Bakhtiari Elyas
Department of Sociology, William and Mary, Williamsburg, VA, USA.
J Racial Ethn Health Disparities. 2025 Aug 15. doi: 10.1007/s40615-025-02589-1.
Many ethno-racial minority populations had disproportionately high rates of COVID-19 infection and mortality during the initial years of the pandemic, but it is currently unclear whether Middle Eastern and North African populations in the USA experienced similar inequalities. This study disaggregates the MENA population in Virginia death records using a long short-term memory machine learning model that can probabilistically identify ethnically-unique names. The results suggest that MENA populations had higher overall excess mortality and a higher percentage of COVID-19 deaths among total deaths, when compared to the non-Hispanic White population of Virginia. The paper highlights the importance of disaggregating the MENA population in health inequalities research and offers a new method that improves upon previous name-based disaggregation methods.
在新冠疫情的最初几年，许多少数族裔群体的新冠病毒感染率和死亡率高得不成比例，但目前尚不清楚美国的中东和北非人群是否也经历了类似的不平等情况。本研究使用一种长短期记忆机器学习模型对弗吉尼亚州死亡记录中的中东和北非人群进行分类，该模型可以概率性地识别具有种族独特性的名字。结果表明，与弗吉尼亚州的非西班牙裔白人相比，中东和北非人群的总体超额死亡率更高，且新冠死亡人数在总死亡人数中所占百分比更高。本文强调了在健康不平等研究中对中东和北非人群进行分类的重要性，并提供了一种改进以往基于名字的分类方法的新方法。
