Little is known about modifiable lifestyle factors beyond quitting smoking that could prevent preterm delivery (PTD, <37 weeks gestation). We examined the individual and joint associations of pre-pregnancy BMI, second trimester exercise and sleep on PTD. We conducted a nested, population-based case-control study interviewing postpartum 344 cases delivering at <37 weeks, as identified by clinical estimate of gestational age from prenatal screening records, and 698 term controls, excluding term low birthweight. Eligible women participated in California's statewide Prenatal Screening Program, worked during pregnancy, and delivered a singleton birth in Southern California in 2002-2003. Modeled separately, moderate (odds ratio [OR] = 0.90; 95% confidence interval [CI] = 0.84-0.96--per hour/week) and vigorous (OR = 0.67; 95% CI = 0.46-0.98 for yes vs. no) exercise during the second trimester were associated with a reduced odds of PTD, and sleep duration was not (OR = 1.09, 95% CI = 0.80-1.48 for <7 h; OR = 0.88, 95% CI = 0.57-1.48 for >8 h vs. 7-8 h). When sleep and exercise variables were modeled together along with pre-pregnancy BMI, only moderate exercise (OR = 0.91; 95% CI 0.85-0.98) continued to be associated with reduced odds of PTD. The benefits of moderate exercise appeared strongest for women with BMI greater than 24 kg/m(2) (OR = 0.85; 95% CI = 0.79-0.93) and weakened with decreasing BMI. No other interactions were found. Moderate exercise is associated with reduced PTD, particularly for women with BMI above the normal range. The results are of public health relevance given that these risk factors are potentially modifiable both pre-conceptionally and during pregnancy and rates of PTD are still high in the United States.