Buszkiewicz James H, Tseng Ashley S, Dai Jane, Ismach Alan, Beese Shawna, Collier Sarah M, Spiker Marie L, Otten Jennifer J
Department of Epidemiology, Center for Social Epidemiology and Population Health, School of Public Health, University of Michigan, Ann Arbor, Michigan, United States of America.
Department of Epidemiology, School of Public Health, University of Washington, Seattle, Washington, United States of America.
PLoS One. 2025 May 14;20(5):e0321585. doi: 10.1371/journal.pone.0321585. eCollection 2025.
Research on COVID-19's impact on food insecurity has primarily relied on cross-sectional data or long recall periods, with limited investigations into longitudinal patterns or the role of food assistance.
We analyzed longitudinal data from 703 respondents participating in at least three Washington State Food Security Survey waves between June 18, 2020, and January 7, 2023. We assessed food security using the United States Department of Agriculture's six-item module, categorizing respondents' trajectories as persistently food secure, persistently food insecure, or experiencing one or more food insecurity transitions. We categorized food assistance use as never used, used before COVID-19 but not at baseline, did not use before COVID-19 but used at baseline, or always used. We descriptively examined sociodemographic factors linked to each food security trajectory and food assistance use pattern. We assessed associations between food assistance use and food security trajectories using modified Poisson regression.
We found that 20.2% of respondents were persistently food insecure, and 22.5% experienced one or more food insecurity transitions. Both patterns were more common among respondents who were aged 35 to 64, had a gender identity other than man or woman, were non-Hispanic Black, were single or divorced, had children, had some college education or less, reported $35,000 or less in household income, or were unemployed. In fully adjusted models, respondents who were newly using food assistance early in the COVID-19 pandemic had a higher probability of being persistently food insecure (marginal effect [ME] = 0.320, 95% CI = 0.204, 0.436) or experiencing one or more food insecurity transitions (ME = 0.216, 95% CI = 0.069, 0.363), than those who never used assistance.
Our findings highlight the importance of examining food security trajectories and food assistance use patterns and implementing policies that help households new to food assistance programs navigate these systems.
关于新冠疫情对粮食不安全影响的研究主要依赖横断面数据或较长的回忆期，对纵向模式或粮食援助作用的调查有限。
我们分析了703名受访者的纵向数据，这些受访者在2020年6月18日至2023年1月7日期间至少参与了三轮华盛顿州粮食安全调查。我们使用美国农业部的六项模块评估粮食安全状况，将受访者的轨迹分类为持续粮食安全、持续粮食不安全或经历一次或多次粮食不安全转变。我们将粮食援助的使用情况分类为从未使用、在新冠疫情之前使用但基线时未使用、在新冠疫情之前未使用但基线时使用或一直使用。我们描述性地研究了与每种粮食安全轨迹和粮食援助使用模式相关的社会人口因素。我们使用修正泊松回归评估粮食援助使用与粮食安全轨迹之间的关联。
我们发现，20.2%的受访者持续粮食不安全，22.5%的受访者经历了一次或多次粮食不安全转变。这两种模式在年龄在35至64岁、性别认同非男非女、非西班牙裔黑人、单身或离异、有子女、接受过一些大学教育或更少、家庭收入报告为35,000美元或更少或失业的受访者中更为常见。在完全调整的模型中，在新冠疫情早期新使用粮食援助的受访者比从未使用过援助的受访者更有可能持续粮食不安全（边际效应[ME]=0.320，95%置信区间=0.204，0.436）或经历一次或多次粮食不安全转变（ME=0.216，95%置信区间=0.069，0.363）。
我们的研究结果强调了检查粮食安全轨迹和粮食援助使用模式以及实施政策以帮助新加入粮食援助计划的家庭适应这些系统的重要性。
