Abstract
OBJECTIVE: To determine whether chronodisruption is associated with achieving pregnancy. DESIGN: Pilot prospective cohort study. SETTING: Academic Medical Center. SUBJECTS: One hundred eighty-three women desiring pregnancy were recruited from the local community of an academic medical center located in the Midwest and provided sleep information between February 1, 2015, and November 30, 2017. EXPOSURE: Sleep and activity data were obtained via actigraphy watches worn continuously for 2 weeks to assess measures of chronodisruption, including sleep period onset, offset, midtime, and duration; as well as variability in each of these measures. MAIN OUTCOME MEASURES: Time to becoming pregnant over 1-year of follow-up. RESULTS: Of the 183 eligible women, 82 became pregnant over a median of 2.8 months of follow-up. Greater interdaily variability in time of sleep onset and variability in sleep duration were associated with a longer time to achieving pregnancy after adjusting for age, body mass index, race, education, income, and smoking status (adjusted hazard ratio [aHR], 0.60; 95% confidence interval [CI], 0.36-0.999 comparing participants with a standard deviation of >1.8 hours to <1.8 hours in daily time of sleep onset; and aHR, 0.58; 95% CI, 0.36-0.98 comparing participants with a standard deviation of >2.3 hours to <2.3 hours in daily sleep duration). In adjusted analyses, no statistically significant associations were observed for average time of sleep onset and offset, midsleep time, and sleep duration, or for variability in time of sleep offset and midtime. CONCLUSION: Higher day-to-day variability in time of sleep onset and sleep duration-two measures of chronodisruption-were associated with a longer time to achieving pregnancy over 1 year of follow-up in women desiring pregnancy. If replicated in additional studies, these findings could point to lifestyle interventions to help women achieve a desired pregnancy.