Trauma history and persistent poor objective and subjective sleep quality among midlife women.

Objectives: Whereas some work links trauma exposure to poor subjective sleep quality, studies largely rely upon limited trauma measures and self-reported sleep at one time point. It is unknown whether trauma is related to persistent poor sleep, whether associations differ based on childhood versus adulthood trauma, and whether trauma exposure is related to poorer objectively assessed sleep. We tested whether childhood or adult trauma associated with persistent poor objectively and subjectively measured sleep at two time points in midlife women.

Methods: One hundred sixty-seven women aged 40-60 at baseline were assessed twice 5 years apart. At baseline, women reported childhood trauma (Child Trauma Questionnaire), adult trauma (Brief Trauma Questionnaire), demographics, depressive symptoms, apnea symptoms, and medical history, and provided physical measures. At both visits, women completed 3 days of actigraphy (total sleep time [TST], wake after sleep onset [WASO]) and reported sleep quality (Pittsburgh Sleep Quality Index). Relations of childhood and adult trauma exposure, respectively, with persistent poor sleep at both baseline and follow-up visits (TST [<6 hours], WASO [>30 minutes], Pittsburgh Sleep Quality Index [>5]) were assessed in logistic regression models, adjusted for age, race/ethnicity, education, body mass index, sleep medications, nightshift work, apnea, depressive symptoms, vasomotor symptoms, and alcohol use.

Results: Childhood trauma was related to persistent high WASO (odds ratio [95% confidence interval] = 2.16 [1.04-4.50], P = 0.039, multivariable). Adult trauma was related to persistent poor sleep quality (odds ratio [95% confidence interval] = 2.29 [1.07-4.93], P = 0.034, multivariable). Trauma was unrelated to persistent short TST.

Conclusions: Childhood and adult trauma, respectively, were related to persistent poor objective sleep continuity and subjective sleep quality in midlife women, independent of risk factors.