Analysis of self-assessment capacity scores related to infectious disease control in the International Health Regulations during the first year of the COVID-19 pandemic
The framework for this study was adopted from the Systemic Rapid Assessment Toolkit (SYSRA). We have adopted the SYSRA because of its consistency with the requirements of the countries in the implementation of the IHR. In SYSRA, there are two types of evaluation; horizontal and vertical. The “Horizontal Assessment” analyzes the health system in which the infectious disease program is integrated from various perspectives. While the second element, ‘vertical assessment’, is used to assess the infectious disease specific component. Thus, both elements index the external environment (political, socio-demographic, economic) and the assessment of needs (morbidity, mortality of the disease) as a consideration in the evaluation of programs for the control of infectious diseases.2,24,25.
Self-assessment capability scores (e-SPAR) related to infectious disease control
To identify changes in country self-assessment capacity (e-SPAR) scores related to infectious disease control during the first year of the COVID-19 pandemic, we calculated the absolute difference between the RSI e-SPAR score in 2019 and 2020 We collected this data from the WHO website in May 2021. There were a total of 13 items in the e-SPAR, including legislation and funding, RSI coordination and national RSI focal point function, zoonotic events and the human-animal interface, food safety, laboratory, surveillance, human resources, national health emergency framework, health service delivery, hazard communication, points of entry, chemical events and radiological emergencies26. Since our study focused on infectious disease control, we excluded chemical event and radiological emergency capacity scores for analysis and used only 11 of the 13 items in e-SPAR.
Case fatality rate (CFR) of COVID-19
We used deaths instead of cases to reduce bias in the data we analyzed. Indeed, there are 3 levels used in the diagnosis of COVID-19 cases; suspected, probable and confirmed cases. Thus, data on the number of confirmed cases of COVID-19 will fluctuate and be unstable due to changes in a patient’s diagnostic status.27. Additionally, in reporting COVID-19 deaths, only deaths confirmed to be caused by COVID-19 were reported.27. As the pandemic is still ongoing, we used the case fatality rate (CFR) from COVID-19 data through March 31, 2021 to represent the CFR of COVID-19 in one year. We have collected the CFR of COVID-19 up to March 31, 2021 from the “Our World in Data” website.27. While until the end of 2021, the CFR of COVID-19 varied from 2 to 3% worldwide28.29, in this study, we used 2.08% as the threshold to classify countries into high or low CFR groups. This number results from the calculation of the average CFR of all the countries that we included in the analysis.
Income level of a country
Countries’ income levels were determined by their gross national income (GNI) per capita. We collected the data for 2019 from the World Development Index on the World Bank website. The income level of a country is determined by the GNI per capita of the country30. We have also adopted the income level classification of countries defined by the World Bank for the analysis, which are low-income countries (LICs), lower-middle-income countries (LMICs), low-income countries upper-middle (PRHM) and high-income countries (PRE)31.
Human Development Index (HDI)
We have used the HDI as an indicator to represent the development levels of countries which reflect the social and environmental status of the country32. HDI data was collected from the Health Development Report (HDR) 2020 on the United Nations Development Program (UNDP) website33. The UNDP HDI classification was used for analysis in the study. Low-development countries are defined as those with index scores below 0.55; while medium, high and very high development countries with scores between 0.55 and 0.69; 0.7 to 0.79; and greater than 0.8 respectively32.34.
Civil liberties (LC)
Although countries’ transparency was reported to be associated with their scores reported by the previous study, we also collected CL score data from the Freedom House website for analysis.2. And the CL level category was also adopted in the study for analysis. “Not free” countries are those whose score is from 0 to 35. While “partially free” and “free” countries are those whose scores are respectively from 35 to 70 and above 70.35,36,37.
Government Effectiveness (GE)
The GE is one of the components of the World Governance Indicators (WGI). This was the indicator reflecting the quality of public services, policy formulation and implementation. We chose GE as one of the variables because the literature mentions that the role of government and good governance were very important in infectious disease prevention and control efforts.38.39. GE data was collected from WGI Project 2020 reports40. Since the GE scores for each country in the report range from -2.5 to 2.5, we have categorized this variable into 2 categories by setting 0 as the cutoff point. Thus, a country with a GE score higher than 0 means that the country has a strong GE, and conversely a country with a GE score lower than 0 is classified as a country with a weak GE.
196 countries reported their e-SPAR scores in 2019 and 2020. Of these, only 154 countries with complete data on all indicators were used for the analysis. We calculated the countries’ average scores on 11 e-SPAR capabilities as well as their average score for each capability in 2019 and 2020, then calculated their absolute deviations over these two consecutive years. Additionally, since we found that the data was not normally distributed, we performed the Wilcoxon Sign-rank test to assess the significance of the difference between the scores.
Next, we divided the 154 countries into two groups based on their score classification, namely the group whose scores increased (n=98) and the group whose scores did not increase (n=56) for a more in-depth analysis. A chi-square test was applied to identify the independence of countries’ e-SPAR scores from their income, HDI, CL, GE and COVID-19 CFR levels. Then we performed a multiple linear regression analysis41 to determine which factors were associated with changes in e-SPAR scores. In the model, the difference between the average e-SPAR scores in 2020 and 2019 was the dependent variable (Y), while the CFR of COVID-19, HDI, CL and GE were the independent variables (X). We did not include country income levels in the model due to its significant correlation with the HDI index. We developed three models in our multiple linear regression analysis to determine which factors were associated with changes in e-SPAR scores across the 154 countries (model 1), countries whose scores increased (model 2), and countries whose scores did not increase (Model 3). In the analysis, we are looking for the adjusted R2 value to represent the proportion of the variance of a dependent variable that is explained by independent variables. We used a p-value of less than 0.05 as the statistical significance level to reject the null hypothesis. All analyzes were performed using SPSS software, version 18.