COVID-19 mortality rates differ across countries. We aimed to construct a model that predicts mortality worldwide, by including only country-level socioeconomic and health system indicators and excluding variables related to short-term measures for pandemic management. Methods: COVID-19 mortality data was collected from Johns Hopkins University resource center. Additional sources were public reports from the United Nations, the World Bank and the Heritage Foundation. We implemented multiple

more »

... ar regression with backward elimination on the selected predictors. Results: The final model constructed on seven Independent variables, significantly predicted COVID-19 mortality rate by country (F-statistic: 29.2, p<0.001). Regression coefficients (95% CI) in descending order of standardized effects: Annual tourist arrivals: 5.43 (4.03, 6.83); health expenditure per capita: 4.43 (2.92, 5.96); GDP (PPP): -4.60 (-6.81, -2.38); specialist surgical workforce per 100000: 2.63 (0.67, 4.59); number of physicians per 1000: -2.32 (-4.3, -0.28); economic freedom score: -1.35 (-2.60, -0.10); and total population: 1.66 (-0.19, 3.52). All VIF values were below 5, showing acceptable collinearity. R-squared (52.65%), adjusted R-squared (50.25%) and predicted R-squared (42.33%) showed strong model fit. Conclusion: limited country-level socioeconomic and health system indicators can explain COVID-19 mortality worldwide; emphasizing the priority of attending to these fundamental structures when planning for pandemic preparedness.