The aim of this study was to identify the reasons for hospitalization associated with abortion and pregnancy termination among women attending the Maternity Unit of IMIP, Recife, Brazil. In a case-control study, 230 women who were hospitalized consecutively between August 1994 and June 1995 due to causes related to pregnancy termination or abortion were chosen as cases. Four controls per case were selected randomly out of a total of 920 women who delivered at the same hospital. For each potential cause considered, the odds ratio (OR) and respective 95 per cent confidence interval (CI) were calculated. Multiple logistic regression analysis was then used to control for confounders. Among all potential causes studied, those that showed significant association with pregnancy termination or abortion were: being single, absence of emotional support from partner, being a working woman, literacy up to the fourth year of school, failure of contraceptive method employed, and experience of previous pregnancies, children, and abortions.