Quality behind quantity: institutions and hospital management Open Access

Healthcare systems represent a central social infrastructure with increasing sustainability problems driven by an ageing population, budgetary constraints (Moro Visconti et al., 2019), burdens associated with healthcare delivery and changing disease patterns (Al-Hanawi and Qattan, 2019). The COVID-19 pandemic has also accentuated the difficulties faced by healthcare institutions in responding to the growing demand for care, lack of medical equipment and shortage of healthcare workers (Sun et al., 2021; Soroush et al., 2022).

Globally, health systems aim to provide services to promote and improve the health status of the population (Roussos and Fawcett, 2000; Singh, 2019): this underlines the social relevance of health systems, since a community, in order to sustain itself and function, must be able to keep its members healthy (Vuong et al., 2017; Kilci, 2021). Healthcare is, therefore, a critical area in terms of social impact, as it is responsible for providing services for human beings to meet their health needs, thus directly affecting the quality of life of patients (Khosravi and Izbirak, 2019). In fact, access to effective health services is increasingly seen as an important aspect of social welfare and, as such, absorbs a substantial part of state public budgets (Carew and Stapleton, 2005). In this sense, most developed and developing countries finance their welfare state through taxation and labour contributions. At the same time, inequality and economic uncertainty are igniting the debate on the sustainability of health financing (Liaropoulos and Goranitis, 2015).

Public revenues are the mainstay of public health financing: although private funding can contribute to guarantee health services, public sources account for the largest share of funds available for health purposes (World Health Organization, 2022). However, in order not to jeopardize the level of health care and, thus, the health of society, health systems should be strengthened not by increasing taxation or introducing horizontal cost cuts, but by reducing inefficient health interventions (Thomson et al., 2009).

Consequently, it is important to pay attention to the efficiency of public health, since resources drawn from the community are used for its functioning. In fact, inefficiency deriving from high levels of indebtedness and operating costs reverberate on the community from which the financial resources originate.

Therefore, the aim of this contribution is to analyse the level of efficiency of public health organizations and to investigate a set of factors that could have an impact on it. We chose to employ Italian hospitals in our research because they are particularly well suited to the aims of the analysis, considering the notable heterogeneity in case reporting across regions and variations in management levels within Italy's healthcare system. This is a predominantly public system inspired by the universalist welfare model. Healthcare, therefore, is considered a service of collective interest (Borgonovi, 2005) financed by general taxation. Moreover, the Italian health system is characterized by the role of the regions, which exercise legislative functions in the field of health and hospital care in compliance with the fundamental principles established by the laws of the State and perform their own administrative functions or those delegated to them. The analysis can provide managerial implications useful also in contexts outside Italy that have similar characteristics: settings with public health services, regional organization and mixed governance, where improving hospital efficiency requires attention to managerial dynamics and resource allocation.

With healthcare management occurring at a regional level, significant differences exist among regions (Berta et al., 2010). In addition, alongside the regional actor, there are local managers (at the level of the individual hospital), who also affect healthcare efficiency.

In this paper, we specifically analyse the impact of these aspects on the technical efficiency of hospitals, investigating the link between the different managerial levels involved and the efficiency of Italian hospitals, which is an element little investigated in the literature (Ohrling et al., 2022).

First, we calculate the efficiency score of Italian hospitals using Network DEA analysis (Tone and Tsutsui, 2009; Lo Storto, 2018). Then, applying a regression analysis, we assess hospital efficiency across various management layers, starting from proxies for hospital management quality and extending to higher layers that gauge the quality of local public management. This approach helps us understand the diverse complexities of health service management, examining how the quality of these different actors collectively influences the overall efficiency of the system.

This article offers an innovative contribution by analysing how the managerial quality (at hospital and a regional level) affects hospital efficiency. Unlike most existing literature, which focuses on quantitative environmental or institutional factors (Alatawi et al., 2020), this study introduces a novel governance perspective by linking regional decision-making profiles to hospital efficiency. The analysis allows for a multidimensional assessment of efficiency rarely explored in prior research and the applied multi-level approach represents a clear advancement in the field of public management and health efficiency studies.

The contribution is structured as follows. The next section contains the literature review from which the research questions arise. The third section is dedicated to the description of the methodology used. The fourth section contains the results obtained from the analysis performed. Finally, the fifth section develops a discussion of the results found and some concluding remarks in terms of policy and managerial implications, further research opportunities and limitation.

Managing the health system appropriately allows the creation of efficient health services capable of providing adequate care (Zakowska and Godycki-Cwirko, 2020). The progressive expansion of the right to health services has generated a growing debate on the appropriate level of efficiency of the health service (Lobo and Araujo, 2017; Bahrami et al., 2018). In the upcoming literature review, we will explore why robust management is essential for hospital efficiency. Additionally, we will remark the importance of coordinating different actors, spanning from specific hospital management to the broader context of local governance. The literature has widely affirmed the importance of measuring health system performance as an important public policy issue (Ginsburg, 2003; Geiger et al., 2019).

Chowdhury et al. (2014) thoroughly explored the importance of providing reliable empirical research, given that “health care providers are often regulated or at least closely monitored by government agencies charged with ensuring that taxpayer funds are used efficiently and productively”. Therefore, performance analysis should be useful to promote innovation and improvements.

