Abstract
Objective: Sleep-related phenotypes, including insomnia symptoms, sleep apnea, and habitual sleep duration, are increasingly recognized as potentially modifiable factors associated with psychiatric disorders. However, their potential causal roles remain uncertain because observational studies are susceptible to confounding and reverse causation. Study Design: Two-sample Mendelian randomization (MR) was performed using self-reported insomnia symptoms, sleep apnea and habitual sleep duration, as exposures against 10 psychiatric outcomes: ADHD, PTSD, MDD, PPD, BIP, ASD, OCS, SCZ, AN, and PD. GWAS summary statistics were obtained from the UK Biobank, Psychiatric Genomics Consortium, and affiliated consortia. Inverse-variance weighted (IVW) was the primary method, with MR-Egger, weighted median, and MR-PRESSO applied as sensitivity analyses. Multivariable MR (MVMR) was conducted to estimate independent associations of the sleep-related phenotypes modeled simultaneously. Results: Genetically predicted insomnia symptoms showed bonferroni-significant associations with increased risk of ADHD (OR 2.50, 95% CI 1.65–3.78, p < 0.001), PTSD (OR 1.32, 95% CI 1.12–1.56, p < 0.001). Nominal associations were observed for MDD (OR 1.57, 95% CI 1.16–2.12, p = 0.003), PPD (OR 2.35, 95% CI 1.07–5.13, p = 0.033), and BIP (OR 1.67, 95% CI 1.03–2.68, p = 0.036). Genetically predicted sleep apnea showed nominal associations with OCS (OR 1.20, 95% CI 1.07–1.35, p = 0.002) and AN (OR 1.57, 95% CI 1.17–2.11, p = 0.003). Genetically predicted longer sleep duration showed nominal associations with increased risk of BIP (OR 1.26) and SCZ (OR 1.67), but did not survive multiple-testing correction. MVMR analyses showed associations of longer sleep duration with ADHD (protective; OR 0.16, 95% CI 0.05–0.54, p = 0.004) and AN (OR 37.66, 95% CI 4.12–345.21, p = 0.001); however, these estimates were imprecise and potentially affected by conditional weak-instrument bias. Sensitivity analyses did not detect significant evidence of directional horizontal pleiotropy based on the MR-Egger intercept test in univariable analyses, although MVMR estimates remained imprecise. Conclusions: The sleep-related phenotypes showed heterogeneous patterns of association with psychiatric disorders. Genetically predicted insomnia symptoms were associated with the broadest range of psychiatric outcomes, whereas sleep apnea showed more outcome-specific associations. These findings highlight the importance of considering multiple dimensions of sleep when evaluating psychiatric disorders.