Abstract
Food insecurity increased during the COVID-19 pandemic, but the impact varied across different age groups during the prolonged public health emergency. This study sought to describe national food insecurity prevalence by adult age group at multiple stages of the pandemic and explore differences by demographic characteristics. Data were from the nationally representative US Census Bureau's Household Pulse Survey from April 2020 to May 2023 (N = 4,153,462). Locally weighted scatterplot smoothing (LOESS) regression analysis identified change points in food insecurity trends, segmenting the timeline into three periods: (1) April 2020-March 2021, (2) April 2021-May 2022, and (3) June 2022-May 2023. Logistic regression models examined associations between age, time period, and self-reported household food insecurity; covariates included demographics, socioeconomic status, household structure, and food support program usage. Overall, 9.3% of respondents experienced food insecurity, ranging from 3.5% among those aged ≥75 to 12.2% for ages 35-44 years. Significant interaction between age group and time period indicated inconsistency in the age-food insecurity association during the pandemic (p < 0.001). From Period 1 to 3, the proportion of food-insecure adults aged ≥65 rose from 9.2% to 13.9%. Across all age groups, higher odds of food insecurity were found among Black, Hispanic/Latino, or Other/Multiracial respondents; those with less than a Bachelor's degree; those with incomes below USD 35,000; those unemployed for reasons other than retirement; and non-homeowners (p < 0.001). The results show that trends and characteristics associated with food insecurity varied across age groups and time periods. Continuous monitoring of food insecurity during emergencies is critical to identify vulnerable populations and timely interventions.