Flood and Rosenberg (2008), in this regard, analytically demonstrated that it is crucial to measure performance in terms of the quality of healthcare services in order to improve the overall healthcare system. Measuring and evaluating the efficiency of health systems has been explored as part of the search for resources to ensure the sustainability of health and social systems (Gavurova et al., 2021).

In a context characterized by globalization processes and pressures on the efficiency of health systems, the analysis and evaluation of the efficiency of healthcare structures assume an increasingly crucial function (Stefko et al., 2018). Indeed, efficiency analyses play an important role in making healthcare managers and policymakers aware of how to achieve efficiency gains (Campanella et al., 2017).

Authors such as Wang and Gao (2017), in assessing the efficiency of hospitals, have focused on the analysis of the reallocation of resources available to hospitals to improve overall healthcare performance. They come to the view that efficiency improvements in this context can be fostered by the way resources (beds, staff, etc.) are allocated.

For any country, public health expenditure is a particular type of social expenditure which, if set at inadequate levels, leads to the provision of poor quality and poorly accessible services. In this sense, De et al. (2012) argue that there are two solutions to improve the accessibility of healthcare: increase the resources allocated to the health sector or improve the efficiency in the use of existing resources. At the same time, in resource-poor settings, it becomes difficult to increase the number of resources available for the health sector. Therefore, it becomes very important to use existing resources efficiently.

In this regard, the contribution of De Nicola et al. (2013) is also relevant, according to which, greater healthcare efficiency can be pursued through an organizational approach. They propose a clear separation between healthcare providers and purchasers, as well as on the freedom of choice of users between public and private providers. Similarly, according to these authors, administrative decentralization from regional governments to local health units is a source of inefficiency.

More recently, Carmo Filho and Borges (2025) conclude that healthcare digitalization offers significant opportunities for process efficiency and cost reduction without compromising service quality.

In any case, according to Kohl et al. (2019), most attempts in the literature to analyse the efficiency of the health sector are almost always theoretical attempts. Policymakers, economists and managers hardly ever actually rely on the results of the efficiency analyses conducted. Hence, there is a need for closer cooperation between the scientific community and healthcare management in order to verify the results and assess the degree to which efficiency improvements have been realized on the basis of the results.

The literature reviewed emphasizes the relevance of assessing the efficiency of public health services, on the careful monitoring of which the very sustainability of health systems depends, given the use of considerable financial resources that health care requires and that come from taxation. Moreover, it emerges from the literature examined how the evaluation of efficiency represents a propaedeutic factor to the analysis of the impact on the same of the managerial levels involved in the management of the health system and, therefore, also in the pursuit of its efficiency.

It follows that the evaluation of efficiency in healthcare is an important topic that, however, may have certain limitations if it is not complemented by an analysis of the factors that in practice affect the level of efficiency achieved. In this sense, the role of research should be to provide policymakers and managers with useful suggestions and recommendations to improve efficiency in healthcare. It is therefore worthy of investigation the assessment of the main multi-level managerial drivers producing the result. Therefore, we aim to identify the “quality behind every quantity”. In healthcare systems around the world, hospitals are the main cost driver and face increasing pressure to improve efficiency (Kohl et al., 2019).

Consequently, the first managerial level that can be considered is the specific hospital management system. The literature provides some elements relating to efficiency hospital management (Table 1) that are recalled below.

Firstly, the health sector has a high level of human capital. Consequently, it seems appropriate to assume that personnel have a major impact on the degree of efficiency of hospitals.

Al-Amin and Li (2020) reported that understanding appropriate staffing levels is an essential prerequisite for pursuing a higher degree of efficiency. Their investigation found the existence of a significant relationship between hospital staffing levels and hospital performance, demonstrating that staffing levels can be a significant predictor of hospital performance.

Of particular interest in this regard is the study conducted by Khomami and Rustomfram (2019) according to which efficiency in healthcare is not only about the effective use of resources but also includes competent care, appropriate use of technology and effective interpersonal relationships with patients. A hospital practice that takes these aspects into account, according to the authors, is a harbinger of efficient care, with a reduction in the length of time patients spend in hospital, thereby generating savings in the costs incurred by hospitals in providing healthcare. At the same time, it should be pointed out that professional managers (who are entrusted with the task of controlling health care costs, monitoring the allocation of human resources and ensuring their efficient utilization) may be inclined to take decisions aimed at reducing healthcare costs by downsizing the nursing staff contingent or reducing the recognized salaries: this leads to an increase in workloads, dissatisfaction of the staff on duty and, consequently, to a lowering of the level of care offered to patients.

Mandal (2018) also confirms the importance of human capital as an essential organizational resource that can lead to positive performance as it can lead healthcare organizations to provide faster responses to patients' treatment needs.

More recently, Orsal and Uckun (2022) identified a positive correlation between the Intellectual Value Added Coefficient and the efficiency of a university hospital. The authors highlight that increasing intellectual resources and the use of high-level skills significantly contribute to improving both efficiency and operational sustainability in the context of public health management. In particular, their findings suggest that the ability to translate internal knowledge and skills into tangible economic value is associated with measurable improvements in operational efficiency and productivity, providing useful implications for strategic management and cost analysis strategies in complex facilities like university hospitals.

On the other hand, Alexander et al. (1994) examined the relationship between the turnover of hospital nurses and the production efficiency of hospitals. They argued that although the literature presents some arguments in favour of the potential positive effects of a certain degree of turnover (such as innovation in the working environment), excessively high levels of staff turnover may reduce efficiency due to the costs of recruiting, training and supervising new staff members.

