Abstract
OBJECTIVE: To investigate whether substituting different intensities of physical activity (PA) for sedentary time (ST) can benefit the prevention of type 2 diabetes (T2D) across varying levels of T2D genetic susceptibility.
PARTICIPANTS AND METHODS: We included 73,272 White British participants from the UK Biobank without prevalent T2D. Wrist-worn accelerometry was used to derive sleep, ST, light PA, and moderate-to-vigorous PA (MVPA) based on machine learning. Polygenic risk scores for T2D were computed based on 138 genome-wide important, unrelated genetic markers. We used compositional isotemporal substitution modeling in logistic regression, with adjustment for putative confounders. Accelerometer data were collected between June 1, 2013, and December 23, 2015. Follow-up continued until December 9, 2022 (England/Wales) and December 19, 2022 (Scotland).
RESULTS: Over a median 8.1-year follow-up, 1399 incident T2D cases were identified. Reallocating 30 minutes/day of ST into time-equivalent light PA and MVPA was associated with 5% (odds ratio, 0.95; 95% CI, 0.94 to 0.97) and 18% (odds ratio, 0.82; 95% CI, 0.79 to 0.85) lower odds of T2D, respectively, independent of genetic risk. Shifting 15 minutes/day of ST to MVPA or 60 minutes/day of ST to light PA was associated with similar odds of T2D.
CONCLUSION: Irrespective of genetic susceptibility to T2D, replacing ST with equivalent physically active time (either at light or moderate-to-vigorous intensity) is associated with lower T2D odds. Individuals at all genetic risk levels may achieve the same degree of T2D odds reduction by substituting a larger volume of light PA for ST as they would by replacing ST with a smaller volume of MVPA.