Mortality-Prevalence Ratio – a novel index evaluating potentially undocumented COVID-19 infection

JMIR Public Health Surveill. 2020 Dec 14. doi: 10.2196/23034. Online ahead of print.

ABSTRACT

BACKGROUND: The prevalence of COVID-19 has caused 200 thousand death-cases since early 2020. The corresponding mortality rate among different countries and time varies.

OBJECTIVE: This study aims to investigate the relationship between mortality rate and the prevalence of COVID-19 within a country.

METHODS: We collected data from Johns Hopkins Coronavirus Resource Center, which included daily cumulative death count, recovered count, and confirmed count for each country. This study focuses on a total of 36 countries with over 10,000 confirmed cases. Mortality is the main outcome and dependent variable, which is computed as the number of COVID-19 deaths divided by the number of confirmed cases.

RESULTS: The result of global panel regression showed that there was a highly significant correlation between prevalence and mortality (Spearmans^’ correlation=0.8304). We found that every increment of one confirmed case in a thousand individuals leaded to 1.29268% increment in mortality after controlling country-specific baseline mortality and time-fixed effects. Over 70% of excess mortality could be explained by prevalence, and the heterogeneity among countries’ mortality to prevalence (MP) ratio was significant. We further showed that China had an abnormally high and significant MP ratio compared to other countries. This unusual deviation of MP ratio disappeared with the removal of data from China collected after February 17th. It is worth noting that the prevalence of a disease relies on accurate diagnosis and comprehensive surveillance, which could be difficult to achieve due to practical or political concerns.

CONCLUSIONS: The mortality-prevalence association was observed and being quantified as MP ratio. Our results highlight the significance of constraining disease transmission on decreasing mortality. Comparison of MP ratio among countries can be powerful in detecting or even quantifying the proportion of undocumented infected individuals.

PMID:33332282 | DOI:10.2196/23034