Comparative Assessment of Health Systems Resilience: A Cross-Country Analysis Using Key Performance Indicators

Abstract

Although organizational resilience is well established, refining the systematic quantitative evaluation of health systems resilience (HSR) remains an ongoing opportunity for advancement. Research either focuses on individual HSR indicators, such as social welfare policy, public expenditure, health insurance, healthcare quality, and technology, or broadly examines socio-economic factors, highlighting the need for a more comprehensive methodological approach. This study employed the Slacks-Based Measure (SBM) within Data Envelopment Analysis (DEA) to analyze efficiency by maximizing outputs. It systematically examined key HSR factors across countries, providing insights for improved policymaking and resource allocation. Taking a five-year (2016–2020) dataset that covered 55 to 56 countries and evaluating 17 indicators across governance, health systems, and economic aspects, the paper presents that all sixteen top-ranked countries with a perfect efficiency score of 1 belonged to the high-income group, with ten in Europe, highlighting regional HSR differences. This paper concludes that adequate economic resources form the foundation of HSR and ensure stability and sustained progress. A properly supported healthcare workforce is essential for significantly enhancing health systems and delivering quality care. Last, effective governance and the equitable allocation of resources are crucial for fostering sustainable development and strengthening HSR.

1. Introduction

1.1. Current Pressures and the “Black Swan Era” on Healthcare Systems

With the advent of an aging society worldwide, the demand for health care for non-communicable diseases has surged dramatically. At the same time, global climate deteriorations accompanied by more frequent extreme weather events like floods, heavy rains, droughts, high temperature, etc., have further strained most countries’ healthcare service sectors since those events generally not only change in disease transmission routes but also incur more prevalent secondary air pollutants due to ozone damage and denser suspended particulates. This results in greater respiratory and cardiovascular diseases or even indirect damage to the population’s mental status, thereby increasing the needs for healthcare services [1,2,3,4,5]. The COVID-19 pandemic has given rise to over 760 million confirmed cases and at least 6.9 million deaths worldwide since December 2019 [6]. The pandemic negatively impacted numerous vulnerable health systems globally.

Society and the financial markets are now seeing more so-called black swan events, in which unexpected and unforeseen occurrences are happening more and more often and increasingly becoming the normality. With the world progressively moving into an era of heightened volatility, uncertainty, and complexity, health systems must prepare themselves for the worst-case scenario. This is especially true with regard to building up national health systems resilience in advance. Recent disruptive events increasingly highlight the emergence of global volatility as the new norm. The COVID-19 pandemic revealed deep systemic vulnerabilities and continues to strain healthcare systems and supply chains. Climate change has amplified the frequency and severity of natural disasters, as seen in the devastating 2024 floods across central Asia, Nigeria, and Europe, followed in 2025 by record-breaking floods in Hunan and Guizhou provinces in China and an unprecedented heat dome in North America that severely tested infrastructure resilience. Geopolitical shocks, such as the prolonged Russia against Ukraine conflict and the 2025 “Twelve-Day War” between Israel and Iran, have disrupted global energy and food security, underscoring the speed at which regional tensions can escalate with global ramifications. The escalation of cyber threats is also putting critical digital systems at risk. Financial volatility has concurrently intensified, as exemplified by sweeping tariff proposals introduced in 2025 by U.S. President Trump that triggered fears of prices going up, slow growth, and widespread concerns about global economic stability. Once exceptional, such disruptions are now recurrent features of the global landscape, reinforcing the urgent need for resilience strategies that extend beyond infrastructure to include adaptive mindsets, integrated planning, and agile governance capable of withstanding systemic shocks.

1.2. Defining and Conceptualizing Health Systems Resilience (HSR)

The term “resilience” is quite popular nowadays and used in various fields with divergent aspects, including military, economic, technological, social, and cultural, to name a few. According to the British Standard Institution’s (BSI’s) standard BS 65000, organizational resilience denotes “the ability of an organization to anticipate, prepare for, respond and adapt to incremental change and sudden disruptions in order to survive and prosper” [7]. For health systems, resilience should necessarily take into consideration all aspects of resilience above.

The Organization for Economic Cooperation and Development (OECD) has been helping its members to think systemically, to collaborate with each other, and to anticipate and respond to crises through resilient health systems that are prepared for sudden shocks and disruption (such as pandemics, economic crises, or the aftermath of climate change), are capable of minimizing the negative consequences, and are able to recover sooner from crises [8]. However, health systems nowadays are more stressed than ever in the aftermath of the COVID-19 pandemic, geopolitical tension, extreme weather conditions, an aging society, economic turmoil, and severe shortages of healthcare professionals, as well as medicine shortages due to supply chain issues. Health systems are thus facing enormous pressure and challenges. In view of the enormity and complexity of the issues involved and for the health systems to perform well, the agile resilience required by modern societies under these conditions demand and require health systems, healthcare professionals, government, and citizens to all heighten their adaptability by formulating strategies founded on better observations, monitoring, predictions, learning, and other resilience capabilities [1,2,3,4,5,9,10,11,12].

To identify and ascertain the factors most influential to HSR capabilities, we address the following research questions through quantitative statistical analysis: (1) Do high-income countries have higher HSR scores? (2) Do European countries have higher HSR scores than others? (3) Does a country with ample healthcare workers have higher disaggregated efficiency? (4) Does a country with better governance have higher disaggregated efficiency?

1.3. Methodologies for Assessing HSR

Efficiency analysis by data envelopment analysis (DEA) has become a popular quantitative analyzing technique for comparing performances between and among countries in their evaluation and calibration of policy implementation. However, few empirical studies have gone beyond aggregated performances to provide insight into disaggregation efficiency. Scant papers have established the status of every indicator’s optimization in spite of the fact that aggregation and disaggregation analysis techniques are valuable tools for manipulating data and determining the appropriate policies to employ. Assuming that this type of model can be traced and examined to check the output efficiency of resources applied, it can help inform and focus future efforts to strengthen HSR by monitoring beneficial strategies from different aspects and via more efficient allocation of various resources [13,14,15].

1.4. Key Determinants of HSR

Different conditions and resources in today’s world, such as family, neighborhood, society, city, and government, intricately influence health outcomes. In recent years, there has been a proliferation of research on public health, inspiring several multidimensional measuring indicators [16,17,18,19,20,21]. Some papers have targeted the overall healthcare system, public expenditure, healthcare expenditure [22], government social welfare policies [23,24], maternal and child health services, health-related services, health insurance coverage [25], life expectancy at birth [26,27,28,29], mortality rate [27,28], maternal mortality rate [16], infant/child mortality rate [30], suicide mortality rate, specific vaccines, advanced health technology, new drugs, environmental factors (clean water, air pollution, or public hygiene), social support network, and more. Others have looked at individual socioeconomic status (SES) [17,26,31] regular physical activity, improving nutrition, health education, and so forth.

Life expectancy is a widely accepted key metric for assessing population health. Preston (1975) set up a mechanism and was able to establish a correlation between declining mortality and increased investment in housing, hospitals, and training programs for healthcare professionals. In other words, the higher a population’s income is, the larger the gains in life expectancy are. Preston also examined the relationship between life expectancy and national income per capita for nations in the 1900s, 1930s, and 1960s by using a scatter diagram, later known as the Preston curve, and revealed that those curves gradually flattened out and shifted upward during the 20th century, illustrating that people living in richer countries, on average, live longer than people living in poorer countries. Among variations in average income, there is a positive relationship with much steeper curvature between national income levels and life expectancy in poorer countries, while life expectancy levels in richer countries are less income sensitive. However, the relationship is changing, with life expectancy increasing overtime at all income levels. Although decades have passed since then, the Preston curve is still used in both global public health policy and academic research in the public health field [27,28,32].

Although it has long been known that sufficient resources are the foundation of HSR, the uneven distributions of resources, unfortunately, go far beyond the lack of them. Horizontal inequalities (HIs) refer to disparities between groups in several key areas: access to and ownership of economic assets; availability and quality of social services such as education, healthcare, and related benefits; political representation and influence; and the societal recognition of languages, customs, traditions, and cultural identities. For a society to be truly just, it must address these inequalities thoughtfully and deliberately as doing so is essential for sustaining peace and fostering long-term security [30].

