Which Comes First?: Depression and Sleep Disturbance Among Sexual Minority Men

Background: Depression and sleep disturbances are known to have a high association, yet the precise causal directionality between these two factors remains unclear. Understanding the mechanisms linking depression and sleep issues could significantly contribute to enhancing our mental and physical health. This is particularly relevant for sexual minority men (SMM), who face a disproportionate prevalence of mental health issues. Despite increased awareness of these disparities, there remains a crucial gap in understanding the dynamics of depression and sleep disturbances within the SMM population. By investigating the cross-lagged effect between different time points of depression and sleep disturbance, we aim to shed light on the intricate association between depression and sleep disturbance among SMM.

Methods: This longitudinal study analyzed 111 SMM recruited through geospatial networking apps, who provided data on all study variables at multiple time points. Participants assessed the level of depression (measured by Center for Epidemiological Studies Depression scale, α=.878) and sleep disturbance (measured by PROMIS-8a questionnaire, α=.796) at baseline and follow-up survey over a three-year period. Autoregressive cross-lagged modeling was employed to investigate the directional relationship between depression and sleep disturbance, specifically determining whether depression predicts sleep disturbance or vice versa. The term 'autoregressive' indicates paths linking the stability of variables over time, while 'cross-lagged' denotes paths between depression and sleep at different time points, suggesting a potential causal relationship.

Results: Confirmatory factor analysis was conducted to assess the stability of the measurement structure over time. Additionally, multiple-group analysis was employed to examine measurement invariable across different time points. The research model demonstrated strong fit indices and measurement stability. All autoregressive paths were significant, confirming the ongoing state of depression and sleep disturbance over time (Depression: β=.524, p<.001, Sleep disturbance: β=.319, p=.002). Particularly noteworthy, the cross-lagged path from depression to subsequent sleep disturbance was statistically significant (β=.173, p<.017). This indicates that higher levels of depression at baseline were predictive of increased sleep disturbances at follow-up, whereas the reverse path—from sleep disturbance to depression—was not significant.

Conclusion: Our study reveals a significant unidirectional effect, with symptoms of depression predicting subsequent sleep disturbance among SMM. While many longitudinal studies suggest that sleep problems precede depressive mood, our findings emphasize the need for further exploration. Previous research often focuses on specific groups, such as particular periods of age or clinic populations. Future research should explore the mechanisms underlying this relationship and develop tailored interventions to improve well-being and quality of life considering unique experience of SMM, potentially filling crucial gaps in addressing their mental health needs.