Evaluating the Health Systems Efficiency of Countries Infected by COVID-19 Using Window Data Envelopment Analysis, Malmquist Productivity Index, and Data Envelopment Analysis with Non-Controllable Input Variables

This paper aims to evaluate the response of different countries against Covid-19 for three consecutive periods from January 2020 to October 2022. 16 countries have been chosen based on their differences in various demographic, cultural, and geographical characteristics.

Design/methodology/approach:

 Window Data Envelopment Analysis (DEA) model, Malmquist Productivity Index and DEA with non-controllable input variables have been used over three periods of 345 days to evaluate the efficiency of health systems of the countries. By the Window DEA and DEA with non-controllable input variables, the efficiency of the countries in several different periods has been compared. Also, the scores for the progress of countries' performance during successive periods have been computed by the Malmquist Productivity Index.

The performance scores calculated with the window DEA indicated that there is a difference in the countries' responses. Based on the obtained results, China had the most efficiency and Italy had the worst efficiency during the three periods. Based on the model of uncontrollable variables, Egypt had the best performance, although the model did not distinguish among most countries. The results of the Malmquist Productivity Index method also indicated that countries had different efficiencies and progress. Based on its results, Malaysia had the most progress, while Singapore had the least progress during the three periods. China despite its high-performance score from the window DEA, had not received a good score from the Malmquist Productivity Index and non-controllable input variables model, which implies the low growth and even retrogression of this country during the three periods. According to the results of the window DEA, Italy had progressed, but this growth has not been enough to make this country's performance efficient.

Research limitations/implications:

 The selection of different and more precise time intervals based on COVID-19 variants (alpha, delta and micron) and a wider selection of countries can be considered in future studies. Considering the amount of vaccination or the type of vaccine and observing its effects dynamically on the DEA model can be also a potential direction for future research. The development of window DEA and the Malmquist Productivity Index for uncontrollable variables and their application to the data of this paper are also suggested. Besides, the selection of input and output indicators based on other goals can be investigated in future studies.

Practical implications: 

To increase countries’ efficiency in dealing with epidemics such as COVID-19, different countries need to know as much as possible about their current situation. One of the ways to get this recognition is to compare them with other countries with the best performance. In such conditions, finding relative efficiency and trying to make the inefficient ones efficient is a solution to increase efficiency. It is possible to identify and extract coping strategies from successful countries with good performance in dealing with COVID-19, and suitable strategies can be implemented in Iran according to the required facilities and infrastructure.

Social implication:

 Appropriate reactions and successful strategies of health systems of countries in dealing with epidemic diseases such as COVID-19 can be very effective in controlling such diseases and reducing their negative social and economic impacts. The results of evaluating the efficiency of different countries along with the review of their strategies can be helpful in this direction.

Originality/value: 

An attempt has been made to present a new picture of countries' response to the COVID-19 pandemic by selecting a diverse range of countries, indicators, and relatively long periods using DEA. Windowed DEA, Malmquist productivity index and DEA with uncontrollable input variables in a period of almost three years were used for the first time to evaluate the efficiency of countries' health systems and their progress in dealing with COVID-19.