This study was performed to examine the effect of stress on pregnancy outcome in women who underwent assisted reproductive technology (ART) procedures. In a controlled clinical study of healthy volunteers in an academic research environment, stress was measured subjectively by administering patient questionnaires and biochemically by examining urinary excretion of cortisol and 6-sulfatoxy-melatonin (6-SM), the primary metabolite of melatonin and a marker of peripheral stress response. A total of 42 women who underwent ART procedures during an 18-month period agreed to participate in the study and were enrolled consecutively. The women collected 24-hour urine specimens on the day after human chorionic gonadotropin administration and concurrently completed the State-Trait Anxiety Inventory questionnaire. We measured the cortisol and 6-SM levels in the urine collection for each of the 42 women and for 10 oocyte donors who served as controls. Pregnancy tests were performed 14 days after embryo transfer. Analysis of covariance was used to compare the relationship of stress ratings to urinary cortisol and 6-SM levels among the women who became pregnant, the women who did not, and the women who served as controls. Other variables were explored by performing chi2 analysis and Fisher's exact test. Neither self-ratings of acute anxiety, nor total daily 6-SM value, nor cortisol levels were associated with pregnancy outcome in the ART procedures. Biochemical markers of stress failed to support a deleterious effect of stress on pregnancy outcome in women who underwent ART procedures. Subjective measurement of stress levels did not differ between women who became pregnant and those who did not.