Pregnancy and other risk behaviors among adolescent girls in Ohio.

PURPOSE: The purpose of this study was to determine whether teenage girls who had been pregnant were more likely to engage in other risk or problem behaviors than girls who had had sexual intercourse without becoming pregnant.

METHODS: The 1993 Ohio Youth Risk Behavior Survey was administered to a random sample of 2461 high school students. A subset of 592 girls (mean age 16.1 +/- 1.1 years, 69.2% Caucasian, 24.1% African-American, 2.4% Hispanic, 4.3% other) reported sexual activity (SA) on the anonymous survey. Demographic factors plus risk and problem behaviors were compared between 98 girls who had been pregnant and 494 girls who had never been pregnant using Chi-square analysis. Risk and problem behaviors were classified into four distinct groups: recent risk behaviors, age of onset of behavior, lifetime behavior, and general behavior. Variables which were univariately significant at p < 0.15 were considered in one of four stepwise multiple logistic regression models based on behavior type, with demographic factors considered in all four models. The final logistic regression model was developed using variables which were significant at p < 0.01. Variables which were significantly associated with pregnancy were summarized as odds ratios (OR) and 99% confidence intervals (CI); these ORs were adjusted for the effects of the other variables in the model.

RESULTS: Pregnancy was more common in girls of color than in Caucasian girls (OR 99% and CI, 2.09, 1.06-4.11) and in older girls than in younger girls (1.52, 1.12-2.08). Among SA girls, weapon carrying in the past 30 days (4.06, 1.75-9.42) was significantly associated with pregnancy, whereas alcohol use in the past 30 days (0.37, 0.18-0.76) was less likely to be associated with pregnancy. The risk of pregnancy increased 1.75 times (1.26-2.43) for each additional sexual partner. Girls who had tried cocaine were 4.88 times (1.40-16.95) more likely to have been pregnant, and the risk of having been pregnant increased 1.43 times (1.14-1.80) for each additional year of SA. Past history of sexually transmitted disease (3.50, 1.28-9.55) was associated with increased pregnancy risk.

CONCLUSIONS: Girls who had been pregnant also had engaged in other risk behaviors, including recent weapon carrying and cocaine use. A history of previous sexually transmitted diseases plus increasing numbers of partners add to the risk of pregnancy. Counseling and educational efforts should continue to identify teens at risk both to prevent pregnancy and to decrease associated risks.