An element of particular relevance regarding staff concerns their gender (Rao and Tilt, 2016; Furtado et al., 2021). This in fact can be correlated with the factors previously identified (e.g. propensity to develop interpersonal relationships with patients or the use of technology), thus proving to be transversal to them and impacting on performance (Ali and Konrad, 2017; Ferrary and Déo, 2023).

The correlation between gender diversity and hospital performance is an aspect that has still been investigated to a limited extent (Klerk and Singh, 2021) and, therefore, a novel element of this article.

Secondly, the literature also reveals the importance of considering the complexity of the different activities and healthcare services of hospitals as an additional factor impacting on the efficiency of hospitals.

All healthcare institutions admit and subsequently discharge numerous cases with different diagnoses, levels of complexity and conditions: the relative proportion of each type of case is referred to as the case-mix, understood as the clinical composition of the inpatient population (Pettengill and Vertrees, 1982). The availability and use of clinical and administrative data have made it possible to better understand the relationship between case-mix and healthcare resource utilization (Costa et al., 2015).

An efficient healthcare service allows for good quality service and cost containment. From this perspective, the case-mix system (a generic term used to describe the mix of patients present in a given healthcare setting) appears to be a good option to achieve this efficient system (Moshiri et al., 2010).

Chowdhury et al. (2014) contributed to the literature by considering the case-mix (the complexity of medical treatments provided by hospitals) for their empirical analysis, showing that models provide different results with and without this element.

Similarly, other contributions (Grosskopf and Valdmanis, 1993; Faye, 2014) represent examples of studies that have used the effect of differences in patient case-mix to measure hospital performance, comparing performance measures with and without adjustment for case-mix. In order to deepen the analysis of the elements that influence hospital efficiency, it is therefore essential to understand their nature. In fact, according to the arguments of Nissi and Sarra (2018), each hospital presents a different level of specialization and can be configured as a complex system that operates through a multitude of interactions between different factors within a complex environment such as the national healthcare system.

A further element representing the complexity that hospitals have to deal with is the differentiation of patients between the various diagnosis-related groups (DRGs). DRGs are a nominal scale for identifying categories or types of patients similar in terms of resource intensity and clinical relevance (Colivicchi et al., 2024).

The heterogeneity of the distribution of discharges across various DRGs is typically quantified through the entropy index (Ugolini and Fabbri, 1998; Bertoli and Grembi, 2019). It is an absolute index that measures the heterogeneity of the distribution of discharges across various DRGs (Colivicchi et al., 2024).

Building on the strands above and explicitly integrating Table 1 evidence, our study contributes in three ways to current literature. First, we conceptualize hospital efficiency not only as a function of resource endowments but also as the outcome of dynamic managerial capabilities and process-management coordinating clinical units. Recent studies show that capability-driven redesign enhances performance (Tenggono et al., 2025; Rosenbäck and Eriksson, 2024; Lima et al., 2024; Al Harbi et al., 2024). Within human capital, adequate staffing emerges as a managerial capability: eliminating understaffing is cost-effective and linked to lower mortality and higher efficiency, whereas reliance on agency staff and unstable skill-mix undermines quality (Saville et al., 2025; Beauvais et al., 2024; Bae, 2024; Clarke and DePesa, 2025). Gender composition further appears as a lever for engagement and performance (Adam and Alfawaz, 2025; Obeng et al., 2025; Sarpong-Danquah and Osman, 2025).

Second, by distinguishing resource and process efficiency through a two-stage network DEA (NDEA), we clarify how quantities and quality of actions interact. In this case, case-mix plays a role (de Mattia et al., 2024; García-Lorenzo et al., 2024; Pai et al., 2024). Post-pandemic evidence confirms a higher CMI baseline alongside renewed mortality improvements, reinforcing the need for acuity adjustment (De Rienzo et al., 2025), in line with broader sustainability work (Gavurová et al., 2024). We further relate DRG heterogeneity to entropy-based organizational design capturing portfolio variety and coordination needs (Winasti et al., 2023; Klemann et al., 2024).

Third, within Italy's multi-level SSN governance, institutional quality (rule of law, regulatory quality, accountability) conditions managerial practices and process performance. Persistent territorial heterogeneity underscores the need for stronger alignment across levels (European Observatory on Health Systems and Policies, 2024; Lega and Prenestini, 2025). Accordingly, our framework tests how governance quality and managerial levers – staffing and composition, process design and case-mix/entropy – jointly explain the “quality behind quantity” shaping hospital efficiency.

These elements can be traced back to the quality of hospital management (Örnerheim and Triantafillou, 2016), and in order to test through empirical evidence what has emerged so far, the literature analysed suggests the following research question:

A step forward can be taken by considering another key player in determining hospital efficiency: the local management where institutions can be considered crucial in setting the rule of the game (North, 1991) and then in driving the efficiency of hospital services. What has been said so far, highlights that among the factors that potentially impact on the efficiency of public health, those relating to the managerial elements are of importance. Together with the elements that refer to the quality of hospital management (Örnerheim and Triantafillou, 2016), it should be not overlook that the managerial quality of institutions outside the hospital $−$ whose activity is influential on the delivery of healthcare services – can also impact its efficiency. For example, in the Italian context – examined in this paper – public healthcare is managed at a regional level: consequently, the managerial profile of the local authorities invested with decision-making powers in the healthcare sector may have repercussions on the efficiency of individual hospitals.

