Adult mortality before and during the first wave of COVID-19 pandemic in nine communities of Yemen: a key informant study
Mervat Alhaffar,Huda Basaleem,Fouad Othman,Khaled Alsakkaf,Sena Mohammed Mohsen Naji,Hussein Kolaise,Abdullah K Babattah,Yassin Abdulmalik Mahyoub Salem,Hannah Brindle,Najwa Yahya,Pasquale Pepe,Francesco Checchi
DOI: https://doi.org/10.1186/s13031-022-00497-3
2022-12-12
Abstract:Introduction: Widespread armed conflict has affected Yemen since 2014. To date, the mortality toll of seven years of crisis, and any excess due to the COVID-19 pandemic, are not well quantified. We attempted to estimate population mortality during the pre-pandemic and pandemic periods in nine purposively selected urban and rural communities of southern and central Yemen (Aden and Ta'iz governorates), totalling > 100,000 people. Methods: Within each study site, we collected lists of decedents between January 2014-March 2021 by interviewing different categories of key community informants, including community leaders, imams, healthcare workers, senior citizens and others. After linking records across lists based on key variables, we applied two-, three- or four-list capture-recapture analysis to estimate total death tolls. We also computed death rates by combining these estimates with population denominators, themselves subject to estimation. Results: After interviewing 138 disproportionately (74.6%) male informants, we identified 2445 unique decedents. While informants recalled deaths throughout the study period, reported deaths among children were sparse: we thus restricted analysis to persons aged ≥ 15 years old. We noted a peak in reported deaths during May-July 2020, plausibly coinciding with the first COVID-19 wave. Death rate estimates featured uninformatively large confidence intervals, but appeared elevated compared to the non-crisis baseline, particularly in two sites where a large proportion of deaths were attributed to war injuries. There was no clear-cut evidence of excess mortality during the pandemic period. Conclusions: We found some evidence of a peak in mortality during the early phase of the pandemic, but death rate estimates were otherwise too imprecise to enable strong inference on trends. Estimates suggested substantial mortality elevations from baseline during the crisis period, but are subject to serious potential biases. The study highlighted challenges of data collection in this insecure, politically contested environment.