Baseline Predictors of Pregnancy Over Two Years Among African-American Teens

Background: Preventing teen pregnancy is a major public health initiative and one of CDC's six priority areas. Greater understanding of factors which predict pregnancy is needed.

Objective: To identify predictors of pregnancy among African-American adolescents.

Methods: Sociodemographic, psychosocial and sexual behavior data were collected via ACASI at baseline and every 6 months for 2 years from 615African-American females (15-19 years) enrolled in an HIV prevention trial. Adolescents who were married, currently pregnant or attempting to become pregnant were excluded from the trial. Pregnancy was confirmed via urine assays at each assessment visit. Multivariate analyses compared study variables for those who did versus did not become pregnant during the 2-year follow-up.

Results: One-hundred twenty (23%) participants became pregnant during follow-up. In analyses adjusting for age, treatment assignment and predictors statistically significant in bivariate analyses, those who became pregnant were less likely to have ever had anal sex (AOR: .48, p=.012) and use hormonal contraceptives at last sex (AOR: .57, p=.020). They were also more likely to have had sex in the past 6 months with a male partner recently released from prison, jail or detention (AOR: 1.8, p=.031).

Conclusions: A substantial proportion of African-American teens who indicated they were not attempting to become pregnant at baseline was not taking hormonal contraceptives and subsequently became pregnant within two years. Data are still needed to better understand how risky social networks either directly or indirectly influence the contraceptive decision-making processes of adolescent girls.