One key social determinant of health in dealing with wealth-related inequalities might be using facility delivery services rather than individual health risks. Most researchers have delved into the main causes of inequity issues via examining factors like healthcare systems, individual income, health insurance coverage, accessibility, waiting times, and healthcare cost. Some have revealed that general practitioner care is distributed equally and often even pro-poor in most OECD countries, whereas the distribution of specialist care is very pro-rich, a trend reinforced when private insurance or private care options were offered. Thus, total doctor utilization appears to lean more toward being pro-rich. While it is conceivable for everyone to gain fair access to healthcare to reach the status of health equity, there is still a long way to go [16,25,33].

Literature reviews on HSR have focused on its ability to adapt, withstand, and recover from unexpected disruptions. Hollnagel et al. pointed out that the capacity of resilient healthcare is the ability to adjust its functioning through events and to sustain required operations under both expected and unexpected conditions. The concept of HSR is relatively new in health policy and systems research and remains in a theoretical development phase with little congruence to theory and frameworks. Saulnier et al. (2021) identified five research areas: (1) measuring and managing systems’ dynamic performance, (2) the linkages between societal resilience and HSR, (3) the effect of governance on the capacity for resilience, (4) creating legitimacy, and (5) the influence of the private sector on HSR [20].

Tan et al. utilized a meta-narrative approach to explore and synthesize evidence about healthcare resilience. They presented a unified framework for future resilience-building by including global health, disaster risk reduction, emergency management, patient safety, and public health under varying hierarchical levels, ranging from micro, the individual level; to meso, the facility or organization level; to macro, the health system level; and finally to meta, the planetary or international level.

While researchers are increasingly taking an all-hazards approach and a process-oriented perspective to resilience studies, their methods are seldom incorporated into frameworks and are difficult to be conceptualized into measurement systems. Waitzberg et al. believed that understanding and comparing health systems are key for cross-country learning and strengthening a health system, and using templates is helpful for developing standardized and coherent descriptions and assessments of health systems. They provided an overview and helped identify the most important aspects and topics to look at when comparing and analyzing health systems. While different countries are endowed with different HSR due to their unique social, political, economic, and cultural situations, some common key factors can be observed, and careful study of them across nations can lead to better HSR. Some key factors include, at a minimum, the following: healthcare system structure (public vs. private healthcare systems and funding) and insurance systems (universal healthcare coverage vs. private systems), public health infrastructure (preventive medicine and public health systems) and emergency response capabilities (emergency medical preparedness and disaster management), healthcare workforce (adequacy and distribution of medical personnel) and training (global mobility and knowledge sharing), digitalization (level of healthcare system digitalization) and information technology (telemedicine and technological innovation), government policy and resource allocation, socioeconomic factors and policy flexibility, global health governance and aid and support mechanisms, and health systems’ sustainability [20,34,35,36].

2. Materials and Methods

2.1. Understanding HSR Through Efficiency Analysis

The resilience dimensions identified in this study encompass a diverse array of governance, health system, and economic indicators, which are conceptually organized into absorptive, adaptive, and transformative capacities. Absorptive capacity reflects a nation’s ability to withstand and buffer external shocks through foundational services and institutions, exemplified by access to basic sanitation and drinking water, clean cooking technologies, government effectiveness, regulatory quality, physician and nurse availability, and sustained health outcomes such as life expectancy. Adaptive capacity captures the system’s responsiveness and flexibility to emerging challenges, demonstrated by infrastructure alignment with societal needs, universal health coverage (UHC) service coverage, labor force participation, private health expenditures, and inclusive governance characterized by voice and accountability. In contrast, transformative capacity denotes the potential for long-term structural change and systemic resilience, represented by economic indicators including GDP per capita, government and pension-related health expenditures, and the fiscal sustainability necessary for continued adaptation and innovation [5,7,8,9,10,11,12].

Through the Slacks-Based Measure (SBM) model within Data Envelopment Analysis (DEA) that analyzes efficiency by maximizing outputs, this research offers more evidence of the determinative factors for HSR in order to promote better policy decision-making and more effective resource allocation. It does so by identifying the existing status of each sample country’s HSR between 2016 and 2020. We examined three aspects of the governance as signified by (1) a population using at least basic sanitation services (%) and basic drinking water services (%) and having access to clean fuels and technologies for cooking (%), as well as regulatory quality and voice and accountability; (2) health systems by the number of inhabitants served per physician, number of inhabitants served per nurse, health infrastructure meeting society needs, universal health coverage (UHC) service coverage index, and life expectancy at birth; and (3) the economy by GDP per capita, current health expenditure (CHE) per capita, general government health expenditure (GGHE) per capita, private health expenditure per capita, pension funding adequacy rating, and labor force participation rate. In this CRS-SBM model, all the indicators above are desirable outputs with the exception of the number of inhabitants served per physician and the number of inhabitants served per nurse, which are undesirable indicators. We evaluated each country’s HSR scores of aggregate efficiency and disaggregated efficiency in order to search for any opportunity for improvement [5,8,10,16,17,18,19,20,21].

2.2. Dataset

This study identifies the sources and key drivers of HSR by examining and analyzing the current status of such systems in the sample countries. As noted above, HSR is affected by many factors that are not isolated but are rather closely interconnected and inseparable. Many, if not all, of these factors lack universal official statistics to measure in the first place. Due to the complexity of the issues and to effectively fulfill the primary objective, this study incorporates 17 variable indicators from the aspects of governance, health systems, and economy, based on integrating relevant data from the WHO’s dataset, the World Bank’s dataset, the International Institute for Management Development (IMD), and the IMD World Competitiveness Yearbook (WCY), as summarized in the National Applied Research Laboratories’ (NAR Labs) dataset, between 2016 and 2020 [37,38,39,40]. Appendix A 

Table A1. Summary of indicators, corresponding data, and sources.

Aspects	Indicators	Data	Source
Governance	Basic sanitation services	Population using at least basic sanitation services (%)	WHO
	Basic drinking water services	Population using at least basic drinking water services (%)	WHO
	Clean fuels and tech. cooking	Access to clean fuels and technologies for cooking (% of population)	World Bank
	Government effectiveness	Government effectiveness: percentile rank	World Bank
	Regulatory quality	Regulatory quality: percentile rank	World Bank
	Voice and accountability	Voice and accountability: percentile rank	World Bank
Health system	Number of inhabitants per physician	Number of inhabitants per physician	PRIDE/IMD
	Number of inhabitants per nurse	Number of inhabitants per nurse	PRIDE/IMD
	Life expectancy at birth	Life expectancy at birth	WHO
	Health infra. meets society needs	Health infrastructure meet society needs presented on a scale of 0 to 10	PRIDE/IMD
	UHC service coverage	Universal health coverage index for essential health services presented on a scale of 0 to 100	World Bank
Economic	GDP per capita	GDP per capita based on purchasing power parity, PPP	World Bank
	Current health expenditure per capita	Current health expenditure per capita, PPP	World Bank
	Government health expenditure per capita	Domestic general government health expenditure per capita, PPP	World Bank
	Private health expenditure per capita	Domestic private health expenditure per capita, PPP	World Bank
	Pension funding	WCY executive survey pension fund adequacy rating based on an index from 0 to 10	PRIDE/IMD
	Labor force participation	The labor force divided by the total working-age population aged 15 to 64	World Bank

summarizes 17 indicators, corresponding data, and sources. Countries were included in the sample if they had complete data for all seventeen indicators in at least one year during the period from 2016 to 2020. We calculated these indicators to compare each country’s HSR and their aggregate output efficiency and disaggregated output efficiency. Data were available for 55 countries in 2016, 56 countries in 2017 and 2018, and 54 countries in 2019 and 2020.

2.3. Statistical Analysis

DEA provides a wealth of useful information for diverse decision-setting based on the concepts of efficiency, which usually refers to the degree to which the goals of an organization are met by comparing the relationship between inputs and outputs. Although people generally look to minimize resource costs to achieve perfect outputs and complete organizations’ goals, under certain circumstances, some inputs are hard to measure properly, and/or some outputs are worthy of weightier importance. At this time and for our purpose, we utilized all output variables to emphasize their significance in the valuation of countries’ HSR.

CRS-SBM refers to a Data Envelopment Analysis (DEA) model that combines Constant Returns to Scale (CRS) and the Slacks-Based Measure (SBM). This approach is used for efficiency evaluation, particularly in cases that count for input and output redundancies. CRS assumes that the efficiency of decision-making units (DMUs) remains constant regardless of scale, meaning that proportional increases in inputs result in proportional increases in outputs. SBM considers inefficiencies caused by input excess and output shortfalls, making the efficiency assessment more precise [41,42,43,44].

We first used the CRS-SBM model to evaluate all countries’ output efficiency scores of HSR from the aspects of governance, health systems, and economy via 17 indicators that were weighted thereof under the assumption of output-oriented SBM of a non-radial DEA approach. This is useful to streamline the understanding of overall technical efficiency by taking the slack variables of excess undesirable output and desirable output shortfall [13,15].

It was assumed that each country’s health system i had both desirable and undesirable outputs, denoted by $y^g$ and $y^b.$ The fractional programming problem was solved by the CRS-SBM model with undesirable outputs but without any inputs for country i’s health system. Each country’s health systems had $n_1$ good outputs and $n_2$ bad outputs; $y^g$ and $y^b$ were the matrices of the good outputs and bad outputs, respectively, with both of them having a value greater than zero; $s^g$ and $s^b$ were matrices of the good output and bad output slacks, respectively; and $\mathsf{\lambda }$ was a constant vector of peer weights. The solved value of $\mathsf{\rho }$ was the overall HSR efficiency score for the $i^{th}$ health systems with the inclusion of undesirable outputs.

We further elaborated on each disaggregated efficiency of the 17 indicators for every country to compare the efficiency scores of each measured by both the actual desirable output over the target desirable output and each target undesirable output over the actual undesirable output. This allowed us to analyze each disaggregated efficiency score in terms of both desirable and undesirable output results. If a country has perfect disaggregated efficiency, then its actual desirable output variable tends to equal the target level with a score approaching 1. In contrast, a country with the least disaggregated efficiency has a score that inclines toward 0 [14].

We used the Mann–Whitney U test as a non-parametric test to check whether there was a difference in the rank sum between two independent groups [45]. Assume that two independent groups, X and Y, have $n_1$ and $n_2$ samples, and the rankings of the two groups are ${ R}_1$ and ${ R}_2$ with $n_1$ and $n_2$ samples. When $n_1$ and $n_2$ are larger than 10, the distribution of U approaches normal. If the absolute value of the z-score is above the critical value at a significant level, then the two groups exhibit a significant difference in their total rankings.

3. Results

This section gives a detailed examination of the data analysis process and empirical findings. It outlines the methodology used to process and interpret the data, ensuring accuracy and reliability in the results. Key trends, correlations, and significant insights are highlighted, offering a comprehensive understanding of the study’s implications.

3.1. Descriptive Statistics

Table 1. Basic statistics of desirable and undesirable output variables.

Aspects	Index	Type	Mean	S.D.	Max.	Min.
Governance	Basic sanitation services	Desirable	94.98	8.17	100.00	60.84
	Basic drinking water services	Desirable	98.17	3.46	100.00	79.40
	Clean fuels and tech. cooking	Desirable	94.60	12.23	100.00	43.30
	Government effectiveness	Desirable	75.59	17.66	100.00	33.33
	Regulatory quality	Desirable	76.80	17.19	100.00	33.33
	Voice and accountability	Desirable	68.60	25.95	99.52	4.83
Health systems	Number of inhabitants per physician	Undesirable	487.22	633.56	4394.09	161.62
	Number of inhabitants per nurse	Undesirable	227.39	234.47	1671.14	47.06
	Life expectancy at birth	Desirable	78.80	4.06	85.00	65.00
	Health infra. meets society needs	Desirable	5.90	2.00	9.25	1.72
	UHC service coverage	Desirable	80.27	7.31	91.00	54.00
Economic	GDP per capita	Desirable	39,148.72	22,557.76	116,283.70	5789.68
	Current health expenditure per capita	Desirable	3236.36	2203.05	11,702.41	179.45
	Government health expenditure per capita	Desirable	2238.86	1580.44	6643.36	54.34
	Private health expenditure per capita	Desirable	992.76	952.01	5631.53	118.95
	Pension funding adequacy rating	Desirable	4.18	1.60	8.32	0.84
	Labor force participation	Desirable	72.48	8.42	89.21	40.68

summarizes the statistics. The results show no value equal to or less than zero for either desirable or undesirable output variables. The four nominal variables—GDP per capita, current health expenditure per capita, government health expenditure per capita, and private health expenditure per capita—are all based on purchasing power parity (PPP) conversion. For example, PPP GDP is gross domestic product converted into international dollars using purchasing power parity rates. An international dollar has the same purchasing power over GDP as the U.S. dollar in the United States.

3.2. Correlation Coefficients

Table 2. Correlation coefficients.

Variables	Basic Sanitation Services	Basic Drinking Water Services	Clean Fuels and Tech. Cooking	Government Effectiveness	Regulatory Quality	Voice and Accountability	Number of Inhabitants per Physician	Number of Inhabitants per Nurse	Life Expectancy at Birth	Health Infra. Meets Society Needs	UHC Service Coverage	GDP per Capita	Current Health Expenditure per Capita	Government Health Exp. per Capita	Private Health Exp. per Capita	Pension Funding Adequacy Rating	Labor Force Participation
Basic sanitation services	1
Basic drinking water services	0.836	1
Clean fuels and tech. cooking	0.848	0.797	1
Government effectiveness	0.615	0.528	0.560	1
Regulatory quality	0.567	0.507	0.580	0.882	1
Voice and accountability	0.256	0.306	0.373	0.596	0.709	1
Number of inhabitants per physician	−0.447	−0.364	−0.672	−0.314	−0.371	−0.300	1
Number of inhabitants per nurse	−0.496	−0.432	−0.697	−0.506	−0.494	−0.392	0.800	1
Life expectancy at birth	0.745	0.659	0.646	0.766	0.734	0.511	−0.456	−0.520	1
Health infra. meets society needs	0.594	0.489	0.433	0.764	0.597	0.285	−0.175	−0.385	0.671	1
UHC service coverage	0.729	0.639	0.693	0.648	0.655	0.524	−0.603	−0.659	0.799	0.466	1
GDP per capita	0.502	0.453	0.513	0.718	0.693	0.317	−0.348	−0.513	0.655	0.614	0.512	1
Current health expenditure per capita	0.527	0.489	0.524	0.782	0.773	0.649	−0.372	−0.541	0.697	0.625	0.657	0.743	1
Government health exp. per capita	0.521	0.485	0.524	0.775	0.778	0.682	−0.375	−0.554	0.706	0.626	0.655	0.732	0.926	1
Private health exp. per capita	0.356	0.328	0.344	0.524	0.497	0.368	−0.237	−0.333	0.443	0.407	0.432	0.501	0.777	0.482	1
Pension funding adequacy rating	0.250	0.225	0.182	0.518	0.447	0.054	−0.007	−0.128	0.324	0.602	0.165	0.478	0.375	0.363	0.265	1
Labor force participation	0.404	0.335	0.383	0.579	0.595	0.327	−0.284	−0.473	0.501	0.380	0.503	0.528	0.486	0.491	0.312	0.352	1

presents a correlation analysis conducted using the average values of desirable and undesirable output variables from 2016 to 2020. Notably, several key indicator variables in HSR exhibit a significant correlation with basic livelihood needs. However, as most correlations remain below 0.8, severe multicollinearity is absent in the model. Consequently, no indicators in the DEA model require replacement, ensuring its solvability and confirming the absence of severe multicollinearity between output variables.

3.3. Efficiency Scores of HSR

The relevant countries’ HSR scores are illustrated across multiple visuals and data sets: Figure 1 presents output efficiency scores categorized by income group, while Figure 2 offers a geographic world map view of resilience scores to enable cross-country comparisons. However, due to the relatively small geographic size of several European nations, some details may not be clearly visible in Figure 2; therefore, Figure 3 provides a dedicated visualization focusing exclusively on European countries. Additionally,

Table 3. HSR scores and ranks for each country.

Country	Score	Rank	Country	Score	Rank	Country	Score	Rank
Australia	1	1	Ireland	0.887	20	Bulgaria	0.506	39
Austria	1	1	Canada	0.887	21	Croatia	0.476	40
Denmark	1	1	South Korea	0.858	22	Brazil	0.454	41
Finland	1	1	Israel	0.833	23	Saudi Arabia	0.447	42
Germany	1	1	Chile	0.791	24	Romania	0.446	43
Greece	1	1	Spain	0.758	25	Malaysia	0.442	44
Iceland	1	1	Italy	0.743	26	Mexico	0.432	45
Japan	1	1	United Kingdom	0.736	27	South Africa	0.401	46
Luxembourg	1	1	United Arab Emirates	0.688	28	Colombia	0.393	47
The Netherlands	1	1	Lithuania	0.678	29	Turkiye	0.381	48
New Zealand	1	1	Cyprus	0.655	30	Kazakhstan	0.380	49
Qatar	1	1	Slovenia	0.654	31	Jordan	0.346	50
Singapore	1	1	Czech Republic	0.595	32	China	0.324	51
Sweden	1	1	Estonia	0.593	33	Peru	0.324	52
Switzerland	1	1	Republic of Latvia	0.565	34	Thailand	0.312	53
United States	1	1	Hungary	0.552	35	Mongolia	0.261	54
Portugal	0.964	17	Argentina	0.531	36	Indonesia	0.224	55
Belgium	0.962	18	Poland	0.519	37	Philippines	0.202	56
France	0.960	19	Slovak Republic	0.511	38	India	0.129	57

ranks and lists HSR efficiency scores for the relevant countries. There are sixteen countries in the top I tier with a perfect score of 1. They all belong to the high-income group in the World Bank’s country classifications of income level based on GNI per capita, including Australia, Austria, Denmark, Finland, Germany, Greece, Iceland, Japan, Luxembourg, The Netherlands, New Zealand, Qatar, Singapore, Sweden, Switzerland, and the United States. However, not every high-income country is able to achieve such perfect efficiency. The scores for France, Canada, South Korea, the United Kingdom, the United Arab Emirates, etc., are all below 1.

3.4. Testing of Hypotheses

Hypothesis 1: High-income countries have higher HSR scores than others. The Mann–Whitney U test results in Appendix B 

Table A2. Hypothesis 1: High-income countries have higher HSR scores than others.

Year	U-Value	p-Value
2016	9.00 ***	<0.0001
2017	10.00 ***	<0.0001
2018	9.00 ***	<0.0001
2019	2.00 ***	<0.0001
2020	2.00 **	<0.0001

Note: One-tailed hypothesis; ***: ≤0.01 and **: ≤0.05.

are all significant from 2016 to 2020, showing a clearly strong relationship between countries’ strong HSR and being in the high-income group. Economic resources are essential, and building up resilient health systems is hard, if possible, to accomplish without them. In fact, Paschoalotto et al. previously found that high-income countries (HICs) usually have greater capacities in several dimensions than do low- and middle-income countries (LMICs) principally because richer countries with greater financial resources are in a better position with regard to obtaining human, physical, and structural resources; adapting to adjust logistics and healthcare supply chain issues; utilizing information and communications technologies; and having better instrumental capacity and tools to plan, monitor, and evaluate in order to hit the targets [46].

Hypothesis 2: European countries have higher HSR scores than countries in other parts of the world. The test results in Appendix C 

Table A3. Hypothesis 2: European countries have higher HSR scores than countries in other parts of the world.

Year	U-Value	p-Value
2016	211.00 ***	0.0033
2017	249.00 **	0.0116
2018	230.50 ***	0.0049
2019	235.50 **	0.0174
2020	230.00 **	0.0139

Note: One-tailed hypothesis; ***: ≤0.01 and **: ≤0.05.

are all significant. Regional HSR differences do exist: ten of the sixteen countries tied with perfect scores are in Europe. European countries generally tend to have the highest HSR scores with an average score of 0.768. Southeast Asian countries have the lowest scores with an average of 0.222. The latter group of countries shows the worst disaggregated efficiency in the economy and the worse one among the health systems.

As to output efficiency in terms of the 17 indicators from the aspects of governance, health systems, and economy, the economy at an overall average score of 0.662 is the most volatile. Health systems at an overall average of 0.799 ranks second in volatility. Governance with an overall average of 0.882 ranks third. Appendix G 

Table A7. Output efficiency of the governance aspect.

Country	Basic Sanitation Services	Basic Drinking Water Services	Clean Fuels and Tech. Cooking	Government Effectiveness	Regulatory Quality	Voice and Accountability
Argentina	0.971	0.998	1.000	0.538	0.412	0.663
Australia	1.000	1.000	1.000	1.000	1.000	1.000
Austria	1.000	1.000	1.000	1.000	1.000	1.000
Belgium	1.000	1.000	1.000	0.982	0.990	0.989
Brazil	0.898	1.000	0.971	0.396	0.488	0.612
Bulgaria	0.867	1.000	0.893	0.569	0.763	0.611
Canada	0.996	0.995	1.000	0.989	0.994	0.994
Chile	1.000	0.999	1.000	0.875	0.913	0.932
China	0.948	1.000	0.795	0.739	0.476	0.078
Colombia	0.945	1.000	0.951	0.529	0.677	0.544
Croatia	0.962	1.000	1.000	0.700	0.691	0.637
Cyprus	0.995	0.998	1.000	0.783	0.836	0.815
Czech Republic	0.992	0.999	1.000	0.799	0.891	0.788
Denmark	1.000	1.000	1.000	1.000	1.000	1.000
Estonia	0.992	1.000	1.000	0.847	0.966	0.903
Finland	1.000	1.000	1.000	1.000	1.000	1.000
France	0.998	1.000	1.000	0.982	0.981	0.969
Germany	1.000	1.000	1.000	1.000	1.000	1.000
Greece	1.000	1.000	1.000	1.000	1.000	1.000
Hungary	0.981	1.000	1.000	0.691	0.733	0.621
Iceland	1.000	1.000	1.000	1.000	1.000	1.000
India	0.669	0.912	0.598	0.657	0.490	0.707
Indonesia	0.844	0.964	0.826	0.636	0.621	0.615
Ireland	0.923	0.969	1.000	0.930	0.980	0.960
Israel	1.000	1.000	1.000	0.930	0.954	0.842
Italy	1.000	1.000	1.000	0.755	0.806	0.856
Japan	1.000	1.000	1.000	1.000	1.000	1.000
Jordan	0.976	0.992	1.000	0.588	0.616	0.307
Kazakhstan	1.000	0.974	0.949	0.531	0.604	0.152
Lithuania	0.933	0.977	1.000	0.828	0.877	0.812
Luxembourg	1.000	1.000	1.000	1.000	1.000	1.000
Malaysia	0.987	1.000	0.989	0.827	0.781	0.416
Mexico	0.914	1.000	0.859	0.479	0.615	0.492
Mongolia	0.778	0.943	0.568	0.507	0.621	0.694
The Netherlands	1.000	1.000	1.000	1.000	1.000	1.000
New Zealand	1.000	1.000	1.000	1.000	1.000	1.000
Peru	0.800	0.970	0.855	0.461	0.746	0.610
Philippines	0.849	0.988	0.480	0.623	0.673	0.609
Poland	0.989	0.926	1.000	0.707	0.813	0.720
Portugal	0.999	0.998	1.000	0.983	0.964	0.983
Qatar	1.000	1.000	1.000	1.000	1.000	1.000
Republic of Latvia	0.924	0.987	1.000	0.794	0.867	0.751
Romania	0.858	1.000	0.872	0.463	0.694	0.668
Saudi Arabia	0.954	0.987	1.000	0.631	0.525	0.057
Singapore	1.000	1.000	1.000	1.000	1.000	1.000
Slovak Republic	0.976	0.998	1.000	0.727	0.801	0.769
Slovenia	0.986	0.995	1.000	0.838	0.776	0.808
South Africa	0.798	1.000	0.919	0.616	0.623	0.762
South Korea	1.000	0.998	1.000	0.928	0.941	0.890
Spain	1.000	0.999	1.000	0.855	0.852	0.870
Sweden	1.000	1.000	1.000	1.000	1.000	1.000
Switzerland	1.000	1.000	1.000	1.000	1.000	1.000
Thailand	0.985	1.000	0.820	0.658	0.568	0.246
Turkiye	1.000	0.985	0.969	0.524	0.583	0.271
United Arab Emirates	0.994	1.000	1.000	0.921	0.871	0.349
United Kingdom	0.996	1.000	1.000	0.948	0.985	0.955
United States	1.000	1.000	1.000	1.000	1.000	1.000
Indicator’s average output efficiency	0.959	0.992	0.953	0.803	0.826	0.760
Total average output efficiency of the governance				0.882

, Appendix H 

Table A8. Output efficiency of the health systems aspect.

Country	Number of Inhabitants per Physician	Number of Inhabitants per Nurse	Life Expectancy at Birth	Health Infrastructure Meets Society Needs	UHC Service Coverage
Argentina	0.922	0.203	0.920	0.515	0.914
Australia	1.000	1.000	1.000	1.000	1.000
Austria	1.000	1.000	1.000	1.000	1.000
Belgium	0.944	0.952	0.998	1.000	1.000
Brazil	0.507	0.560	0.907	0.244	0.960
Bulgaria	0.949	0.251	0.911	0.340	-
Canada	0.763	0.839	0.998	0.925	1.000
Chile	0.851	0.500	0.976	0.743	0.981
China	0.563	0.157	0.983	0.643	0.982
Colombia	0.495	0.075	0.950	0.306	0.962
Croatia	0.789	0.376	0.938	0.517	0.924
Cyprus	0.948	0.304	0.970	0.537	0.942
Czech Republic	0.959	0.487	0.949	0.772	0.971
Denmark	1.000	1.000	1.000	1.000	1.000
Estonia	0.799	0.353	0.938	0.643	0.913
Finland	1.000	1.000	1.000	1.000	1.000
France	0.944	0.950	1.000	0.973	1.000
Germany	1.000	1.000	1.000	1.000	1.000
Greece	1.000	1.000	1.000	1.000	1.000
Hungary	0.779	0.372	0.914	0.329	0.919
Iceland	1.000	1.000	1.000	1.000	1.000
India	0.287	0.152	0.895	0.695	0.727
Indonesia	0.122	0.123	0.913	0.740	0.674
Ireland	0.907	0.899	0.997	0.592	0.950
Israel	0.849	0.714	0.998	0.825	1.000
Italy	0.961	0.477	0.996	0.813	0.988
Japan	1.000	1.000	1.000	1.000	1.000
Jordan	0.729	0.205	0.931	0.785	0.799
Kazakhstan	0.915	0.360	0.872	0.527	0.955
Lithuania	1.000	0.625	0.908	0.643	0.860
Luxembourg	1.000	1.000	1.000	1.000	1.000
Malaysia	0.472	0.203	0.943	0.879	0.928
Mexico	0.627	0.181	0.902	0.442	0.871
Mongolia	0.744	0.193	1.000	0.333	0.916
The Netherlands	1.000	1.000	1.000	1.000	1.000
New Zealand	1.000	1.000	1.000	1.000	1.000
Peru	0.585	0.159	0.960	0.286	0.907
Philippines	0.088	0.055	0.959	0.713	0.750
Poland	0.547	0.284	0.933	0.371	0.953
Portugal	1.000	0.971	0.997	0.958	1.000
Qatar	1.000	1.000	1.000	1.000	1.000
Republic of Latvia	0.751	0.252	0.904	0.444	0.878
Romania	0.690	0.408	0.909	0.308	0.913
Saudi Arabia	0.595	0.307	0.922	0.673	0.849
Singapore	1.000	1.000	1.000	1.000	1.000
Slovak Republic	0.809	0.324	0.926	0.356	0.965
Slovenia	0.721	0.571	0.971	0.565	0.977
South Africa	0.166	0.202	0.847	0.415	0.891
South Korea	0.817	0.762	0.998	0.924	1.000
Spain	0.963	0.497	0.996	0.938	0.999
Sweden	1.000	1.000	1.000	1.000	1.000
Switzerland	1.000	1.000	1.000	1.000	1.000
Thailand	0.136	0.215	0.964	0.837	0.950
Turkiye	0.426	0.124	0.942	0.757	0.911
United Arab Emirates	0.690	0.490	0.964	0.898	0.925
United Kingdom	0.665	0.535	0.985	0.698	0.994
United States	1.000	1.000	1.000	1.000	1.000
Indicator’s average output efficiency	0.780	0.573	0.963	0.735	0.946
Total average output efficiency of the health systems				0.799

, and Appendix I 

Table A9. Output efficiency of the economic aspect.

Country	GDP per Capita	Current Health Expenditure per Capita	Government Health Expenditure per Capita	Private Health Expenditure per Capita	Pension Funding Adequacy Rating	Labor Force Participation
Argentina	0.325	0.266	0.526	0.142	0.276	0.793
Australia	1.000	1.000	1.000	1.000	1.000	1.000
Austria	1.000	1.000	1.000	1.000	1.000	1.000
Belgium	0.970	0.957	1.000	0.887	0.912	0.972
Brazil	0.213	0.177	0.232	0.151	0.302	0.834
Bulgaria	0.308	0.194	0.345	0.126	0.344	0.831
Canada	0.832	0.871	1.000	0.707	1.000	0.987
Chile	0.653	0.659	0.641	0.771	0.901	0.875
China	0.231	0.104	0.181	0.067	0.660	0.946
Colombia	0.214	0.151	0.333	0.064	0.554	0.885
Croatia	0.401	0.240	0.610	0.063	0.286	0.793
Cyprus	0.580	0.351	0.584	0.230	0.574	0.869
Czech Republic	0.571	0.380	0.967	0.087	0.403	0.911
Denmark	1.000	1.000	1.000	1.000	1.000	1.000
Estonia	0.503	0.302	0.697	0.113	0.519	0.932
Finland	1.000	1.000	1.000	1.000	1.000	1.000
France	0.942	0.951	1.000	0.884	0.931	0.977
Germany	1.000	1.000	1.000	1.000	1.000	1.000
Greece	1.000	1.000	1.000	1.000	1.000	1.000
Hungary	0.445	0.260	0.555	0.119	0.480	0.854
Iceland	1.000	1.000	1.000	1.000	1.000	1.000
India	0.103	0.019	0.012	0.028	1.000	0.703
Indonesia	0.184	0.040	0.045	0.036	1.000	0.945
Ireland	0.998	0.873	1.000	0.672	0.692	0.947
Israel	0.784	0.678	0.760	0.582	0.960	0.931
Italy	0.689	0.573	1.000	0.345	0.689	0.828
Japan	1.000	1.000	1.000	1.000	1.000	1.000
Jordan	0.147	0.080	0.092	0.065	1.000	0.527
Kazakhstan	0.375	0.101	0.194	0.057	0.691	0.945
Lithuania	0.542	0.324	0.519	0.185	0.466	0.943
Luxembourg	1.000	1.000	1.000	1.000	1.000	1.000
Malaysia	0.400	0.126	0.184	0.094	1.000	0.860
Mexico	0.297	0.132	0.196	0.103	0.497	0.792
Mongolia	0.201	0.070	0.127	0.036	0.389	0.892
The Netherlands	1.000	1.000	1.000	1.000	1.000	1.000
New Zealand	1.000	1.000	1.000	1.000	1.000	1.000
Peru	0.190	0.084	0.156	0.047	0.711	1.000
Philippines	0.139	0.036	0.029	0.043	0.997	0.878
Poland	0.454	0.253	0.551	0.110	0.340	0.836
Portugal	0.914	0.901	0.899	0.903	0.919	0.989
Qatar	1.000	1.000	1.000	1.000	1.000	1.000
Republic of Latvia	0.430	0.238	0.436	0.142	0.610	0.928
Romania	0.406	0.198	0.486	0.060	0.400	0.804
Saudi Arabia	0.691	0.345	0.781	0.145	0.904	0.683
Singapore	1.000	1.000	1.000	1.000	1.000	1.000
Slovak Republic	0.448	0.265	0.653	0.080	0.339	0.864
Slovenia	0.540	0.402	0.900	0.165	0.538	0.880
South Africa	0.219	0.153	0.278	0.092	0.650	0.801
South Korea	0.838	0.750	0.879	0.691	0.845	0.929
Spain	0.679	0.587	0.922	0.379	0.541	0.913
Sweden	1.000	1.000	1.000	1.000	1.000	1.000
Switzerland	1.000	1.000	1.000	1.000	1.000	1.000
Thailand	0.264	0.079	0.142	0.038	0.885	0.950
Turkiye	0.410	0.147	0.355	0.048	0.630	0.686
United Arab Emirates	1.000	0.504	0.731	0.387	0.943	0.992
United Kingdom	0.756	0.667	1.000	0.276	0.691	0.959
United States	1.000	1.000	1.000	1.000	1.000	1.000
Indicator’s average output efficiency	0.637	0.535	0.667	0.460	0.763	0.910
Total average output efficiency of the economic aspect				0.662

