Prior observational research has suggested an association between increased physical activity during pregnancy and improved maternal outcomes including a reduced risk of gestational diabetes mellitus (GDM). Despite this, randomized controlled trials of prenatal exercise are sparse and only one had been conducted in a high-risk population. Therefore, we examined the association between physical activity during pregnancy and risk of adverse pregnancy outcomes in the Behaviors Affecting Baby and You (B.A.B.Y.) Study, a randomized controlled trial of a 12-week exercise intervention among a diverse group of inactive pregnant women at high-risk for GDM. The first project examined the effect of participation in the B.A.B.Y. exercise intervention on reducing the risk of GDM. Among the 251 pregnant women at high-risk for GDM included in the analysis, those who participated in the exercise intervention had a lower odds of GDM as compared to those who participated in the health and wellness intervention, although results were not statistically significant. The second project examined the effect of participation in the B.A.B.Y. exercise intervention on gestational weight gain (GWG) were assessed. Of the 241 participants included in the analysis, there was no statistically significant impact of the exercise intervention on total GWG, rate of GWG, and compliance with the Institute of Medicine recommendations for GWG versus the health and wellness intervention. The third paper investigated the correlates of three health risk behaviors in early pregnancy: smoking, low levels of exercise, and high levels of sedentary behavior. Data on health risk behaviors were collected through self-report prior to randomization (n=383). Lower education, lower income, not being married or living with a partner, not living with any other adults, drinking alcohol pre-pregnancy and being sedentary were associated with smoking. Younger age and not having children in the household were associated with low levels of activity. Lower income, being unmarried or not living with a partner, not having children in the household and smoking were associated with higher levels of sedentary behavior. To summarize, this dissertation adds to current research by assessing the effectiveness of an individually-tailored exercise intervention in reducing risk of pregnancy complications in a diverse population.