Physical activity (PA) and sleep are both important to mental health. However, their joint effects on mental distress have not been well explored. The aim of this study was to investigate the joint effects of PA and sleep on mental health, as well as the dose-response relationships between PA and mental health under different sleep health statuses. A longitudinal panel study was adopted to evaluate the relationship between PA, sleep, and mental health among 66 healthy Chinese college students with four online questionnaire surveys. A mixed-effect model with individual-level random effect was used to analyze the interactive regulation effect of PA and sleep on mental health, and a generalized additive model with splines was further fitted to analyze dose-response relationships between variables. When sleep was at a healthy level, no significant difference in mental health was observed between different levels of PA (p > 0.05). However, poor sleepers with moderate and high PA levels indicated significantly fewer negative emotions than those with low PA levels (p = 0.001, p = 0.004). Likewise, poor sleepers who engaged in more moderate intensity PA could significantly reduce negative emotions (β = -0.470, p = 0.011) in a near-linear trend. In summary, both sleep and PA benefit mental health, and they probably regulate mental health through an interactive compensation mode. For good and poor sleepers, PA plays a different role in maintaining and improving mental health. Increasing moderate intensity PA up to moderate-and-high levels is recommended for those who simultaneously suffer from sleep and psychological health problems.
Keywords: interactive compensation effect; longitudinal panel study; mental health; physical activity; sleep.