Objective To analyze the survival of patients with Chagas disease, beneficiaries of social security and social assistance, in Brazil, from 1942 to 2016. Methods This is a retrospective cohort study with data from the Brazilian Ministry of Social Security. The event of interest was death, and the survival functions were estimated by the Kaplan-Meier and Cox regression methods. Results In the period “onset of the disease until death”, women (HR=0.54; 95%CI 0.43–0.53) and receiving social security benefits (HR=0.13; 95%CI 0.11–0.23) were associated with longer survival. Lower survival was associated with the cardiac form of the disease (HR=2.64; 95%CI 2.23–3.12), living in a rural area (HR=1.23; 95%CI 1.14–1.21), and manifestation of the disease between the years 2000 and 2016 (HR=5.32; 95%CI 4.74–5.93). Likewise, in the period “work disability until death”, women (HR=0.51; 95%CI 0.41–0.52) and receiving social security benefits (HR=0.24; 95%CI 0,14–0.45) were associated with longer survival, as well as the cardiac form of the disease (HR=1.95; 95%CI 1.83–2.13), living in a rural area (HR=1.31; 95%CI 1.21–1.54), and manifestation of the disease between 2000 and 2016 (HR=1.53; 95%CI 1.33–1.71) were associated with lower survival. Conclusion The main predictors of mortality and survival of patients with Chagas disease who receive social security and assistance benefits in Brazil were presented. These findings can guide the definition of priorities for follow-up actions by Primary Health Care, currently recommended for the longitudinal management of the disease.