Postpartum Psychosis: could genetic vulnerability to insomnia or short sleep duration be protective?

Postpartum Psychosis (PP) is a severe and understudied perinatal mental illness which disproportionately affects women with bipolar disorder (BD). A relationship between sleep disturbance and PP is often assumed, but is poorly understood. From a cohort of 2099 individuals with BD, 343 parous women were identified and screened for perinatal psychiatric complications. We compared 117 women who developed PP with 226 who did not. Polygenic Risk Scores (PRS) for BD, schizophrenia, insomnia, short sleep, long sleep, sleep efficiency and sleep duration were computed using PRS-CS. Logistic regression was used to model the effect of each PRS on PP. Higher PRS for insomnia and short sleep were associated with reduced risk of PP. Individuals in the lowest decile for insomnia PRS (RR 1.96, 95% CI 1.25-3.07, p = 3.50 × 10⁻³) and short sleep PRS (RR 2.23, 95% CI 1.40-3.54, p = 7.94 × 10⁻⁴) had approximately double the risk of PP than individuals in the highest decile. The other PRS were not associated with PP. Mendelian Randomisation analyses did not support a causal relationship between sleep traits and PP. However, we demonstrate that the integration of PRS with bipolar subtype can improve prediction accuracy. Individuals with genetic vulnerability to insomnia or short sleep may develop a heightened tolerance to sleep disruption earlier in life, mitigating the impact of childbirth on mood. These findings suggest that genetic susceptibility to sleep disturbance may be important in the aetiology of PP, offering a new potential avenue for risk stratification and targeted prevention.