PIP: The purpose of this prospective study of 120 black and Puerto Rican adolescent mothers recruited from an urban adolescent health center was 1) to identify multiple predictors of repeat pregnancies and their outcomes, and 2) to assess the relative contribution of these predictors to repeat pregnancies. Variables include demographic factors, frequency of sexual activity, contraceptive method used, occurrence of pregnancies before the first delivery, educational attainment, and levels of depressive symptoms by 3 weeks postpartum. Pregnancy outcomes included mothers who terminated their pregnancy, who delivered a live birth, and who had a miscarriage; data were available 12 months postpartum. The mean age was 17.13 years. 52% were black and 44% were Puerto Rican and of low socioeconomic status. Reliability of self-reports was 95%. 60.3% had no repeat pregnancy and 39.6% became pregnant during the 1-year followup, of which 15.3% had a therapeutic abortion, 6.3% miscarried, and 18.1% carried to term. There were no significant differences in the characteristics of the repeat and nonrepeat pregnancy groups. The procedure involved interviews at 2-4 weeks (time 1), 6-7 months (time 2), and 12-13 months (time 3). Measures for delayed grade placement, reading achievement, career aspirations and socioeconomic status of the family, and school attendance are described. Multivariate analysis, univariate analyses, and Duncan multiple range tests were conducted as well as logistic regression model construction. The significant predictors of pregnancy outcome (repeat pregnancy) were having 2 or more previous pregnancies. The abortion group was twice as likely to have had a pregnancy before the birth of the first child (41%) than the full-term (20%) and the no-repeat groups (15%). There were no differences in frequency of sexual activity or contraceptive method used. In the no-repeat and miscarriage groups, grade placement was less delayed. In the no-repeat group, reading achievement scores were higher than in the full-term group. There were no differences in career aspirations. 65% of the no-repeat group were attending school at time 2 while only 35% of the therapeutic and 24% of the full-term groups were attending school. Depressive scores were significantly higher for the abortion group. Delayed grade placement was the only predictor independently associated with the increased log odds of repeat pregnancy.