Food insecurity and associated factors during the COVID-19 pandemic in a vulnerable population in Rio de Janeiro: A primary care registry-based survey

Despite documented increases in food insecurity (FI) during the COVID-19 pandemic, the specific impact on highly vulnerable households within distinct urban contexts, as well as the interplay of various household-level factors in exacerbating this vulnerability, remain understudied. This study investigated FI prevalence and household-level factors associated with moderate/severe FI in Complexo de Manguinhos, a complex of slums in Rio de Janeiro, Brazil, during the COVID-19 pandemic. The Comvida-1 survey was conducted between September/2020 and February/2021 and included individuals aged ≥1 year. FI was measured using the Brazilian Food Insecurity Scale. Multiple Correspondence Analysis (MCA) identified associations between household variables (family composition, income, welfare benefits, job/income losses). Multilevel logistic regression assessed associations between the dimensions identified by the MCA and moderate/severe FI. Among 3864 households, moderate/severe FI prevalence was 14.7% (95% confidence interval [CI]: 13.5-15.9) and ranged from 4 to 28% across the 16 slums comprised in the Complexo de Manguinhos. Two MCA dimensions were associated with FI: Dimension 1 represented chronic socioeconomic vulnerability, with highest contributions from Bolsa Família benefit (10.73%), multiple children in Bolsa Família (9.56%), and presence of children in the household (6.84%). Dimension 4 captured acute COVID-19 economic impacts, including job/income loss (17.24%), reduced working hours (11.44%), and reduced family income (11.30%). Each dimension increased the odds of moderate/severe FI by approximately three-fold (Dimension 1 adjusted odds ratio [aOR]=2.72, 95%CI: 2.19-3.38; Dimension 4 aOR=3.32, 95%CI: 2.42-4.55, both p