Age-specific mortality and immunity patterns of SARS-CoV-2.

Area of Science:

• Epidemiology
• Infectious Disease Modeling
• Public Health

Background:

• Estimating the scale of the COVID-19 pandemic and SARS-CoV-2 infection severity is challenging due to data inconsistencies.
• Observed COVID-19 deaths represent a fraction of infections, and variations in reporting, especially for older individuals and in nursing homes, complicate comparisons.
• Direct comparisons of mortality rates and transmission levels across countries are hindered by heterogeneous burdens in long-term care facilities and variable reporting practices.

Purpose of the Study:

• To investigate the consistency of infection and fatality patterns across multiple countries using age-specific COVID-19 death data and seroprevalence studies.
• To develop robust estimates of population infection rates and infection fatality ratios (IFRs) by leveraging consistent age-specific mortality data.
• To assess the impact of population age structures and nursing home burdens on the heterogeneity of IFRs between countries.

Main Methods:

• Utilized age-specific COVID-19-associated death data from 45 countries.
• Incorporated results from 22 seroprevalence studies to cross-validate findings.
• Developed a modeling framework to estimate infection rates and IFRs, analyzing age distributions and population structures.

Main Results:

• The age distribution of deaths in individuals under 65 years was consistent across settings, enabling reliable estimation of population infection shares.
• The infection fatality ratio showed a log-linear increase with age after 30 years, with the lowest rates in 5-9-year-old children.
• Estimated that approximately 5% of the population in the 45 included countries had been infected by September 1, 2020, with higher transmission rates suggested in some Latin American countries.

Conclusions:

• Age-specific death data, particularly in younger age groups, provide a robust method for estimating pandemic progression and population infection rates.
• Population age structure and nursing home dynamics partially explain variations in IFRs between countries, but other factors likely contribute.
• The developed modeling framework offers a valuable tool for countries to assess their pandemic status using readily available age-specific death data.