Harrigan Sean P, Fibke Chad D, Velásquez García Héctor A, Mak Sunny, Wilton James, Prystajecky Natalie, Tyson John, Wang Linwei, Sander Beate, Baral Stefan, Mishra Sharmistha, Janjua Naveed Z, Sbihi Hind
British Columbia Centre for Disease Control, Vancouver, British Columbia, Canada.
University of British Columbia Centre for Disease Control, Vancouver, British Columbia, Canada.
Am J Epidemiol. 2025 May 7;194(5):1352-1361. doi: 10.1093/aje/kwae266.
The mechanisms facilitating the relationship between low income and COVID-19 severity have not been partitioned in the presence of SARS-CoV-2 variants of concern (VOCs). To address this, we used causal mediation analysis to quantify the possible mediating role infection with VOC has on the relationship between neighborhood income (exposure) and hospitalization due to COVID-19 among cases (outcome). A population-based cohort of 65 629 individuals residing in British Columbia, Canada, was divided into 3 periods of VOC co-circulation in the 2021 calendar year, whereby each period included co-circulation of an emerging and an established VOC. Each cohort was subjected to g-formula mediation techniques to decompose the relationship between exposure and outcome into total, direct, and indirect effects. In the mediation analysis, the total effects indicated that low income was associated with increased odds of hospitalization across all periods. Further decomposition of the effects revealed that income is directly and indirectly associated with hospitalization. The resulting indirect effect through VOC accounted for approximately between 6% and 13% of the total effect of income on hospitalization. This study underscores, conditional on the analysis, the importance of addressing underlying inequities to mitigate the disproportionate impact on historically marginalized communities by adopting an equity lens as central to pandemic preparedness and response from the onset.
在存在值得关注的严重急性呼吸综合征冠状病毒2（SARS-CoV-2）变异株（VOC）的情况下，低收入与新冠病毒病（COVID-19）严重程度之间关系的促进机制尚未得到区分。为解决这一问题，我们采用因果中介分析来量化感染VOC对社区收入（暴露因素）与COVID-19病例住院（结局）之间关系可能产生的中介作用。对居住在加拿大不列颠哥伦比亚省的65629名个体组成的基于人群的队列，按照2021年日历年内VOC共同流行的3个时期进行划分，每个时期包括一种新出现的VOC和一种已存在的VOC共同流行。对每个队列采用g公式中介技术，将暴露因素与结局之间的关系分解为总效应、直接效应和间接效应。在中介分析中，总效应表明低收入与所有时期住院几率增加相关。效应的进一步分解显示，收入与住院存在直接和间接关联。通过VOC产生的间接效应约占收入对住院总效应的6%至13%。本研究在分析的条件下强调，从一开始就采用公平视角作为大流行防范和应对的核心，对于解决潜在的不公平现象以减轻对历史上边缘化社区的不成比例影响至关重要。