list detailed output efficiency of governance, health systems, and the economy.

For the economic aspect, the output efficiency of per capita private health expenditure scores the lowest at 0.460. It is followed by per capita current health expenditure at 0.535, per capita GDP at 0.637, per capita government health expenditure at 0.667, pension funding adequacy rating at 0.763, and labor force participation at 0.910.

Hypothesis 3a/3b: Countries with more abundant healthcare workers–physicians/nurses have higher disaggregated efficiency. The significance test results in Appendix D 

Table A4. Hypothesis 3a: Countries with more abundant healthcare workers–physicians have higher disaggregated efficiency.

Year	U-Value	p-Value
2016	54.00 ***	<0.0001
2017	28.00 ***	<0.0001
2018	60.00 ***	<0.0001
2019	43.00 ***	<0.0001
2020	134.00 ***	0.001

Note: One-tailed hypothesis; ***: ≤0.01.

and Appendix E 

Table A5. Hypothesis 3b: Countries with more abundant healthcare workers–nurses have higher disaggregated efficiency.

Year	U-Value	p-Value
2016	3.00 ***	<0.0001
2017	3.00 ***	<0.0001
2018	9.00 ***	<0.0001
2019	2.00 ***	<0.0001
2020	23.00 ***	<0.0001

Note: One-tailed hypothesis; ***: ≤0.01.

show that countries with ample healthcare workers–physicians/nurses have higher disaggregated efficiency. In terms of health systems, output efficiency for the number of inhabitants served per nurse is the lowest at 0.573, followed by health infrastructure meeting societal needs at 0.735, the number of inhabitants served per physician at 0.780, UHC (universal health coverage) service coverage at 0.946, and life expectancy at birth at 0.963.

Hypothesis 4: Countries with better governance have higher disaggregated efficiency. The significant test results in Appendix F 

Table A6. Hypothesis 4: Countries with better governance have higher disaggregated efficiency.

Year	U-Value	p-Value
2016	8.00 ***	<0.0001
2017	9.00 ***	<0.0001
2018	17.00 ***	<0.0001
2019	13.00 ***	<0.0001
2020	26.00 ***	<0.0001

Note: One-tailed hypothesis; ***: ≤0.01.

denote that countries with better governance have higher disaggregated efficiency. In terms of governance, output efficiency for voice and accountability is the lowest at 0.760, followed by government effectiveness at 0.803, regulatory quality at 0.826, clean fuels and technologies for cooking at 0.953, basic sanitation services at 0.959, and basic drinking water services at 0.992.

4. Discussion

4.1. Diverging HSR: Key National Drivers

HSR depends on numerous social and economic factors. It must be addressed via country-level aggregate efficiency analyses and micro-managerial disaggregated efficiency analyses for every desirable output and undesirable output. Utilizing DEA, we explore some significant issues on this topic that could be extremely helpful for countries to better deal with this very issue.

The IMD WCY is particularly useful in analyzing long-term trends to trace those countries at the top of the list, whereby each has its own unique approach to becoming more competitive. The 2020 yearbook covers 64 economies and ranks them based on four dimensions: competitiveness of infrastructure, government efficiency, economic performance, and business efficiency [40]. The construction methods of the Healthcare Access and Quality (HAQ) Index are updated for the Global Burden of Disease Study (GBD) 2019, which is based on death rates from 32 causes of death that could be avoided by timely and effective medical care (also known as amenable mortality). The HAQ Index shows strong convergence validity and provides a comprehensive measure reflecting a health system’s capacity for effectively detecting, managing, and preventing risks [47]. Our research findings of the ranking trend on HSR are mostly consistent with the rankings in the 2020 WCY and the HAQ index in 2019. Appendix J 

Table A10. List of HSR scores, IMD WCY ranking, and HAQ index ranking.

Country	HSR Score Ranking	WCY Ranking 2020	HAQ Index Ranking 2019	Country	HSR Score Ranking	WCY Ranking 2020	HAQ Index Ranking 2019	Country	HSR Score Ranking	WCY Ranking 2020	HAQ Index Ranking 2019
Argentina	36	62	47	Hungary	35	47	31	Poland	37	39	34
Australia	1	18	6	Iceland	1	21	1	Portugal	17	37	23
Austria	1	16	11	India	57	43	57	Qatar	1	14	32
Belgium	18	25	17	Indonesia	55	40	55	Republic of Latvia	34	41	38
Brazil	41	56	51	Ireland	20	12	7	Romania	43	51	37
Bulgaria	39	48	41	Israel	23	26	26	Saudi Arabia	42	24	43
Canada	21	8	4	Italy	26	44	9	Singapore	1	1	20
Chile	24	38	35	Japan	1	34	14	Slovak Republic	38	57	33
China	51	20	36	Jordan	50	58	40	Slovenia	31	35	12
Colombia	47	54	45	Kazakhstan	49	42	48	South Africa	46	59	54
Croatia	40	60	28	Lithuania	29	31	39	South Korea	22	23	18
Cyprus	30	30	19	Luxembourg	1	15	15	Spain	25	36	8
Czech Republic	32	33	27	Malaysia	44	27	50	Sweden	1	6	5
Denmark	1	2	22	Mexico	45	53	52	Switzerland	1	3	2
Estonia	33	28	30	Mongolia	54	61	53	Thailand	53	29	44
Finland	1	13	13	The Netherlands	1	4	3	Turkiye	48	46	42
France	19	32	10	New Zealand	1	22	21	United Arab Emirates	28	9	49
Germany	1	17	16	Peru	52	52	46	United Kingdom	27	19	25
Greece	1	49	24	Philippines	56	45	56	United States	1	10	29

Source: Authors’ elaboration based on data from IMD 2020 WCY ranking and 2019 HAQ index ranking.

lists detailed rankings.

The three distinct strategies that drove success in 2020 WCY are those of Singapore, founded on its strong economic performance; Denmark, based on its robust economy, labor market, health, and education systems; and Switzerland, rooted in its strong international trade, which fuels its economic growth. Each one exhibits a perfect score for HSR and ranks 20th, 22nd, and 2nd, respectively, in the 2019 HAQ index ranking.

