Sleep quality, particularly daytime dysfunction and sleep disturbance, and circadian preference are robust, prospective, and modifiable predictors of adolescent anxiety, suggesting that sleep should be prioritized as a transdiagnostic therapeutic target.
Key Findings
Results
Overall sleep quality showed one of the strongest cross-sectional associations with adolescent anxiety among all sleep-related factors examined.
Cross-sectional analyses included k=117 independent samples from a total of N=416,426 participants
Overall sleep quality showed a correlation of r=0.34 with anxiety
The odds ratio for poor overall sleep quality and anxiety was OR=2.93
This was among the strongest associations found across all sleep dimensions examined
Results
Sleep disturbance and daytime dysfunction showed the strongest cross-sectional correlations with adolescent anxiety among specific sleep quality components.
Sleep disturbance showed a correlation of r=0.36 with anxiety
Daytime dysfunction showed the highest correlation at r=0.38 with anxiety
These two dimensions outperformed other sleep-related factors in cross-sectional association strength
Both were identified as 'robust, prospective, and modifiable predictors of adolescent anxiety'
Results
Sleep onset latency showed a significant but moderate cross-sectional association with adolescent anxiety.
Sleep onset latency correlated with anxiety at r=0.20
This association was weaker than sleep disturbance and daytime dysfunction but remained statistically significant
It was identified as part of the broader sleep quality construct linked to anxiety
Results
Short sleep duration and evening circadian preference showed significant but weaker cross-sectional associations with adolescent anxiety.
Short sleep duration correlated with anxiety at r=-0.07 and OR=1.52
Evening circadian preference correlated with anxiety at r=-0.16 and OR=1.58
Both associations, while statistically significant, were described as weaker compared to sleep quality measures
The negative sign of the correlations reflects directionality conventions (shorter duration or more evening preference associated with higher anxiety)
Results
Longitudinal analyses showed that baseline poor sleep quality prospectively predicted anxiety, with predictive power peaking within a 12-month window.
Longitudinal analyses included k=32 independent samples
Baseline poor sleep quality predicted later anxiety with r=0.28 and OR=1.45
Predictive power peaked within a 12-month follow-up window
This prospective relationship supports a directional role of sleep quality in anxiety development
Results
Short sleep duration as a continuous measure was a significant longitudinal predictor of anxiety, but its categorical risk was non-significant.
Short sleep duration as a continuous predictor showed r=-0.11 in longitudinal analyses
When treated as a categorical variable (short vs. not short), the odds ratio was non-significant at OR=1.07
This discrepancy suggests that the graded nature of sleep duration may matter more than a simple threshold classification
This contrasts with the significant categorical OR found in cross-sectional analyses (OR=1.52)
Results
RCT evidence showed that poor sleep quality, sleep restriction, and evening chronotype increased anxiety, while sleep interventions reduced anxiety.
RCT analyses included k=19 independent samples
Conditions involving poor sleep quality, sleep restriction, or evening chronotype were associated with increased anxiety (SMD=0.43)
Sleep interventions, including cognitive behavioral therapy for insomnia (CBT-I), reduced anxiety (SMD=-0.47)
The near-symmetrical effect sizes for worsening (SMD=0.43) and improvement (SMD=-0.47) suggest bidirectional modifiability of the sleep-anxiety relationship
Methods
The meta-analysis synthesized a large and methodologically diverse evidence base covering cross-sectional, longitudinal, and experimental study designs.
A total of 129 independent samples were included with a combined N=416,426 participants
Databases searched included PubMed, Embase, PsycINFO, Web of Science, and Scopus from inception to October 10, 2023
Eligible studies required validated anxiety and sleep measures or clinical diagnostic criteria
Study designs included cross-sectional (k=117), longitudinal (k=32), and RCT (k=19) samples
What This Means
This research synthesized findings from 129 studies involving over 416,000 adolescents to understand how different aspects of sleep are connected to anxiety. The researchers looked at studies that simply measured sleep and anxiety at one point in time, studies that followed adolescents over time, and experiments that deliberately changed sleep to see how it affected anxiety. Across all three types of evidence, poor sleep quality — especially experiencing disturbances during the night and feeling impaired during the day as a result — was consistently and meaningfully linked to higher anxiety levels in teenagers. The strength of association was stronger for sleep quality measures than for simply sleeping too few hours.
This research suggests that the relationship between sleep and adolescent anxiety is not just a snapshot association but a prospective one: teenagers who had poor sleep quality were more likely to develop or worsen anxiety over the following months, with the predictive window strongest within 12 months. Experimental evidence from randomized controlled trials further supported this, showing that when sleep was restricted or disrupted, anxiety increased, and when sleep was improved through interventions like cognitive behavioral therapy for insomnia (CBT-I), anxiety decreased by a meaningful amount. Evening chronotype — a natural tendency to stay up late — also emerged as a modifiable risk factor.
The practical implication of these findings is that sleep quality may be an important and treatable contributor to anxiety in adolescents. Rather than treating sleep problems and anxiety as separate issues, this research supports the idea that improving sleep — particularly addressing nighttime disturbances and daytime impairment — could be a useful part of efforts to prevent or reduce anxiety during this vulnerable developmental period. The authors suggest sleep should be considered a 'transdiagnostic therapeutic target,' meaning it may be relevant across different types of anxiety and mental health concerns in young people.
Zhang J, Wan J, Wang Y, Li H, Lei Y. (2026). Explore the impact of sleep-related factors on anxiety in adolescents: A meta-analysis of cross-sectional, longitudinal and RCT studies.. Journal of affective disorders. https://doi.org/10.1016/j.jad.2026.121320