The Global Case-Fatality Rate of COVID-19 Has Been Declining Since May 2020

• Published in: The American journal of tropical medicine and hygiene Vol. 104; no. 6; pp. 2176 - 2184
• Main Authors: Hasan, Mohammad Nayeem, Haider, Najmul, Stigler, Florian L., Khan, Rumi Ahmed, McCoy, David, Zumla, Alimuddin, Kock, Richard A., Uddin, Md. Jamal
• Format: Journal Article
• Language: English
• Published: United States Institute of Tropical Medicine 01.06.2021; The American Society of Tropical Medicine and Hygiene
• Subjects: Coronaviruses; COVID-19; COVID-19 - epidemiology; COVID-19 - mortality; COVID-19 diagnostic tests; COVID-19 Testing; Fatalities; Global Health; Humans; Incidence; SARS-CoV-2; Time Factors
• ISSN: 0002-9637; 1476-1645; 1476-1645
• DOI: 10.4269/ajtmh.20-1496

The objective of this study was to evaluate the trend of reported case fatality rate (rCFR) of COVID-19 over time, using globally reported COVID-19 cases and mortality data. We collected daily COVID-19 diagnoses and mortality data from the WHO’s daily situation reports dated January 1 to December 31, 2020. We performed three time-series models [simple exponential smoothing, auto-regressive integrated moving average, and automatic forecasting time-series (Prophet)] to identify the global trend of rCFR for COVID-19. We used beta regression models to investigate the association between the rCFR and potential predictors of each country and reported incidence rate ratios (IRRs) of each variable. The weekly global cumulative COVID-19 rCFR reached a peak at 7.23% during the 17th week (April 22–28, 2020). We found a positive and increasing trend for global daily rCFR values of COVID-19 until the 17th week (pre-peak period) and then a strong declining trend up until the 53rd week (post-peak period) toward 2.2% (December 29–31, 2020). In pre-peak of rCFR, the percentage of people aged 65 and above and the prevalence of obesity were significantly associated with the COVID-19 rCFR. The declining trend of global COVID-19 rCFR was not merely because of increased COVID-19 testing, because COVID-19 tests per 1,000 population had poor predictive value. Decreasing rCFR could be explained by an increased rate of infection in younger people or by the improvement of health care management, shielding from infection, and/or repurposing of several drugs that had shown a beneficial effect on reducing fatality because of COVID-19.