The analysis of this aspect represents an innovative element in this study, underscoring the originality of examining local management as a determinant of hospital efficiency. By shifting the focus from traditional external factors to the managerial profile of local authorities with decision-making power in healthcare, the paper contributes a novel perspective on how governance at the regional and local level can shape service performance. While the literature often emphasizes environmental and quantitative factors, our approach highlights qualitative dimensions – such as staff characteristics, service complexity, and the managerial capacities of non-hospital institutions – that influence efficiency outcomes. In the Italian context, where public healthcare is organized regionally, this innovative angle can reveals how the governance quality of local authorities reverberates on individual hospital performance, offering new insights beyond the mainstream focus of prior studies (Alatawi et al., 2020; Ghahremanloo et al., 2020; Yousefi Nayer et al., 2022). In short, this study advances the discourse by suggesting that efficiency in health systems is not only a matter of internal hospital management or external environment, but could be significantly affected by the managerial profiles of the surrounding institutions and the qualitative attributes of their staff and operations. This approach also allows us to link theory to practical political and managerial implications.

Furthermore, apart from Lo Storto (2016) who measured the efficiency of Italian municipalities, to our knowledge the analysis of the impact of factors on the efficiency of Italian hospitals is a relatively unexplored objective within the public health management literature.

In order to help bridge this gap – expanding the RQ1 – we can include the institutional management related to hospital efficiency, formulating a second research question.

In what follows, the reader will find the main passages of our analysis. The ultimate goal of the paper is to investigate the contextual determinants of Italian hospital efficiency in light of different levels of management, from hospital-specific metrics to local government quality. The data included in the analysis are derived from various sources using the most recent available data from 2019: (1) the Databank of the Ministry of Health (https://www.salute.gov.it/portale/documentazione/p6_2_8_1.jsp?lingua=italiano), (2) the ISTAT portal (https://www.istat.it/dati/banche-dati/) and (3) the institutional quality index dataset (https://sites.google.com/site/institutionalqualityindex/home). This results in a final dataset of 513 observations reporting hospital-level information across the whole Italian Territory (see Supplementary material for more details).

Following the most recent literature, this paper employs the two-stage NDEA (Kawaguchi et al., 2014; Khushalani and Ozcan, 2017; Alfonso et al., 2024), hereafter referred to as NDEA. Unlike traditional non-parametric DEA methods (Jung et al., 2023), this approach considers the interrelationships among the different sub-units within the observed Decision-Making Unit (DMU). Specifically, hospital services can be viewed as sequential organizational processes, where the first stage evaluates the effectiveness of resource management, while the second assesses the efficiency of service delivery, given the resources already utilized. This two-stage framework offers the advantage of accounting for the sequential nature of hospital processes, as well as providing distinct efficiency measures for each stage.

In line with Alfonso et al. (2024), we hypothesize a first stage of NDEA focused on the use of physical capital (bed capacity) and labour (doctors, nurses and other staff), examining performance relative to the hospital's activity level. We use proxies such as the number of beds occupied, admissions and surgeries performed to evaluate which hospital unit achieves the highest efficiency in utilizing resources at a given activity intensity. Thus, the first efficiency measure is termed resource efficiency.

The second stage examines efficiency in managing the hospital processes, considering the activity level achieved in the first stage as an intermediate output. Here, the focus is on the number of discharged patients and the speed of achieving this outcome. This measure is labelled process efficiency, as it pertains to how resources are applied to enhance service management efficiency, rather than resource intensity.

In the second level of the analysis, both resource and process efficiency indicators are used to measure overall Hospital Efficiency, incorporating various managerial aspects at multiple levels. The first level concerns hospital-specific management, with key indices such as the case-mix index, which reflects the complexity of treated cases and managerial decisions regarding specialization within the hospital. The entropy index, which measures the diversity of cases across departments, also plays a role, indicating whether the hospital pursues economies of scale (low entropy) or economies of scope (high entropy).

At the provincial and regional levels, we include indicators related to the quality of management within local institutions (Nifo and Vecchione, 2014), reflecting the public sector dynamics of the healthcare sector.

To ensure replicability and transparency, we provide additional details on our empirical implementation. The final dataset consists of 513 public hospitals covering all Italian regions in 2019, selected from the Databank of the Ministry of Health after excluding hospitals with missing values for key variables. Each hospital represents a DMU. The analysis uses a two-stage NDEA model, implemented using an output-oriented variable returns to scale (VRS) assumption. This approach models hospital operations as a two-stage process, with the first stage measuring Resource Efficiency – based on input variables such as number of doctors, nurses, other staff and total bed capacity – and the second stage measuring Process Efficiency – based on outputs such as discharged patients and hospitalization days, using the intermediate outputs (beds used, surgeries, and admissions) linking the two stages. All variables were selected following a comprehensive literature review (see Supplementary Material), in particular drawing from Fazria and Dhamayanti (2021) and were normalized to ensure consistency across units. The NDEA model was implemented using R, with model code and data processing procedures available upon request to ensure full replicability.

Figure 1 outlines the roadmap of the analysis conducted. Further details regarding the methodology and variables are provided in the next subsections.

As anticipated, the efficiency analysis is conducted by using the Databank of the Ministry of Health.

