stabilisation of Patient Satisfaction and of Hospital Reputation in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Informal Payments for early Surgery/Admission longer Waiting List for Surgery or Admission, longer Waiting Time in ER & Outpatient Services and, finally, to lower Survival Rate and Patients’ Health Status & Quality of Life after treatment. (Dynamic Hypothesis 2). (3) The existence of Informal Payments can be explained by the Loop R2 – Informal Payments & Corruption, which leads to a perpetual increase of private spending and to the outspread of corruption between the case hospital doctors, given the good reputation of the case hospital and the long waiting lists that are already in place. This phenomenon is sustained by the current policies in place, which favour the creation of long waiting lists. However, this phenomenon is also sustained by factors external to the case hospital and to our model, such the relative tolerance of the Ministry of Health and of the authorities, and the widespread idea between patients in Greece that informal payments are necessary for a timely and proper treatment. (Dynamic Hypothesis 3). (4) High Mortality Rates can be explained by the Loops B3 - Actual Time Available & Errors, and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which lead to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at less Proper Communication & Attendance to Patients’ Needs, more Errors and Complications, longer Length of Stay, higher Nosocomial Infections Rate, and, finally, to higher Failure & Mortality Rates. (Dynamic Hypothesis 4). (5) Numerous Medical Errors can be explained by the Loop B3 - Actual Time Available & Errors, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient and of the Adherence to Guidelines & Protocols in the long run as in the Limits to Success archetype, resulting at higher Difficulty of Shift Schedule for nurses and doctors, less Proper Communication & Attendance to Patients’ Needs and, finally, to more medical, nursing and patients’ Errors (Dynamic Hypothesis 5). (6) High Nosocomial & Multidrug-resistant bacteria Infections Rates can be explained by the loops R5 – Multidrug Resistance in the General Population and B4 - Actual Time Available and Adherence to Guidelines & Protocols, both of which cause Nosocomial Infections to increase in the long run, resulting at more Complications and higher Multidrug Resistance in the General Population (Dynamic Hypothesis 6). (7) Low Adheremce to Clinical Guidelines and Treatment Protocols can be explained by the loop B4 - Actual Time Available and Adherence to Guidelines & Protocols, which leads to a gradual decrease and stabilisation at a low level of the Actual Time Available per Patient in the long run, as in the Limits to Success archetype, resulting at increased Difficulty of Shift Schedule for nurses and doctors, low Availability of Equipment, ICT, Standard Procedures & Digital Forms and, finally, to low Adherence to Guidelines & Protocols. In order to test those seven hypotheses, a quantified SD model (a stock-flow diagram) would be needed, as that would enable us to run simulations and test our hypothesis in different scenarios to analyse the loop dominance. Such a model is out of the scope and purposes of the present, qualitative study and is not included, but is recommended for future research. However, we used the Dynamic Performance Management analysis as an alternative method, in order to: (1) identify Strategic Resources, Performance Drivers and End Results of hospital performance and show their role in the hospital performance management and measurement; (2) show how the time factor influences the overall hospital performance; (3) understand the contribution of each one of the four hospital divisions (the Medical, the Nursing, the Administrative & Financial and the Technical division) on the End Results (i.e., the final hospital services produced); (4) allow the division managers to start concentrating on the core intermediate, administrative products that divisions are required to deliver on the process that leads to the final end-results; (5) map the ultimate and intermediate services value chain provided to both external and internal users of the case hospital; (6) make performance measures (i.e., the drivers and end-results associated with the delivery of products) explicit and then link them to the goals and objectives of the division managers of the case hospital; (7) discuss the insights that the DPM analysis offers us for a sustainable Performance Management in Greek public hospitals in general, and in the case hospital in particular. The identification of Strategic Resources, Performance Drivers and intermediate End Results, as well as the different views that our DPM analysis offered (i.e., instrumental, dynamic, subjective, objective) provided the hospital decision-makers with signs of potential future shift in End Results, and can help public hospital managers in Greece interpret and calculate the consequences of an incident or the implications of a policy; show possible discrepancies on performance; and try to mitigate it. The performance measures we identified could be helpful to foresee possible changes in the financial and clinical results of public hospitals in Greece. When framed in a wider sense than budgetary control, transaction cost drivers can provide hospital managers and policy makers in Greece with valuable information for strategic planning, such as the opportunity to identify trade-offs in space and in time (e.g., higher costs for investments and for managerial capacity building in the short-run, versus investments in equipment, ICT, and facilities that would increase performance in the long run). Thus, the performance management policies adopted at the case hospital during the healthcare reform ( i.e., structure and process reforms undertaken) and their overall impact for Greek public hospitals’ outputs and outcomes, can now be examined through a different “lenses” by the hospital managers; lenses that will allow them overcome the seven counterintuitive, negative outcomes documented, and align the hospital’s and the different division’s and departments’ goals and actions to achieve improved efficiency and effectiveness, along with better hospital service quality for patients.
Performance.,Management.,Healthcare.,Unintended.,Counterintuitive.,Consequences.,Reform.,Public.,Hospital.,Dynamic Performance Management.,DPM.,Greece.,Greek.,System Dynamics.,Group Model Building.,SD.,GMB.,Causal Loop Diagram.,CLD.,System.,Systemic.,
(2021). Why do Traditional Performance Management Systems in Healthcare not always lead to Improved Performance? Outlining the Unintended Consequences of the Greek Healthcare Reform in a Public Hospital through a Dynamic Performance Management Approach..
File in questo prodotto:
File Dimensione Formato  
PhD Thesis - Angeliki Lenakaki.pdf

accesso aperto

Descrizione: Tesi Completa
Tipologia: Tesi di dottorato
Dimensione 50.6 MB
Formato Adobe PDF
50.6 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/514805
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact