An ecological study design was used in this study to investigate the correlation between IHR core capacity scores and the UHC service coverage index (Fig. 1). The study is based on analyses of data collected from 96 countries for JEE, SPAR, and the UHC service coverage index. The IHR core capacity scores, which were measured by JEE 2016-2019 and SPAR 2019 from the World Health Organization (WHO), were used as independent variables. The dependent variable, i.e., the UHC service coverage index, was extracted from the 2019 UHC monitoring report.
This study identified the confounding variables that may affect UHC service coverage. This study selected the confounding variables that may affect UHC service coverage. UHC supports the idea that health service is available to everyone without causing financial hardship. Tracking the progress towards UHC uses two specific indicators, health services coverage and financial risk protection coverage. Previous studies on UHC identified that socio-demographic index, government health expenditure and governance show positive association with UHC service coverage. The selection principle of confounding variables is related to both core explanatory variables and dependent variables. Thus, the confounding variables in this study include GDP per-capita, current health expenditure, infant mortality rate, life expectancy at birth, hospital beds, medical doctors, nursing and midwifery personnel, population ages under five, population ages 65 and above [9, 10].
Four models were demonstrated to understand the factors of JEE and other independent variables affecting UHC service coverage, and SPAR and other independent variables affecting UHC service coverage. Model 1 includes the global health security index of either JEE or SPAR overall mean score and population variables (population age under five, population age 65 and above) as independent variables, Model 2 incorporates the variables used in Model 1 and economic variables (GDP per capita, current health expenditure), model 3 includes the variables used in model 2 and variables related to medical resources (hospital beds, medical doctors, nursing, and midwifery personnel), and Model 4 includes the variables used in Model 3 and variables related to health status (infant mortality rate, life expectancy at birth).
JEE 2016-2019, SPAR 2019, and UHC service coverage index 2017 data from 1st March, 2021 to 31st March, 2021 were extracted [13,14,15]. Online databases from the World Bank, WHO Global Health Observatory, and United Nations provided GDP per capita, current health expenditure, infant mortality rate, life expectancy at birth, hospital beds, medical doctors, nursing and midwifery personnel, population age under five, population age 65 and above. GDP per capita, current health expenditure, and infant mortality rates were available from the World Bank [16,17,18]. Data on life expectancy at birth, hospital beds, medical doctors, and nursing and midwifery personnel were available from the WHO Global Health Observatory [19,20,21,22]. The United Nations website was the other source of data for population age under five and population age 65 and above [23, 24] (Table 1).
The JEE tool contains 19 technical areas, represented by 48 indicators. Each technical area represents the mean scores of the indicators. The indicator’s scoring system is based on a five-point ordinal scale from 1 to 5, reflecting higher capacity as the score increases. The SPAR tool consists of 24 indicators from the 13 IHR capacities needed to detect, assess, notify, and respond to public health events of national and international concern. The level of SPAR is expressed as the average of all indicators, which are calculated as a percentage of performance based on a scale of 1 to 5. Recognizing the conceptual similarities between JEE and SPAR, the technical areas were matched and categorized into 15 areas. The overall mean values of JEE and SPAR were retrieved and used for the statistical analysis, except for the spider diagram comparing JEE and SPAR, where the value of JEE was multiplied by 20 to ensure that both JEE and SPAR could have the same scale.
Statistical analyses were performed using SPSS version 25.0. A spider diagram was used to visually explain the relationship between the JEE and SPAR scores. The correlation analysis between JEE and SPAR further supported this relationship. Furthermore, Pearson correlation analysis was used to test the association between the IHR core capacity scores and the UHC service coverage index. Descriptive analysis and one-way analysis of variance (ANOVA) were used to describe the general characteristics of the selected countries and compare the JEE scores, SPAR scores, and the UHC service coverage index by population, economic index, human resources for health, and health indicators of countries. The scatterplots between the JEE score and the UHC service coverage index, as well as the SPAR score and the UHC service coverage index, were presented to support this relationship. Lastly, multiple regression analysis was used to test the effect of the IHR core capacity scores on the UHC service coverage index. We used a variance inflation factor (VIF) to confirm that multicollinearity did not occur between the explanatory variables (VIF < 10). The level of statistical significance was set at p < 0.05.