COVID-19 global risk evaluation: rankings, reducing surveillance bias, and infodemic

This study examines how public health institutions estimate regional COVID-19 burdens, pursuing two primary objectives: (1) to analyze the methodologies employed for regional risk assessment, and (2) to perform spatial and Spearman rank correlation analyses of risk metrics that incorporate testing data across 101 countries. Classification methods used to assess COVID-19 risk often treat testing as a secondary, qualitative factor, overlooking its value as a quantitative input. Integrating testing data with case counts can improve the accuracy of regional infection probability estimates. Spatial analysis revealed that probabilistic metrics-such as the local probability of infection-showed stronger spatial synchronization of epidemic patterns compared to observed-to-expected case ratios. The death-to-population ratio displayed the strongest positive correlation with the observed-to-expected cases ratio. Conversely, the case fatality rate exhibited only a weak positive correlation with probabilistic metrics, though these correlations were not consistently statistically significant. The findings underscore the potential of probabilistic metrics, such as the local probability of infection, in predicting COVID-19 risk. Further research is warranted to explore the predictive capacity of probabilistic metrics concerning death-related outcomes.