The age-dependent risk of SARS-CoV-2 infection resulting in death is consistent across countries for people below the age of 65, but the relative risk in older individuals varies considerably between countries, according to a new modelling study. The numbers of deaths in over-65s (an established risk factor for COVID-19-related deaths) from European countries, for example, are substantially higher than expected. The study, which uses data from 45 countries, provides information that can be used to assess countries’ infection levels and response effectiveness.
Monitoring the levels of SARS-CoV-2 transmission and severity of infection is an important tool for guiding responses to the pandemic, but can be challenging because of inconsistencies in the available data. Serological surveys, which estimate the percentage of people in a population who have antibodies against SARS-CoV-2, may provide variable data depending on testing capabilities.
Data on deaths are considered a reliable indicator of SARS-CoV-2 transmission rates and infection severity, although these data may also be inconsistent, depending on the locations of outbreaks. Care homes, for example, have anomalously high transmission and death rates compared with the general population.
To investigate patterns of infection and fatality, Megan O’Driscoll and colleagues at the department of genetics, University of Cambridge, and the Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, created a model using COVID-19 death data from 45 countries and results from 22 national-level serological surveys. They provide an estimate of the infection fatality ratio – the proportion of people infected with the virus that are likely to die as a result – which can be used to compare reported and expected numbers of deaths. The relative risk of death by age in under 65s is shown to be consistent across countries; the lowest risk of infection-related deaths is in children aged 5–9 years old, with the infection fatality ratio increasing by 0.59% for every 5-year increase in age.
The authors estimated the number of deaths expected in the absence of nursing home transmission for people aged over 65 years. They then compared these estimates with reported deaths in this age group and identified disparities in numbers between countries. For example, South America reported fewer deaths in individuals over 65 than expected, which is probably due to the under-reporting of COVID-19 deaths among the elderly. Meanwhile, in most European countries, the number of reported deaths among the over 65s was higher than expected, as a result of the many outbreaks in care homes.
These findings demonstrate how combining the results of serological surveys and death data can be used to assess trends in the transmission and fatality rates of SARS-CoV-2 infection. Model-based estimates of infection rates may help to identify countries in which excess deaths in vulnerable populations may have occurred, the authors conclude.
Estimating the size and infection severity of the SARS-CoV-2 epidemic is made challenging by inconsistencies in available data. The number of COVID-19 deaths is often used as a key indicator for the epidemic size, but observed deaths represent only a minority of all infections1,2. Additionally, the heterogeneous burden in nursing homes and variable reporting of deaths in elderly individuals can hamper direct comparisons across countries of the underlying level of transmission and mortality rates3. Here we use age-specific COVID-19 death data from 45 countries and the results of 22 seroprevalence studies to investigate the consistency of infection and fatality patterns across multiple countries. We find that the age distribution of deaths in younger age groups (<65 years) is very consistent across different settings and demonstrate how this data can provide robust estimates of the share of the population that has been infected. We estimate that the infection-to-fatality ratio (IFR) is lowest among 5-9 years old, with a log-linear increase by age among individuals older than 30 years. Population age-structures and heterogeneous burdens in nursing homes explain some but not all of the heterogeneity between countries in infection-fatality ratios. Among the 45 countries included in our analysis, we estimate approximately 5% of these populations had been infected by the 1st of September 2020, with much higher transmission likely to have occurred in a number of Latin American countries. This simple modelling framework can help countries assess the progression of the pandemic and can be applied wherever reliable age-specific death data exists.
Megan O’Driscoll, Gabriel Ribeiro Dos Santos, Lin Wang, Derek AT Cummings, Andrew S Azman, Juliette Paireau, Arnaud Fontanet, Simon Cauchemez, Henrik Salje