Importance: Short sleep duration over a prolonged period in childhood could have a detrimental impact on long-term mental health, including the development of psychosis. Further, potential underlying mechanisms of these associations remain unknown.
Objective: To examine the association between persistent shorter nighttime sleep duration throughout childhood with psychotic experiences (PEs) and/or psychotic disorder (PD) at age 24 years and whether inflammatory markers (C-reactive protein [CRP] and interleukin 6 [IL-6]) potentially mediate any association.
Design, setting, and participants: This cohort study used data from the Avon Longitudinal Study of Parents and Children. Data analysis was conducted from January 30 to August 1, 2023.
Exposures: Nighttime sleep duration was collected at 6, 18, and 30 months and at 3.5, 4 to 5, 5 to 6, and 6 to 7 years.
Main outcomes and measures: PEs and PD were assessed at age 24 years from the Psychosislike Symptoms Interview. CRP level at ages 9 and 15 years and IL-6 level at 9 years were used as mediators. Latent class growth analyses (LCGAs) were applied to detect trajectories of nighttime sleep duration, and logistic regressions were applied for the longitudinal associations between trajectories of nighttime sleep duration and psychotic outcomes at 24 years. Path analyses were applied to test CRP and IL-6 as potential mediators.
Results: Data were available on 12 394 children (6254 female [50.5%]) for the LCGA and on 3962 young adults (2429 female [61.3%]) for the logistic regression and path analyses. The LCGA identified a group of individuals with persistent shorter nighttime sleep duration across childhood. These individuals were more likely to develop PD (odds ratio [OR], 2.50; 95% CI, 1.51-4.15; P < .001) and PEs (OR, 3.64; 95% CI, 2.23-5.95; P < .001) at age 24 years. Increased levels of IL-6 at 9 years, but not CRP at 9 or 15 years, partially mediated the associations between persistent shorter sleep duration and PD (bias-corrected estimate = 0.003; 95% CI, 0.002-0.005; P = .007) and PEs (bias-corrected estimate = 0.002; 95% CI, 0-0.003; P = .03) in young adulthood.
Conclusions and relevance: Findings of this cohort study highlight the necessity of addressing short sleep duration in children, as persistence of this sleep problem was associated with subsequent psychosis. This study also provides preliminary evidence for future targeted interventions in children addressing both sleep and inflammatory responses.