Background: Abortion is a public health problem in Latin America and is more common among women living with HIV.
Objective: to verify the incidence and factors associated with induced abortion in a cohort of women living with HIV assisted in a reference service for care for individuals with HIV/AIDS in Rio de Janeiro/Brazil.
Methods: Prospective cohort during the period 1996-2016. We estimated the incidence of induced abortions during follow-up in the cohort by calculating person-time incidence rates [per 100 persons-years (PY)] and investigated the factors associated with the outcome "induced abortion" using a generalized linear mixed model.
Results: 753 women and 210 pregnancies were included in the present analysis. We estimated an induced abortion incidence rate of 0.68/100 persons-years (95% confidence interval [CI]: 0.47; 0.94) in the study period, with a significant reduction after 2006. The main factors associated with an induced abortion were currently living with a partner (adjusted OR [AdjOR] 0.32 95% CI: 0.10-0.98), number of children (2 children AdjOR 0.12, 95% CI: 0.02-0.95) and the type of antiretroviral treatment used (regimen without Efavirenz: AdjOR: 0.11, 95% CI 0.02-0.70).
Conclusions: We showed a significant reduction in the incidence of induced abortions in a cohort of women living with HIV in Rio de Janeiro, Brazil, probably due to a decrease in the incidence of pregnancies observed in the same period. The factors associated with a lower occurrence of induced abortion suggest a good integration between the clinical and reproductive assistance offered to those women.
Keywords: Abortion, induced; Brazil; HIV infections; cohort studies; incidence.