All of 417 pregnancies in 1987-1990 in a rural community were retrospectively evaluated with emphasis on age, parity, previous spontaneous or induced abortion, occupation during pregnancy and sick leave from work. No difference was found in prevalence of sick leave between the employed women and the unemployed. Thirty-five percent of the women experienced sick leave at some time during pregnancy. The greatest single diagnostic group was pelvic insufficiency, which accounted for 33.5% of all days of sick leave up to the 32nd week of gestation. The incidence of pelvic insufficiency was high: 76 in 1000 deliveries. This may in part be due to a change in diagnostic criteria. Logistic regression analysis revealed that being a multipara was associated with a significantly higher risk of sick leave in the present pregnancy. Furthermore, being employed as a nurse, a shop assistant, a seamstress or a factory worker was associated with a significantly higher risk of sick leave during pregnancy in comparison with employment as a secretary. When investigating relationships between occupational conditions and the outcome of pregnancy, sick leave during pregnancy because of the same conditions may interfere significantly.