To decide which input, intermediate and output variables to introduce into our model, we started from the study conducted by Fazria and Dhamayanti (2021), which analyses the variables most often used to assess efficiency in hospitals when applying the DEA method. The results of this study show that the five most frequently used inputs in the literature are the number of beds, medical staff, non-medical staff, medical technical staff and operating costs. While the most frequently used outputs are the number of inpatients, surgeries, emergency visits, outpatient services and inpatient days.

The input variable “number of beds”, which among other things is considered by part of the literature as a proxy for hospital capital, is in fact used very often in the literature (Nepomuceno et al., 2018, 2020; Cavalcante et al., 2017; Omrani et al., 2018), as is the input variable “number of medical staff” (Stefko et al., 2018).

In particular, the input variable “doctors” finds many applications (Azreena et al., 2018; Stefko et al., 2018; Kounetas and Papathanassopoulos, 2013), just as the variable “nurses” is just as often used (Vafaee Najar et al., 2018; Azreena et al., 2018; Stefko et al., 2018; Kounetas and Papathanassopoulos, 2013). All this also applies to the input variable “other members’ (Lee et al., 2023; Azreena et al., 2018; Stefko et al., 2018; Kounetas and Papathanassopoulos, 2013).

As specified in the study of Fazria and Dhamayanti (2021) also the intermediate and output variables chosen in our work are also widely used in the literature to assess the efficiency of the hospital sector.

Specifically, the intermediate variable “beds used” (See et al., 2024; Pinto, 2016); the variable “surgery treatment” (Mosadeghrad et al., 2017; Hofmarcher et al., 2002) and the variable “hospitalization number” (Hasni et al., 2022; Gurrieri and Lorizio, 2016).

The output variables chosen in our paper are the “discharged patients” (Azreena et al., 2018; Ferrier and Trivitt, 2013) and, finally, the variable “hospitalization days” (González-de-Julián et al., 2021; Eckmann et al., 2013).

More details will be provided in Tables 2 and 3.

As defined above, the two-stage NDEA used analyses the efficiency of hospitals in two stages. To this end, we adapt the conceptual framework of previous studies using NDEA to investigate the two-stage efficiency of hospitals (Alfonso et al., 2024). In the first stage, efficiency in resource usage (physical capital and labour) is assessed in relation to the level of activity, such as the number of beds occupied, admissions and procedures performed. This is referred to as resource efficiency. In the second stage, efficiency in hospital service management is examined, using the activity from the first stage as input and the number of discharged patients and the speed of discharge as output. This stage is referred to as process efficiency.

In the first stage, input variables include: bed capacity (beds), number of doctors (doctors), number of nurses (nurses) and other staff members (staff others). Intermediate variables (first-stage output, second-stage input) include the number of beds used (beds used), the number of surgeries performed (surgery treatments) and the number of hospitalizations (hospitalizations number). These variables become input when, given a specific level of hospital activity intensity (which these variables are proxies for), we study efficiency in the management process. In the second stage, the final output is the average length of hospitalization, with the variable flipped since longer average stays may indicate greater inefficiency in service. Flipping the variable, we interpret an increase in value as indicating faster service. Additionally, the number of discharged patients (discharged patients) serves as an indicator of the effectiveness of the care provided.

We use an input-oriented two-stage NDEA with VRS. As previously mentioned, we consider two stages of efficiency.

The two DEA-derived efficiency indices (resource efficiency and process efficiency indices) will be used as dependent variable in the second-level regression analysis (Lewis and Linzer, 2005). The main regressors of interests will be introduced in the following step.

The concept of multi-level management in healthcare includes the complex network of decision-making and coordination across different levels of government, institutions and organizations involved in healthcare delivery (Schreyögg et al., 2011). Following this approach, the resulting quality and efficiency of the healthcare services released derive from the interplay between specific healthcare management and institutional entities. The aim of this work is to identify the sensitivity of the hospital efficiency to the quality of the performance of these actors involved in the sector.

At micro-level, the primary contributor and responsible of the healthcare quality is the specific hospital management. A proper hospital management is crucial for guaranteeing an adequate quality of care (Jha and Epstein, 2010). It influences efficient resource allocation, regulatory compliance, transparency, ethics, financial stability and risk management. A well-governed hospital is more likely to provide safe, effective and high-quality care to its patients. From an empirical perspective, the assessment of hospital management is a challenging task due to the lack of available data. That is, existing literature is basing the assessment and the evaluation by mostly employing survey data (Walsh et al., 2020), or, alternatively, performance indicators. In this context, we opt for the second strategy by introducing three main metrics that can characterize the vision and the organizational efficiency of each healthcare structure. In a broader sense, we consider the management including different aspects. With the first two indicators, namely the Case-Mix Index and the Entropy Index, we consider the specialization and the quality of care, while with the staff gender diversity we account for the sociological aspects of workplace.

Firstly, we introduce the Hospital Case-Mix Index (Hornbrook, 1982) that is a positive numerical value used in healthcare to represent the diversity and complexity of patient cases treated within a hospital. Therefore, higher values might be proxy of higher efficiency in the management given the higher ability to manage different cases favouring the accessibility to different types of care.

Secondly, we introduce the Hospital Entropy Index, which has been less employed in existing literature but may provide valuable insights into the management strategy of hospital administration. In fact, as reported in Bertoli and Grembi (2019), the entropy index is a composite indicator reflecting the diversity of cases treated in different departments. A higher value of the indicator suggests greater differentiation in the cases handled, and consequently, a more heterogeneous activity across departments. Therefore, a high value of the indicator identifies a strategy by the hospital board to pursue economies of scope, leveraging the know-how and relationships between different and specialized departments. A low value of the indicator, on the other hand, indicates a strategy aimed at pursuing economies of scale, focusing on specialized care for a specific type of treatment, thereby reducing the marginal cost of treating additional similar cases.

A third factor is the staff gender diversity, meant as the female over male coexistence rate. Literature agreed on the managerial implications of gender staff differences on performance (Ferrary and Deo, 2023), given the importance of open-mindedness and diversity in the improvement of decision-making processes.

Institutions play a vital role in the society, setting the rules of the game and driving the quality and the efficiency of the surrounding economic system (North, 1991). As a result, the quality of institutions can pose a solid basis for efficient public services (Alesina and Tabellini, 2007, 2008). While this is well-known in the provision of different services, the quality of institutions might also enhance the quality of infrastructures jointly determining the economic growth (Zergawu et al., 2020). Therefore, this turns worthy of investigation in the analysis of the healthcare infrastructures (De Luca et al., 2021), given the pivotal role that improvements in public health and healthcare systems play in fostering economic development and sustainable growth (Madsen, 2018), favouring productivity, well-being and less social inequalities. This assumes particular relevance in healthcare systems such as Italy's, the management of which is highly dependent on choices made in regional public administrations. In light of this, we analyse the role that the quality of the local institutions plays in determining the efficiency of healthcare infrastructures. To this end, we introduce the series of institutional quality indexes (IQI) proposed by Nifo and Vecchione (2014). The components of IQI are freely available (https://siepi.org/institutional-quality-index-dataset-disponibile/) and encompass five primary pillars for measuring institutional quality at the provincial level: (1) Voice and accountability, meant as citizen participation in public elections, engagement in civic and social associations; (2) Government quality as a measure of urban and infrastructure development index; (3) Regulatory quality: this dimension examines the openness of the economy, the rate of business closures, various indicators related to the business environment and the density of businesses operating within the provinces; (4) Rule of Law, proxying the efficacy of the legal system, then including the on crimes against individuals and property, productivity of magistrates, trial durations, instances of tax evasion and the size of the shadow economy; (5) the control of corruption identifying the level of crimes against the Public Administration, the number of local administrations overruled by federal authorities. The data distribution can be found in Table 2.

Considering the hospitality efficiency measure and the different proxies of management, we introduce a second-level regression. The regression analysis on the two efficiency scores obtained by applying the DEA method (first stage). This practice has been used in many studies to investigate the impact on efficiency of a number of environmental or contextual factors (Balaguer-Coll et al., 2013; Kalb, 2010) without employing the same set of variables used in the first level to avoid endogeneity. Therefore, we consider the Efficiency Indices obtained in the first level as the dependent variable and identify different independent variables characterizing the quality of multilevel management, which are shown in Table 3. The methodology is based on the premise that efficiency scores derived from quantitative inputs and outputs through the NDEA model can reflect both operational performance and underlying strategic choices. To maintain interpretability and practical relevance, we adopt a second-stage regression to explore how qualitative factors influence efficiency in a clear and additive way.

From Table 2, it can be also observed that we account for demand factors including the population density calculated at municipality level considering the population over the territorial surface (ISTAT data). For completeness, we remember that healthcare data are sourced from the databank of the Health Ministry, while the Institutional Quality data are downloaded from the online portal provided by Nifo and Vecchione (2014). In this case, we consider the most disaggregated data at provincial level, while we avoid the use of the aggregate regional metrics due to collinearity issues with both the provincial ones and the macro-area controls.

In this section, we will present the results following the structure outlined in Section 3 and Figure 1. Initially, we will display the outcomes of the NDEA followed by the results pertaining to the regression models.

The map depicted in Figure 2 illustrates the average efficiency indexes among various provinces. Remarkably high levels of efficiency are evident, with slightly lower levels observed in central Italy. The distribution appears to be skewed towards higher and more uniform average levels in northern Italy. This observation aligns with the information presented in Table 4, where average values for macro areas are detailed. The data suggest a consistent pattern of generally high and evenly distributed efficiency across north provinces, contrasting with the moderately lower yet still noteworthy efficiency levels in central Italy. The spatial illustration on the map and the values of Table 3 jointly describe the main results of NDEA analysis providing interesting heterogeneity that will be analysed in the regression analysis. In the Supplementary material, there are hospital-specific efficiency scores.

Table 5 reports the main results of the regression analysis. Column 1 presents the results using resource efficiency as the dependent variable, while Column 2 reports the results using process efficiency. The baseline model employs the most disaggregated provincial institutional quality indicator from 2019, and the analysis is repeated in Columns 3 and 4 using the 2009 indicator as a robustness check to avoid endogeneity related to simultaneity (hospital efficiency influencing regional quality and vice versa). This allows for the examination of the direction of the effect of past institutional quality improvements on current hospital efficiency. In this case, it cannot be assumed that current hospital efficiency causes an improvement in past institutional quality.

As can be observed, among the three proposed indicators, the most significant effects are attributed to the case-mix index (Hospital_icm) and the entropy index. In fact, when examining the relationship with resource efficiency, we find that these two indicators have a positive impact. Therefore, we can conclude that a hospital management approach focused on achieving economies of scope, promoting service diversification and fostering greater synergies and knowledge flows among specialized staff, leads to efficiency gains (Schiavone, 2008).

