Background. Flight attendants (FAs) may experience circadian rhythm disruption due to travel through multiple time zones, which causes desynchronization of biological rhythms and physiologic and behavioral disturbances including sleep-wake disturbance. Overnight melatonin production, sleep disturbance, and flight history information were analyzed as part of a menstrual cycle function biomonitoring study. The first objective was to determine whether FAs are at higher risk of circadian disruption compared to teachers, as measured by melatonin production. The second objective was to identify the best metrics of circadian rhythm disruption for studies in which biomonitoring is infeasible. Methods. Each day for one menstrual cycle, 45 FAs and 26 teachers kept a daily diary, collected and measured their overnight urine, and wore an activity monitor, from which activity and sleep disturbance measures were calculated. The principal urinary metabolite of melatonin, 6-sulfatoxymelatonin, was measured by radioimmunoassay. The relation between melatonin production and FN/teacher status was analyzed with linear and multiple logistic regression. Melatonin production was expressed as 1) overnight melatonin production rate, 2) within-woman variance of the rate, and 3) the proportion of low melatonin study days. Relations between measures of sleep disturbance, melatonin, and measures derived from flight histories (including time zones crossed) were examined with exploratory factor analyses. Results. Flight attendants experience increased circadian rhythm disruption as measured by a higher adjusted melatonin rate variance than teachers (2.69x(l0)5 vs 1.17x(l0)5 (ng/h)2 respectively; p=0.02) and the likelihood of being in the highest quartile of melatonin variance (Odds Ratio=2.3; 95%. Confidence Interval: 0.4-11.9). Circadian rhythm disruption was characterized by three groups of measures (factors). The first factor contained most sleep disturbance and activity measures and the number of time zones crossed. The second factor included time zones crossed plus the overnight rate and variance of melatonin production. The third factor included sleep efficiency (one measure of sleep disturbance) and the proportion of low melatonin days. Time zones crossed was included in the first two factors and thus was related to both melatonin production and most measures of sleep disturbance. An exception was sleep efficiency, which was distinct from all other sleep measurements evaluated. Conclusions. Our observations suggest that FAs experience increased circadian rhythm disruption, as measured by more variable melatonin rates. For studies in which melatonin measurement is infeasible, time zones crossed is a useful measure of both sleep disturbance and melatonin desynchronization.