| The purpose of this study was to: (1) describe sleep patterns in menstruating women sleeping at home, (2) determine the effect of a nighttime magnetic field and other exposures on sleep, and (3) examine the association between sleep and urinary sex hormones during the luteal menstrual phase. Data are from a randomized crossover design trial, comparing an intervention magnetic field (0.5 to 1.0 microTesla above ambient levels) to ambient levels, during two 5-night measurement periods. Subjects collected an overnight urine sample on the fifth night. Subjects were not taking oral contraceptives, 20--40 years old, not pregnant/breast-feeding during the previous year, and had regular menstrual cycles. Sleep outcomes were measured via actigraphy. Of 640 nights, 91 (14%) were missing some data. A wide range of sleep patterns was observed. The intra-class correlation was low (rho < 0.21) for total sleep time, sleep onset, and time in bed, but higher (rho = 0.40--0.51) for sleep efficiency, total wake time, wake after sleep onset, awakenings, and awakenings ≥3 minutes. Sleep outcomes were not significantly different between intervention and ambient measurement periods. However, higher ambient magnetic field exposure was significantly associated with less total sleep time, lower sleep efficiency, and more awakenings ≥3 minutes. Unusual bed or rise times, medication use, employment, day of week, daylight hours, menstrual cycle length, and body mass index were associated with sleep patterns, while age, alcohol consumption, exercise over the previous month, and perceived stress were not. We found few associations between sleep and urinary sex hormone concentrations. However, increased awakenings were modestly associated with higher urinary luteinizing hormone concentrations. Anovulatory cycles were significantly associated with lower sleep efficiency and more awakenings ≥3 minutes. Our results suggest that actigraphy is a feasible method of measuring sleep patterns and assessing associations with various exposures in menstruating women sleeping at home. The intervention exposure had no effect on sleep patterns, however other exposures were associated with worse sleep. Our results also support the hypothesis that sleep patterns and hormonal regulation processes are interrelated. More research is necessary to elucidate the complex relationships between these exposures and sleep. |