Regarding process efficiency, the effect of the two variables is opposite: the Case-Mix Index has a negative impact on process efficiency. This can be explained by attributing a loss of efficiency in terms of patient treatment speed due to the higher complexity of the cases being handled. However, the presence of economies of scope (increased entropy index) has a positive impact in this case as well.

The relationship with the quality of institutions within the territory is more complex and deserves further attention.

As anticipated, the index of Nifo and Vecchione (2014) can be further disentangled into its five components. This can help us in identifying the most impactful dimension. Considering the high correlation among sub-indices, we conduct the analysis with separate regression models, and the related coefficient is reported in Figure 3.

We present the analysis at both regional and provincial levels, using data from both 2019 and 2009, following the same framework proposed in the regressions.

The results reveal several important findings. First, a greater impact of regional quality is observed, regardless of the reference year, indicating that institutional quality consistently influences hospital efficiency over time. Additionally, the results show heterogeneity in the effect of the various sub-indices, suggesting that not all regional and provincial quality indicators affect hospital efficiency in the same way.

Considering the impact of lagged variables, a stronger effect of institutional variables is noted, with a higher number of cases showing statistical significance. This supports the causal link hypothesis, where prior institutional improvements lead to better hospital management. Finally, the impact is more pronounced in process efficiency than in resource efficiency, suggesting that changes in hospital processes are more critical to improving efficiency than changes in resource use.

The quality of the administrative and legal system is therefore paramount for improving organizational efficiency. This can be due to several reasons, since a proper legal framework, less crime and more political stability can ensure safety, patient right protection and quality of care. Following this line, timely and fair conflicts resolutions can prevent delays than promoting efficiency, and it can be also possible to efficiently manage bureaucracy. From a behavioural perspective, the efficiency of the local government can lead to more public trust and then to higher confidence in the healthcare system. This poises the Institutional Quality as one of the key factors for the economic development (Haggard et al., 2008; Roxas et al., 2012).

Following the hypotheses, the results are summarized below.

RQ1: The identified managerial variables show interesting (and sometimes contrasting) effects depending on how efficiency is defined. This has significant implications for managerial decisions related to operational strategies and for selecting appropriate performance indicators. Specifically, choosing to pursue economies of scope (as indicated by a high entropy index) consistently results in improved efficiency, regardless of the approach taken. Additionally, a willingness to take on more complex cases positively influences resource management efficiency, though it can slow down process speed.

RQ2: Our observation indicates elevated levels of efficiency, with slightly lower levels noted for Central Italy and slightly higher for North-East Italy. The quality of institutions at the regional land local level has a greater impact in explaining the efficiency of processes. Therefore, regional disparities might increase inequalities in the healthcare access.

To close the loop between theory and evidence, we map the constructs introduced – dynamic managerial capabilities and multi-level governance – onto our two-stage outcomes. The consistently positive effect of the entropy index (Table 5) operationalizes economies of scope: portfolio variety combined with cross-department integration enhances performance by enabling knowledge sharing and flexible capacity. By contrast, the case-mix index reveals a boundary condition of capability theory: greater clinical complexity raises resource efficiency but can depress process speed, motivating acuity-adjusted pathways, proactive discharge planning and throughput management. At the system level, provincial/regional institutional quality primarily lifts process efficiency, extending governance theory by showing how rules, accountability and coordination act as an enabling infrastructure for hospital routines (Figure 3). Taken together, these links advance the “quality behind quantity” proposition and translate into actionable levers – stable core staffing and skill-mix, portfolio design to improve scope and hospital–region alignment on joint Key Performance Indicators – to improve management practice. In what follows, we discuss more in details the results found.

The healthcare system consists of a different hierarchical operational structure and different factors and indicators that can monitor its performance. This paper investigates the contextual determinants of hospital efficiency in Italy, focusing on the interplay between managerial levels and their influence on healthcare provision. By employing two-stage NDEA and regression analysis, the study finds a positive correlation between the quality of regional and local management and hospital efficiency, highlighting the critical role of institutional governance in enhancing healthcare performance. This confirms the findings of other studies, as in the paper of Johannessen et al. (2017), where the authors examined the impact of the centralization of the Norwegian hospital sector since 2002, which resulted in increased hospital efficiency.

Results underscore the importance of the case-mix (which allows hospitals to be ranked according to efficiency and productivity parameters (Chowdhury et al., 2014)) and Entropy (which is a very suitable index for measuring the necessary information and designing the efficiency of organizational structure (Winasti et al., 2023) indices in determining efficiency, while revealing that higher-quality regional institutions significantly contribute to improve hospital outcomes. These findings align with the health-growth nexus, suggesting that strong institutional contexts foster both healthcare efficiency and broader economic and social prosperity, a finding that is consistent with the research conducted by Sethi et al. (2020), who found that in South Asian countries (between 1996 and 2018), institutional quality had a positive effect on healthcare spending.

An important and only apparently trivial concept emerges from the analysis conducted: quality matters in defining the quantity and efficiency of the provision of health services. This trade-off is also confirmed in the literature where, for example, in the work of Navarro-Espigares and Torres (2010) shows that in Andalusian hospitals between 1997 and 2004, there was a positive correlation between efficiency and quality indicators. This seemingly simple finding hides within it several levels of complexity that are analysed and discussed in the paper. In fact, we observed high level of heterogeneity in the metrics proposed.