Abnormalities do exist for some countries in the 2020 WCY rankings. The United States came in at 10th place (3rd in 2019). Trade wars seem to have damaged the economies of both China and the U.S., reversing their positive growth trajectories. China also dropped to 20th position in 2020 (14th in 2019). Although the U.S. still has a perfect score on HSR, it ranked 29th in the 2019 HAQ index ranking. Despite its strong defensive medical capabilities and ample medical resources, its healthcare systems still seemingly were dysfunctional when confronted by COVID-19. In fact, the U.S. had the highest COVID-19 confirmed cases and highest fatality rates in the world.

During the pandemic, the Global Health Security (GHS) Index aimed to assess the capacities of 195 countries to better prepare for future epidemics and pandemics. The results showed that the U.S. performs well in the GHS Index. However, there are still challenges in responding to pandemics. The U.S. does not perform well on public trust in government—a fact that may affect public compliance with anti-pandemic measures. The U.S. also ranks low in the proportion of healthcare resources per capita since it does not have universal health insurance, resulting in many people’s inadequate medical support for emergencies. Additionally, its federal regulatory restrictions have led to a lack of a nationwide testing strategy and ongoing equipment and supply shortages, resulting in individual states having to develop their own pandemic prevention strategies. Therefore, although the U.S. has extensive experience in assisting other countries in responding to pandemics, it lacks a comprehensive national strategy, leading to a lack of effective use of its own expertise and resources [48].

Although Greece has faced significant economic challenges, including a financial crisis that led to tightening fiscal measures affecting various sectors, including healthcare, it still has a perfect score for HSR. Even though the financial crisis there did strain the country’s healthcare system with challenges to access, resources, and healthcare delivery, the country was nonetheless able to demonstrate remarkable resilience through community support, adaptation, and international collaboration. This helped it to sustain a relatively adequate level of healthcare services [49].

Qatar, United Arab Emirates (UAE), and Saudi Arabia are all high-income countries. However, each shows dramatic differences in HSR, ranking 1st, 28th, and 42nd, respectively. High-income countries do not necessarily translate into or ensure efficient allocation or utilization of healthcare resources. Qatar is considered an urbanized country. The majority of its population (about 2.6 million in 2021), its primary economic activities, and its political and cultural hubs are all concentrated in its capital, Doha. At the same time, Saudi Arabia (a population of 35.9 million) and UAE (a population 9.4 million) both rank low in efficiency scores of HSR, mostly because of serious shortages of healthcare workers (HCWs).

South Korea, being advanced in digital health innovation and having a relatively high level of execution efficiency in economic development, as well as implementation of public policies, nevertheless is only 22nd ranking in HSR. In 2019, there were over fifty million people in South Korea, and about 90% of them lived in urban regions. It was facing a severe shortage of hospital beds and healthcare workers. The COVID-19 pandemic, its rapidly aging society, and the increasing life expectancy of its population have heavily burdened its health systems [50].

Mongolia, ranked 54th in HSR, is listed among the thirty countries with a high tuberculosis burden. In 2021, approximately 10-11% of tuberculosis cases were in children, or higher than the global average of 6.0%. Prioritizing prevention strategies for tuberculosis should clearly go first for the country. However, its economy relies on agriculture and mining, resulting in limited financial resources to tackle this problem and to build a better healthcare infrastructure. Mongolia’s extreme climate with lifestyle-related illnesses like cardiovascular diseases and its vast geography also give rise to huge difficulties in patients’ access to medical facilities and HCWs’ delivery of healthcare services to patients. Moreover, Mongolia has always had shortage problems of HCWs, particularly in rural areas. This scarcity has severe negative effects on the quality of healthcare services and has deepened the country’s healthcare inequities [51].

India’s strengths in the medical and health industries are known worldwide. However, it is among the lowest ranked in HSR. Several country-specific conditions appear to contribute to this phenomenon, including the fact that India has the largest population in the world at more than 1.4 billion people and has wide wealth disparity. A large portion of its economy depends on daily wages, and 9 million people are living in Mumbai slums with poor living conditions. Even though the country has made progress in healthcare infrastructure, the uneven distribution of both medical facilities and access to essential medications weakens its health system’s ability to respond to emergencies in vast areas. Socioeconomic disparities have substantially worsened health inequities, and the country also faces a shortage of HCWs, including doctors, nurses, and support staff. Most citizens rely on out-of-pocket expenses for their healthcare, and so sickness often leads to financial disaster or even ruins. The country’s limited health insurance coverage further hinders patients from accessing quality care. All of the above factors have contributed to India’s problem of lacking a resilient health system [52].

The world population is presently over 8 billion. India and China are the two most populous countries, each with over 1.4 billion people. Indonesia and Brazil also have significant populations. It is not encouraging in view of the fact that these populous countries all have lower scores of HSR, ranking 57th, 51st, 41st, and 55th. They are all facing several common and prominent challenges that impact their HSR, e.g., limited financial resources for both individuals and the government; a shortage of HCWs, particularly in remote and underserved areas; and more frequent and larger-scale public health challenges caused by infectious diseases, sanitation-related illnesses, and age-related chronic diseases.

From the above discussion, it is apparent that economic resources are essential for people to live better and for health systems to progress more resiliently. However, adequate economic resources are just one factor in raising and strengthening HSR. Truly resilient health systems are influenced and determined by a multitude of factors, including resource allocation, population structure, government effectiveness, healthcare policies and management, and public health infrastructure, among others.

4.2. Wealth, Equity, and Global HSR

This study focuses on gaining a deeper understanding of the available data and procuring further insights on real-world evidence in order to conclude that adequate economic resources are the foundation of HSR. Moreover, sufficient healthcare workers are able to support substantial health systems. Last, superior governance and equitable utilization of resources are both essential to HSR.

The sixteen countries at the very top tier with a perfect score in HSR all belong to the high-income group in the World Bank’s country classifications of income level. Regional differences do exist. Ten of the sixteen countries tied with perfect scores are located in Europe. The results show a clear and strong relationship between a country’s HSR and its economic resources as reflected by its income level. This finding is akin to and compatible with the resource-based theory (RBT) in organization management, which deems an organization’s internal resources (either tangible, such as capital and equipment, or intangible, such as brand, technology, or organizational knowledge and culture) are the crucial and most valuable elements for businesses to gain competitive advantages and achieve long-term success. Nevertheless, our study shows in terms of HSR that not every high-income country performs as superbly and ideally as it could in view of their more than adequate resources, although it is apparent that the shortage of such resources has hindered and stymied some, or even most, low- and middle-income countries’ efforts to achieve higher efficiency.

The sudden advent of the SARS and COVID-19 pandemics caused the halt and collapse of most global activities. High-income countries have since shifted from weak resistance–good recoverability to good resistance–weak recoverability in national economic resilience, while an estimated one hundred million people in low- and middle-income countries have fallen into distressing poverty. While most countries are reviewing and reinforcing their HSR in response, many low- and middle-income countries are still struggling to cope with the consequences of some more infectious diseases, such as measles or malaria. A large portion of the world population are still living under conditions with poor HSR [21,41].

The 2030 Agenda for Sustainable Development, adopted in 2015 by all members of the United Nations, provides a shared blueprint for the basic peace and prosperity of people and the planet. At its heart are 17 Sustainable Development Goals (SDGs), constituting an urgent call for action by all countries for the benefit of humanity. Good health and well-being (SDG3) is among the top three SGDs, which also include no poverty (SDG1) and zero hunger (SDG2). These priorities are apparently consistent with our findings herein. Human societies, of course, must endeavor to end hunger and poverty everywhere, achieve food security and improved nutrition, and promote sustainable agriculture so that everyone can ensure healthy lives and well-being [53].

