This study aimed to characterize users of a government soup-kitchen program and the association with family food insecurity, using a cross-sectional design and random sample of 1,637 soup-kitchen users. The study used a questionnaire with socioeconomic variables and the Brazilian Food Insecurity Scale, and measured weight and height. The chi-square test was applied, and the crude and adjusted prevalence ratios (PR) were calculated using Poisson regression. Prevalent characteristics included per capita income ranging from one-half to one minimum wage (35.1%), complete middle school (39.8%), and food security (59.4%). Users in the North of Brazil showed the worst data: incomplete primary school (39.8%), per capita income up to one-half the minimum wage (50.8%), and food insecurity (55.5%). Prevalence ratios for food insecurity were higher among users with per capita income up to one-fourth the minimum wage (p < 0.05). Income was the only variable that remained associated with higher prevalence of food insecurity in the adjusted PR. Knowing the characteristics of soup-kitchen users with food insecurity can help orient the program's work, location, and operations.