The Potentially Causal Interplays among Sleep Quality, Perceived Friend Support, and Psychological Distress among Hong Kong Young People

Methodological innovations, such as random-intercept cross-lagged panel models (RI-CLPMs), allow for the disaggregation of within- and between-person variance in psychological processes, enabling stronger causal inference at the within-person level. Yet, RI-CLPMs have rarely been applied to examine both mediational processes and long-term reciprocal and feedback effects among sleep quality, perceived social support, and psychological distress in young people, particularly within the Hong Kong context. The present 3-wave study analyzed data from a prospective cohort of 1,626 young people (Mbaseline age = 27.16; SD = 4.97) recruited in Hong Kong in 2022 (T1), 2023 (T2), and 2024 (T3) via the RI-CLPMs. At the within-person level, worse sleep quality at T1 predicted lower social support from friends at T2 (β = 0.28), which in turn predicted higher psychological distress at T3 (β = -0.22), supporting a temporal indirect pathway from worse sleep quality (T1) → eroded support from friends (T2) → increased psychological distress (T3) (β = -0.062). Notably, the effect sizes for both the a- and b-paths in this mediation were substantial—each roughly twice the magnitude of the “large” threshold proposed in the literature—highlighting their practical significance. In addition, sleep quality and social support from friends showed within-person positive bidirectional associations and yielded an indirect effect (worse sleep quality at T1 → reduced social support from friends at T2 → further declines in sleep quality at T3; indirect effect: β = 0.060). These results suggest a dynamic feedback effect and highlight the long-term consequences of fluctuations in sleep quality beyond immediate or next-timepoint effects. Together, these findings underscore social support from friends as a key within-person mediator linking fluctuations in sleep quality to later distress and sleep quality over time.