OBJECTIVE: To estimate the effect of long working hours and shift work on time to pregnancy. METHODS: Cross sectional samples with retrospective data collection from two 700 bed hospitals at secondary to tertiary care level in Hatyai district, Songkhla Province, Thailand. The study was conducted from March 1995 to November 1995 among 1496 pregnant women attending the antenatal clinics. Subfecundity was defined as time to pregnancy longer than 7.8, 9.5, or 12 months (time to pregnancy was calculated from the date at which the couples started having sexual relations without any contraception until last menstrual date). RESULTS: The descriptive analyses were restricted to 1201 planned pregnancies and the analytical part to 907 working women. Separate analyses on primigravid women were also done. Logistic regressions adjusted for age, education, body mass index, menstrual regularity, obstetric and medical history, coital frequency, and potential exposure to reproductive toxic agents, showed an odds ratio (OR) associated with female exposure to long working hours of 2.3 (95% confidence interval (95% CI) 1.0 to 5.1) in primigravid and 1.6 (1.0 to 2.7) in all pregnant women. Male exposure to long working hours and shiftwork showed no association with subfecundity. The OR of subfecundity was highest when both partners worked > 70 hours a week irrespective of the cut off point used (OR 4.1 (95% CI 1.3 to 13.4) in primigravid women; OR 2.0 (95% CI 1.1 to 3.8) in all pregnant women). CONCLUSIONS: Long working hours is a risk factor for subfecundity especially for women. Shiftwork was not associated with subfecundity in this study.