BACKGROUND: Pregnancy-induced hypertension (PIH) and gestational diabetes mellitus (GDM) are among the leading causes of disability and death for women and their babies. Identifying risk factors for these pregnancy-related complications is essential to their prevention. Studies identifying preventive models for PIH and GDM are few. OBJECTIVES: This study was designed to evaluate lifestyle and personal predictors of PIH and GDM in a cohort of nearly 20,000 pregnant women. METHODS: The exposure data for the study were derived from a combination of a telephone interview and a questionnaire completed approximately 2 months after conception during the period from 1984 to 1987. The initial questionnaires asked for information on three periods: 3 months before conception, at conception, and 2 months after conception. Subjects included 19,312 women, aged 18-<45 years, who did not have excessive intakes of alcohol or food, were neither underweight (BMI >18.5) nor extremely overweight (BMI <40), and did not use illegal drugs during the first trimester of pregnancy. Outcome data on the mother and baby were collected approximately one year after the expected data of delivery. Logistic regression models were used to estimate the odds ratios (OR), and 95% confidence intervals (CI), as well as receiver operating characteristic (ROC) curves predicting PIH and GDM. Akaike Information Criteria (AIC) were used to select the best predictors of these two outcomes. Factors found not to affect PIH or GDM (based on a two-unit decrease in the AIC) were excluded from the final models. RESULTS: Based on the outcome data collected, there were 204 PIH cases, 358 GDM cases, and 538 who had PIH and/or GDM. After selecting the outcome predictors using AIC values, we identified three predictive models—one each for PIH, GDM, and either PIH or GDM. Factors found to predict PIH included age, previous hypertension or type 1 or 2 diabetes, pre-pregnancy BMI, parity, exercise, red meat consumption, margarine consumption, cigarette smoking, and weight change at 2 months. The final AIC value for PIH was 2084.12 and the AUC value was 0.76. GDM was predicted by age, previous GDM (in an earlier pregnancy), pre-pregnant BMI, height, exercise, race, dairy consumption, and cigarette smoking, with an AIC value of 3288.74 and an AUC value of 0.70. The combined model (predicting either PIH or GDM) was best predicted by age, history of GDM in a previous pregnancy, pre-pregnant BMI, previous history of hypertension, height, exercise, dairy consumption, red meat consumption, parity numbers, cigarette smoking, and weight change at 2 months with an AIC value of 3288.74 and an AUC value of 0.71. CONCLUSIONS: In these analyses, separate models predicting PIH and GDM were better than a combined model predicting PIH or GDM. These final models indicate that we can reasonably identify women who are at increased risk for adverse maternal outcomes associated with hypertensive disorders or diabetes during pregnancy.