An assessment of excess mortality during the COVID-19 pandemic, a retrospective post-mortem surveillance in 12 districts – Zambia, 2020-2022

Abstract Background The number of COVID-19 deaths reported in Zambia (N=4,069) is most likely an underestimate due to limited testing, incomplete death registration and inability to account for indirect deaths due to socioeconomic disruption during the pandemic. We sought to assess excess mortality during the COVID-19 pandemic in Zambia. Methods We conducted a retrospective analysis of monthly-death-counts (2017-2022) and individual-daily-deaths (2020-2022) of all reported health facility and community deaths at district referral health facility mortuaries in 12 districts in Zambia. We defined COVID-19 wave periods based on a sustained nationally reported SARS-CoV-2 test positivity of greater than 5%. Excess mortality was defined as the difference between the pandemic monthly death counts (2020-2022) and the median pre-pandemic monthly death counts (2017-2019) using a Microsoft Excel-based tool. We compared median daily death counts, median age at death, and the proportion of deaths by place of death (health facility vs community) by wave period using the Mann-Whitney-U test and chi-square test respectively in R. Results A total of 112,768 deaths were reported in the 12 districts between 2020 and 2022, of which 17,111 (15.2%) were excess. Wave periods had higher median daily death counts than non-wave periods (median [IQR], 107 [95-126] versus 96 [85-107], p<0.001). The median age at death during wave periods was older than non-wave periods (44.0 [25.0-67.0] versus 41.0 [22.0-63.0] years, p<0.001). Approximately half of all reported deaths occurred in the community, with an even greater proportion during wave periods (50.58% versus 53.11%, p<0.001), respectively. Conclusion There was excess mortality during the COVID-19 pandemic in Zambia, with more deaths among the elderly, and occurring within the community during wave periods. This analysis suggests more COVID-19 deaths likely occurred in Zambia than suggested by officially reported numbers. Mortality surveillance can provide important information to monitor population health and inform public health programming during pandemics.