Two hundred eighty-four women who had preterm deliveries and matched controls who had full-term deliveries were analyzed using a conditional logistic regression model to assess the effects of employment and socio-medical factors on preterm birth. In the analysis, employment outside the home, in general, turned out not to be a significant risk factor for preterm birth. Of the social factors, unmarried status (odds ratio 2.0) and current smoking (odds ratio 2.4) were associated with preterm birth, as were medical factors such as hypertension (odds ratio 7.3), intrauterine growth retardation (odds ratio 3.9), and fetal malformations (odds ratio 5.2).