The prevalence of diabetes mellitus (DM) is a significant public health concern, especially among individuals with short sleep duration. Understanding the relationship between physical exercise and DM in this population is crucial for developing effective prevention strategies. However, the presence of a potential threshold effect of exercise on DM risk remains unclear.
Using data from the National Health and Nutrition Examination Survey (NHANES) spanning from 2007 to 2018, this population-based study investigated the association between physical exercise and DM in individuals with short sleep duration (no more than 7 hours per night). Weighted logistic regression analyses were conducted, adjusting for demographic and lifestyle factors. Additionally, a two-piecewise linear regression model was employed to identify any threshold effect of exercise on DM risk.
This study included 15,092 participants identified with short sleep duration. Demographic characteristics stratified by DM status indicate higher prevalence among certain groups, such as middle-aged and older adults, males, and non-Hispanic Whites. The analysis revealed an inverse association between exercise levels and DM prevalence among the short sleep population. In the fully adjusted model, individuals engaging in sufficient exercise (> 600 MET-minutes/week) exhibited significantly reduced odds of developing DM [OR (95% CI): 0.624(0.527,0.738), p < 0.001]. Furthermore, the segmented regression model identified an inflection point at 2000 MET-minutes/week, below which a significant correlation between exercise and DM was observed.
This study provides evidence of a threshold effect of physical exercise on its association with DM in individuals with short sleep duration. Tailored exercise interventions targeting this population may help mitigate DM risk and improve overall health outcomes. Further research is warranted to validate these findings and explore optimal exercise thresholds for DM prevention strategies.