The association between subjective and objective sleep parameters and internalizing and externalizing problems in adolescents

• Disturbances in sleep have been previously associated with mental illness • Subjective and objective measurements of sleep were combined into a measure of ‘sleep health’. • Sleep health is inversely associated with reported psychopathological symptoms. • Subjective measures of sleep health are associated with symptoms of psychopathology • Objectives measures of sleep health are not associated with symptoms of psychopathology. Sleep practices change markedly in adolescence, leading to reductions in both duration and quality. Poor sleep health is a core component of many psychiatric disorders, yet most previous studies have mainly used singular rather than multidimensional sleep measures, which fail to account for all aspects of sleep as a behavior. In an adolescent cohort oversampled for high psychopathology risk ( n = 494, mean age 17.88), we used a validated 0–6 composite sleep health score based on sleep diary and actigraphy data. The Youth Self Report (YSR) quantified psychopathology symptoms as a total score, as broadband problem scales (e.g., externalizing problems), and narrowband scales (e.g., aggressive problems). Confounder-adjusted multiple linear regression models examined the association between the sleep composite score and total, broadband, and narrowband problem scores. A higher sleep composite score, indicating better sleep health, was linked to a lower total problems score (B -3.66, 95 % CI [-5.20; -2.12]) and was inversely related to all specific problem scales, including internalizing and externalizing problems, with the sole exception of aggressive problems. Secondary analyses revealed consistent associations between subjective sleep domains and various problem subscales, whereas objective domains showed none. Our study underscores the complementary roles of subjective and objective sleep measures in mental health research, illustrating that subjective measures are consistently associated with psychopathological symptoms and may offer personal relevance, while objective measures might provide accuracy and consistency by capturing detailed sleep patterns. However, limitations include the study's cross-sectional design and potential selection bias in the sleep measurement group.