A Meta-Regression of psychosocial factors associated with sleep outcomes in mindfulness-based intervention trials.

Objectives: In this meta-regression, we aimed to explore associations between changes in psychosocial factors and changes in sleep disturbance during mindfulness-based interventions (MBIs). We also investigated participant-specific and methodological factors associated with sleep disturbance during MBIs.

Method: We utilized data from a published meta-analysis of 40 randomized controlled trials of MBIs (published from inception to 2020) with a sleep disturbance outcome measure in healthy and clinical adult populations. We conducted meta-regressions to test associations between sleep improvements following MBIs and psychosocial factors, as well as demographic and methodological factors.

Results: MBIs were associated with significant reductions in sleep disturbance (SMD = -0.523; 95% CI = -0.678 to -0.368) and psychosocial factors (SMD = -0.213 - -0.894). Reductions in sleep disturbance were associated with reductions in stress (r = 0.74, p = .02) and depression (r = 0.90, p < .001).

Conclusions: MBIs improve sleep disturbance across a wide range of healthy and clinical populations. Stress and depression may be important psychosocial factors associated with sleep disturbance. Future RCTs should include measures of additional factors and should investigate longitudinal associations between psychosocial, demographic, and methodological factors with changes in sleep disturbance to test mechanisms and to identify "active ingredients" of MBIs.