There have been claims of an increased risk for gestational trophoblastic disease (i.e., hydatidiform mole and choriocarcinoma) in Vietnam since the period of Agent Orange sprayings. In 1990, we conducted a case-control study in Ho Chi Minh City to investigate risk factors for gestational trophoblastic disease in Vietnam. Eighty-seven married women, all of whom had a recent pathologic diagnosis of gestational trophoblastic disease, identified in the Obstetrical and Gynecological Hospital, were included in the study. Eighty-seven married women who were admitted mainly in the surgery departments of the same hospital were the controls, and they were matched to cases for age and area of residence. Odds ratios (ORs), adjusted for matching variables and other potential confounders, were estimated with unconditional logistic regression. A statistically significant trend in risk was observed with previous live births (p = .01). Cases were found to eat less meat per wk (OR = 0.4, 95% confidence interval [95% CI] = 0.2-0.9 for > or = five meat dishes) and to own fewer consumer goods than controls. An increase in risk was associated with the breeding of pigs (OR = 5.7, 95% CI = 1.2-27.6 for raising three or more pigs). A cumulative Agent Orange exposure index was constructed, using the patient's complete residence history. No significant difference was found between cases and controls for this index (OR = 0.7, 95% CI = 0.2-1.8 for high-exposure category), nor was such a difference noted for the agricultural use of pesticides.