(PS9-C63) Associations Between Chronotype and Suicidal Thoughts and Behaviors: A Meta-analysis

Background: Suicide is a leading cause of death worldwide and, as such, a major public health issue. With the goal of identifying modifiable risk factors for suicidal thoughts and behaviors (STBs), there has been a surge of research examining associations between sleep and suicidality. Two recent meta-analyses found several features of sleep, including sleep disturbances, insomnia, and nightmares, prospectively predict future STBs. However, circadian rhythms, a related yet distinct construct, has received relatively little attention regarding its association with suicide. As such, the goals of this study were to provide a quantitative synthesis of associations between chronotype, an indicator of circadian preference, and suicide.

Methods:   A systematic search of the literature was conducted in MEDLINE and PsycINFO to identify studies relevant to this meta-analysis. All research results were evaluated for inclusion by two independent coders. To be included, studies must be a) provide quantitative data on the association between chronotype (e.g., evening preference, morning preference, or neither) and suicide, b) assess STB distinctly from other constructs, and c) written in English. Hedge’s g was used to calculate pooled effect size. All analyses were conducted using Comprehensive Meta-Analysis.

Results: A total of 28 eligible studies were identified for inclusion in this meta-analysis. Evening chronotype was significantly associated with STBs, yielding a small-to-medium effect size (g = .30, p< .001). This association remained significant in analyses examining associations with suicidal ideation (SI; k=18, g=.28, p</em>< .001), but not in analyses of suicide attempts (SA; k=9, g=.20, p=.07). Further, effect sizes were greater (ps< .XX) in analyses comparing evening types to morning types (g [STB] =.55, p< .001; g [SI] =.63, p< .001), relative to analyses comparing evening types to individuals without a circadian preference (g [STB] =.35, p< .001; g [SI] =.32, p< .001). Morningness, relative to individuals without a circadian preference, was not associated with STB (g=-.14, p=.15), SI (g=-.20, p=.20), or SA (g=-.18, p=.30).

Conclusions: Findings suggest that eveningness is a robust correlate of STB. These effects appear to be specific to SI, and not SA, although fewer studies investigated the association between chronotype and SA. Findings were consistent across studies of clinical and community samples. However, all studies were cross-sectional, which precludes causal inference. It remains unclear whether circadian preference is a risk factor or concomitant of STB. Future research examining sleep and circadian factors as predictors of future STB would benefit by employing a prospective study design and considering the interactive effects of circadian rhythms (i.e., circadian alignment of sleep).