Considering the hospital management indexes, we observed a strong highly significant effect related to the case-mix and entropy indices, underlying that the quality of the care released is closely related to the efficiency of hospitals. What we have stated is also verified in the literature, in fact it has been shown that the case-mix index was lower in public hospitals than in private ones, as they are less efficient due to having fewer financial resources available (Mendez et al., 2014).

However, more efforts are needed in identifying the social dimensions of management quality. The choice of the gender diversity index, which in other sectors, has proven effective in justifying the positive impact on business efficiency (Santi and Rupjyoti, 2021), in our specific case is not able to capture the variability of efficiency, suggesting the need of further explorations of the social aspects of workplace and, more generally, of the working conditions that can improve hospitals productivity.

Another interesting result is related to the quality of underlying local institutions. Although the overall indicator of quality of institutions is positively correlated with efficiency, the decomposition of the indicator leads to several interesting results.

In line with regional development theory (Stough, 2001), our findings indicate that the quality of institutions significantly enhances the efficiency of resource allocation and hospital processes. Effective governance and decision-making at the regional level contribute to improved healthcare outcomes, highlighting the importance of institutional quality in managing hospital performance across Italy.

A particularly interesting result obtained from the analysis conducted concerns the relevance of managerial quality at the institutional and hospital levels in developed societies, allowing the development of some concluding remarks in terms of practical implications.

Regarding the policy implications, the institutions–growth nexus considers both economic and political science literature, with a primary focus on the relationship between the rule of law and economic growth (Barro, 1996, 2000). From the pioneering work of Acemoglu et al. (2022), the main ingredient of this theory is the idea that government inefficiency, corruption and instability is the primary constraint on economic growth. In this virtuous context, efficiency is more likely to be observed in various public service delivery contexts and in the specific case of health services. The latter prove to be extremely important for health, prosperity and economic growth, as evidenced by various studies and economic theories regarding health-led growth (Madsen, 2018). Thus, the idea is that the “institutions–growth” nexus and the “health–growth” nexus might converge into a broader “institutions–health–growth” nexus where the quality of the legal and public system can amplify the health-led economic growth and social prosperity. Incorporating institutional quality into health-led growth theories suggests that strong institutions play a key role in improving hospital efficiency and health outcomes. Effective governance and sound healthcare policies lead to better resource management, enhancing hospital performance and patient care. A healthier population, resulting from high-quality institutions, boosts workforce productivity, reduces healthcare costs and drives economic growth. This framework highlights how institutional improvements in the healthcare sector contribute to both better health outcomes and sustainable economic development. Thus, the relationship between institutional quality, health and economic growth is cyclical, reinforcing each other over time.

The result obtained on a national scale is particularly interesting because the country under analysis (Italy) exhibits significant within-country heterogeneity. Future research could extend the findings in several directions. The first, in a narrow sense, concerns the analysis of efficiency. As demonstrated in several studies referenced in this work, the efficiency analysis of a complex structure like hospitals can encompass a multitude of aspects related to the very definition of efficiency. This can not only align with the standard production theories focused on resources and processes but also incorporate equity concepts such as the speed and ease of access to care.

This study thus opens up interesting avenues for using more advanced methodologies, like NDEA with second-level analysis, which include broader characteristics and aim to connect the internal dynamics of the hospital sector with broader socio-institutional objectives. It therefore becomes important to understand how the different facets of hospital sector efficiency might influence and be influenced by the socioeconomic dynamics of a country.

Regarding the managerial implications, the findings lead to the conclusion that ever greater efforts should be made to improve managerial quality. This could be done, for example, through appropriate training paths (Ravaghi et al., 2021). Training both physicians (involved in hospital management) and policymakers (whose choices impact health services) on managerial issues would enable healthcare management at different levels to learn principles and tools useful for pursuing efficiency.

In addition, in our opinion, the analysis suggests a reassessment of managerial networking (Song et al., 2021): closer cooperation between hospital managers and policymakers is considered essential so that they know each other's needs and critical issues. Only in-depth knowledge on the part of politicians about the critical operational issues that need to be addressed by hospital managers, and, speculatively, knowledge on the part of hospital managers about the constraints and opportunities that need to be addressed politically, allows for real (institutional-hospital) convergence with a view to improving healthcare efficiency, for the ultimate benefit of the community.

From the analysis emerges the need to a further focused investigation of the qualitative social indicators that take into account the diversity of the operational structure on different levels. It would be interesting to further investigate and analyse the result related to gender diversity and extend the concept of diversity of sociological characteristics to include other different factors regarding workplace relationships (Danna and Griffin, 1999; Zarrin et al., 2022).

Additionally, being a developed country, it becomes interesting to develop new research extending the result obtained on a global scale by distinguishing between developed and developing countries.

The main limitation of this paper is that the empirical analysis is confined to a single fiscal year, 2019, due to data availability. This narrow time frame, however, offers a deliberate advantage: it allows the study to examine a pre-COVID period, thereby avoiding distortions caused by the extraordinary shock associated with the pandemic. As a result, the data and the resulting findings can be interpreted without the confounding effects of the crisis, providing a cleaner baseline from which to assess underlying relationships. That said, this restriction also narrows the scope of the conclusions and may limit the generalizability to other periods or to post-pandemic conditions. To address this, future research could extend the analysis to additional years.

The supplementary material for this article can be found online.