Submitted:
16 November 2026
Posted:
18 November 2026
You are already at the latest version
Abstract
Background: Sleep disturbance and anxiety often co-occur in young people. Earlier syntheses report bidirectional longitudinal links between sleep and internalizing symptoms [4,6] , but it is unclear how analytic model (cross-lagged panel model [CLPM], random-intercept CLPM [RI-CLPM], cross-lagged panel network [CLPN]), time lag, informant and anxiety subtype shape conclusions about direction. Objective: To map recent longitudinal cross-lagged studies of sleep and anxiety in non-clinical children, adolescents and young adults, and to chart how the reported direction relates to model, lag, informant and anxiety type. Methods: Scoping review following JBI guidance and PRISMA-ScR. PubMed/MEDLINE was searched with ten keyword-specific query sets for studies published from January 2022 to November, 2026; exports were merged and de-duplicated, and two reviewers screened records. Eligible studies were multi-wave panel studies of non-clinical participants aged 3–25 years with at least one cross-lagged or mediation/moderation analysis involving sleep and anxiety. Data were charted descriptively; risk of bias was not assessed. Results: Fifteen reports from 14 cohorts were included (baseline N = 52,548 across 13 cohorts; sample size not reported in 1 cohort). Eleven reports tested sleep–anxiety paths directly (three sleep→anxiety only; five bidirectional; one anxiety→sleep only; one developmental shift from anxiety→sleep to reciprocal; one symptom-level network study). Four reports addressed sleep and anxiety within mediation or moderation models. Direction appeared to vary with model, lag and anxiety type, but studies were few and heterogeneous. The only social anxiety study reporting anxiety→sleep had 71 participants; a larger cohort reported sleep predicting later social anxiety. Conclusions: Recent PubMed-indexed longitudinal evidence in young people is consistent with a reciprocal association, with sleep-to-anxiety paths reported somewhat more often. Evidence for age-specific or subtype-specific reversal is limited and confounded by design. Future studies should report within-person models, justify lags, use multiple informants and objective sleep measures. These findings do not establish causality.
Keywords:
sleep disturbance
; anxiety
; cross-lagged panel model
; random-intercept cross-lagged panel model
; adolescents
; scoping review
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.