Sleep duration and post-traumatic stress disorder symptoms: a twin study

Long and short sleep duration are associated with greater risk of developing post-traumatic stress disorder (PTSD); however, it is unknown how genetic and environmental influences affect this relationship. Thus, we investigated the association between sleep duration and PTSD symptoms using twin models.

Data were obtained from 1865 monozygotic and 758 dizygotic twin pairs enrolled in the community-based Washington State Twin Registry. PTSD symptoms were assessed using the Impact of Events Scale (IES). A classical twin model decomposed the variances of sleep duration and IES score into additive genetic, shared environmental, and unique environmental components. We used correlated factor models to examine the moderation of variance components of sleep duration and IES.

Shorter and longer sleep duration were associated with higher IES scores with a quadratic association (p < 0.001). The heritability of sleep duration was 36%, and IES 31%. Variance in sleep duration attributable to shared (b1C1 = 2.91, 95% CI = 1.40 to 4.43; p < 0.001) and unique (b1E1 = 0.18, 95% CI = 0.10 to 0.27; p < 0.001) environment was moderated by IES score. Similarly, but to a lesser extent, variance in IES attributable to additive genetics (b1A2 = -0.23, 95% CI = -0.45 to 0.00; p = 0.048) was moderated by sleep duration.

Greater PTSD symptom severity was associated with short and long sleep duration. Increasing PTSD symptoms increased variability in sleep duration primarily via shared environmental factors, whereas decreasing sleep duration increased variability in PTSD symptoms primarily via additive genetic factors. This suggests childhood experiences affect variability of sleep duration and genetic factors affect the variability of PTSD symptoms in trauma-exposed individuals.