A population-based interview study of 207 case-control pairs investigated reproductive, dietary, and other factors thought likely to increase thyroid cancer risk among women of reproductive age in Shanghai. Of particular interest were factors that might help explain the striking female over male excess in this age group. Risk was strongly associated with prior goiter or benign nodules (odds ratio [OR] = 7.0, 95 percent confidence interval [CI] = 2.5-27.5) and miscarriage as outcome of first pregnancy (OR = 9.9, CI = 2.0-48.4). Weaker associations were seen for women who were ever-pregnant (OR = 2.1, CI = 1.1-4.2), ever had an induced abortion (OR = 1.6, CI = 0.9-2.9), and ever used oral contraceptives (OR = 1.7, CI = 1.0-3.1). Compared with controls, cases gained significantly more weight from menarche to highest nonpregnant weight (P trend < 0.01). Overall, cases ate more fish and shellfish, but there was no trend with level of consumption. More cases had a parent, sibling, or child with thyroid disease (OR = 5.2, CI = 2.5-12.1). Our findings relating to goiter and benign nodules and various reproductive factors support earlier studies. Consumption of seafood was difficult to evaluate; more detailed dietary data are needed to assess iodine intake.