Paglino Eugenio, Lundberg Dielle J, Zhou Zhenwei, Wasserman Joe A, Raquib Rafeya, Hempstead Katherine, Preston Samuel H, Elo Irma T, Stokes Andrew C
Department of Sociology and Population Studies Center, University of Pennsylvania, Philadelphia, PA.
Department of Global Health, Boston University School of Public Health, Boston, MA.
medRxiv. 2023 Jan 18:2023.01.16.23284633. doi: 10.1101/2023.01.16.23284633.
Accurate and timely tracking of COVID-19 deaths is essential to a well-functioning public health surveillance system. The extent to which official COVID-19 death tallies have captured the true toll of the pandemic in the United States is unknown. In the current study, we develop a Bayesian hierarchical model to estimate monthly excess mortality in each county over the first two years of the pandemic and compare these estimates to the number of deaths officially attributed to Covid-19 on death certificates. Overall, we estimated that 268,176 excess deaths were not reported as Covid-19 deaths during the first two years of the Covid-19 pandemic, which represented 23.7% of all excess deaths that occurred. Differences between excess deaths and reported COVID-19 deaths were substantial in both the first and second year of the pandemic. Excess deaths were less likely to be reported as COVID-19 deaths in the Mountain division, in the South, and in nonmetro counties. The number of excess deaths exceeded COVID-19 deaths in all Census divisions except for the New England and Middle Atlantic divisions where there were more COVID-19 deaths than excess deaths in large metro areas and medium or small metro areas. Increases in excess deaths not assigned to COVID-19 followed similar patterns over time to increases in reported COVID-19 deaths and typically preceded or occurred concurrently with increases in reported COVID-19 deaths. Estimates from this study can be used to inform targeting of resources to areas in which the true toll of the COVID-19 pandemic has been underestimated.
准确及时地追踪新冠死亡病例对于一个运转良好的公共卫生监测系统至关重要。美国官方公布的新冠死亡人数在多大程度上反映了这场疫情的真实死亡人数尚不清楚。在本研究中，我们开发了一个贝叶斯分层模型，以估计疫情头两年每个县的月度超额死亡率，并将这些估计值与死亡证明上官方认定的新冠死亡人数进行比较。总体而言，我们估计在新冠疫情的头两年，有268,176例超额死亡未被报告为新冠死亡病例，占所有超额死亡病例的23.7%。在疫情的第一年和第二年，超额死亡与报告的新冠死亡之间的差异都很大。在山区、南部以及非都市县，超额死亡被报告为新冠死亡的可能性较小。除新英格兰和中大西洋地区外，在所有人口普查分区中，超额死亡人数都超过了新冠死亡人数，在这些地区，大都市地区以及中小都市地区的新冠死亡人数超过了超额死亡人数。未被认定为新冠死亡的超额死亡人数随时间的增加模式与报告的新冠死亡人数增加模式相似，并且通常先于报告的新冠死亡人数增加或与之同时发生。本研究的估计结果可用于指导将资源投向那些新冠疫情真实死亡人数被低估的地区。
