Estimates of the COVID-19 Infection Fatality Rate for 48 African Countries: A Model-based Analysis
23 Pages Posted: 10 Aug 2020
Date Written: July 18, 2020
Introduction: The infection fatality rate (IFR) is key to determining the effect of the pandemic at population level, as well as the effects of public policies and regulations. We examine global data from 48 African countries to estimate the SARS-CoV-2 IFR.
Methods: We analyzed time series data on the confirmed cases and deaths from COVID-19 disease outbreak across Africa. We define IFR as the ratio of the number of deaths caused by COVID-19 (numerator) and the total number of people in the population who were infected by the virus (denominator). We controlled for the upward bias associated with the denominator, to accommodate for the untested individuals by adjusting for population density, population aged 65 years and older, population with basic handwashing facilities, extreme poverty, diabetes prevalence, and death rate from cardiovascular disease in a Bayesian prediction model based on the technique of Monte Carlo.
Results: We analyzed data on the 135,126 confirmed cases and 3,922 deaths from COVID-19 disease outbreak in Africa through May 30, 2020. After adjusting for potential risk factors in the Bayesian model, we predicted a total of 1,686,879 COVID-19 infections, corresponding to 13 infections per confirmed case. The IFR in Africa was estimated to be 0.23% (95%CI: 0.14% to 0.33). Country-specific rates varied from 0.004% in Botswana and Central African Republic, to 1.53% in Nigeria, respectively. The estimated IFR is twelvefold higher than WHO reported estimate (0.02%) from the 2009 H1N1 influenza pandemic. The inverse distance weighted (IDW) interpolation map shows concentrations of extreme IFR in four countries: Morocco, Nigeria, Cameroon, and South Africa.
Conclusion: The infection fatality rate of COVID-19 can vary substantially across different locations, and this may reflect differences in demographics, underlying health issues in the population, capacity of the healthcare system, positive health seeking behavior, as well as other factors. Variability in testing and cause-of-death data across African countries might have impacted on the results. Our model and our estimates can help disease and policy modelers to obtain more accurate predictions for the epidemiology of the disease and the effect of alternative policy strategies to contain this pandemic.
Note: Funding: O Keiser was funded by grants from the Swiss National Science Foundation(no 163878 and 196270).
Declaration of Interest: None to declare
Keywords: COVID-19, Infection Fatality Rate, Bayesian prediction, Monte Carlo method, Influenza, Africa
Suggested Citation: Suggested Citation