The inequity of health problems among different socioeconomic backgrounds is an equally crucial issue around the world at this time, and even advanced countries are facing seemingly growing health disparities. To mitigate the inequity of health problems and to build resilient health systems, substantial economic resources are required. After all, adequate economic resources are the very foundation for resilient health systems. Nonetheless, unless equitable utilization of resources is properly and concurrently addressed, adequate economic resources alone can hardly substantially raise worldwide HSR at least for the near future.

4.3. Exploring the Applications and Limitations of Slack-Based Measure (SBM) Models

While the SBM model offers a powerful, non-radial, and non-oriented approach in DEA, it has notable limitations. Primarily, it struggles to distinguish between efficient DMUs as all efficient units receive a score of 1, preventing further ranking or differentiation and thus reducing its discriminatory power when many DMUs are efficient. In this study, there are sixteen countries all in the top I tier with a perfect score of 1, as revealed in Figure 1 and Table 3. Additionally, SBM treats input and output slacks independently, which can limit the realism and interpretability of results in scenarios where these are interrelated. The model also exhibits weak discrimination in high-dimensional data, often classifying numerous units as efficient when there are many variables but few DMUs, leading to efficiency overestimation. Furthermore, the standard SBM model does not support super-efficiency, making it incapable of distinguishing among efficient DMUs that achieve an efficiency score of 1. As a result, ranking these efficient units requires extended methodologies. The Super SBM model, developed as an extension of the standard SBM framework, addresses this limitation by allowing efficiency scores to exceed 1. This enhancement facilitates a meaningful ranking and comparison of fully efficient DMUs, thereby improving the discriminatory power of DEA analyses.

We reclassified two undesirable output variables (health systems by number of inhabitants served per physician and number of inhabitants served per nurse) as input variables and re-computed the model using the Super SBM approach, in order to compare the results with the standard SBM model presented in Appendix K 

Table A11. Cross-country comparison of aggregate HSR efficiency scores using Super SBM models.

	Output-Oriented Super SBM with Undesirable Items as Inputs
Country	Aggregate HSR Score	Rank
Greece	1.048	1
Finland	1.038	2
Switzerland	1.016	3
Austria	1.008	4
Luxembourg	1.007	5
Germany	1.005	6
Denmark	1.004	7
Sweden	1.001	8
Iceland	0.898	10
The Netherlands	0.751	11
Australia	0.720	12
United States	0.651	14
New Zealand	0.516	21
Singapore	0.434	26
Japan	0.397	30
Qatar	0.294	38
Portugal	0.962	9
Belgium	0.550	19
France	0.530	20
Ireland	0.673	13
Canada	0.461	23
South Korea	0.363	33
Israel	0.480	22
Chile	0.319	37
Spain	0.596	15
Italy	0.572	16
United Kingdom	0.436	25
United Arab Emirates	0.338	35
Lithuania	0.568	17
Cyprus	0.555	18
Slovenia	0.424	27
Czech Republic	0.451	24
Estonia	0.410	28
Republic of Latvia	0.369	32
Hungary	0.354	34
Argentina	0.398	29
Poland	0.241	41
Slovak Republic	0.326	36
Bulgaria	0.376	31
Croatia	0.284	39
Brazil	0.194	44
Saudi Arabia	0.203	43
Romania	0.229	42
Malaysia	0.161	46
Mexico	0.188	45
South Africa	0.074	53
Colombia	0.155	47
Türkiye	0.124	52
Kazakhstan	0.241	40
Jordan	0.139	49
China	0.134	50
Peru	0.129	51
Thailand	0.039	54
Mongolia	0.140	48
Indonesia	0.016	56
Philippines	0.009	57
India	0.016	55

. In the Super SBM model the undesirable outputs are inputs because as the inputs increase, the efficiency scores decrease, given other things being equal. The results reveal among the sixteen countries previously with a perfect efficiency score of 1 under the standard SBM that Singapore, Japan, and Qatar dropped in rankings—falling to 26th, 30th, and 38th, respectively.

Both the standard and Super SBM models are subject to discontinuity issues, where minor variations in input or output data can lead to abrupt changes in efficiency scores. This instability compromises the reliability and interpretability of efficiency assessments. To address these limitations—particularly the discontinuity inherent in the Super SBM model—Chen [54] introduced the Joint SBM (J-SBM) model. It was developed to unify the evaluation process by simultaneously computing SBM scores for inefficient DMUs and super-efficiency scores for efficient ones. Traditionally, these two groups require separate analyses using standard SBM (Table 3) and Super SBM (Appendix K Table A11) models. The J-SBM model integrates both into a single, continuous framework, enhancing consistency and analytical robustness.

To the best of our knowledge, this study is the first to quantitatively assess the aggregated efficiency of country-level HSR via the standard SBM model within the DEA framework (Table 3). We explore disaggregated efficiency to provide further practical insights. Future researchers may consider employing the Joint SBM (J-SBM) model to address the discontinuity issues inherent in the standard and super-efficiency SBM approaches, thereby helping mitigate this study’s methodological limitations [14,15,54,55].

4.4. Reimagining Governance, Health Systems, and Economic Indices with WHO’s HSR Toolkits

The WHO in 2014 introduced a comprehensive package of HSR indicators to strengthen global health systems. This package serves as a structured resource for measuring and monitoring resilience during both routine operations and crises, helping to bridge a critical data gap in assessing system vulnerabilities and progress. By aligning with initiatives such as the HSR Toolkit and WHO’s policy recommendations for universal health coverage (UHC), the indicator package reinforces its strategic role in promoting resilient, adaptive health systems worldwide.

The package of HSR indicators is grounded in a functional definition of resilience, encompassing four core capacities: forecasting and preparing for public health needs, maintaining essential health services under all conditions, adapting to evolving demands, and learning from experience while advancing long-term goals. It provides technical specifications for 64 indicators, organized into seven domains: service delivery, health workforce, health information, access to medicines and health technologies, health financing, governance and leadership, and general/composite indicators.

To operationalize these indicators, the package outlines a six-step process: mapping existing frameworks to identify gaps; selecting context-relevant, complementary indicators; defining rational and measurable targets through participatory processes; building data collection capacity at national and subnational levels; conducting ongoing measurement and monitoring; and using data to inform improvements aligned with health sector planning. This approach emphasizes integration, stakeholder engagement, and continuous learning to support resilient, adaptive health systems [56].

This study conceptualizes a framework for assessing HSR across governance, health systems, and economic aspects, employing seventeen indices partially derived from WHO’s HSR indicators. It seeks to construct a comprehensive global profile of resilience and to establish a methodological foundation for advancing the rigor, depth, and interdisciplinary integration of future research in this critical domain.

5. Conclusions, Research Limitations, and Future Suggestions

5.1. Conclusions

The COVID-19 pandemic has vividly demonstrated the ease with which infectious diseases can transcend borders, highlighting the necessity for global collaboration in health security. It underscores the reality that all individuals are stakeholders in a shared public health ecosystem and that coordinated efforts are crucial for fostering safer and more resilient societies [6]. Our paper emphasizes the urgent need to build a globally resilient framework that promotes sustainable and inclusive growth to ensure collective health and well-being. To achieve this, high-income countries should actively support low- and middle-income nations in strengthening health infrastructure, enhancing system adaptability, and improving pandemic preparedness. Greater global cooperation and solidarity are vital for co-creating an interconnected network of robust health systems capable of mitigating future public health crises.

5.2. Research Limitations and Future Suggestions

Due to the challenges associated with large-scale cross-national data collection and the lack of data completeness, the initial analysis herein focuses on governance, health systems, and economic aspects using seventeen indices. Fortunately, in 2024, the WHO introduced a comprehensive framework for HSR, featuring sixty-four indicators across seven domains. This structured package offers a robust resource for measuring and monitoring resilience in both routine operations and public health emergencies, helping to close critical data gaps in evaluating system vulnerabilities and tracking progress [56].

Each country possesses distinct economic, political, cultural, and geographical characteristics that significantly influence resource availability, allocation, and health system approaches. Recognizing these national and regional heterogeneities, future research may benefit from more granular analyses that explore such differences in depth, thereby contributing to a more nuanced understanding of HSR.