Abstract
BACKGROUND: Suboptimal sleep is prevalent in both children and adults and is linked to adverse health outcomes. Understanding life course sleep duration trajectories is important for identifying at-risk individuals and informing intervention strategies. This study aimed to describe sleep duration trajectories from childhood to mid-adulthood and to identify childhood predictors. METHOD: Data were from a national cohort of children (n = 759, mean baseline age 12.0 [2.0] years, 38.6% male). Sleep duration in childhood was calculated from self-reported bedtime and wake-up time and reassessed approximately 25 and 30 years later using validated questionnaires. Suboptimal sleep (yes/no) was defined using age-specific recommendations. Trajectories were constructed based on changes in suboptimal sleep over time. Log-multinomial regression models identified childhood sociodemographic, school-related, and health predictors. RESULTS: Five sleep duration trajectories were identified: persistently optimal (46.1%), improving (15.4%), inconsistent (10.4%), worsening (23.5%), and persistently suboptimal (4.6%). Relative to the persistently optimal trajectory, female sex predicted the improving trajectory. Speaking a language other than English at home and being overweight/obese predicted the inconsistent trajectory. Male sex, lower parental occupation, more siblings, skipping breakfast, and poorer self-rated health predicted the worsening trajectory. Speaking a language other than English at home, lower school engagement, worse learner self-concept, and smoking predicted the persistently suboptimal trajectory. DISCUSSION: Nearly half of participants maintained optimal sleep from childhood to mid-adulthood, while 40% followed unfavourable trajectories and only 15% improved. Early-life sociodemographic, school-related, and health factors may shape long-term sleep duration trajectories and should inform strategies to support healthy sleep across the life course.