Abstract
BACKGROUND: It is well established that all types of movement behaviors, including moderate-to-vigorous physical activity (MVPA), light-intensity physical activity (LIPA), sedentary behavior (SB), and sleep, are associated with the risk of incident dementia, all-cause mortality, and premature death. However, it remains unclear whether reallocating time from one type to another is associated with these outcomes. In addition, the extent to which genetic susceptibility modifies the association between physical activity and dementia risk still warrants further investigation. METHODS: This study included 94 086 dementia-free participants from the UK Biobank with valid accelerometer and genomic data. Time spent MVPA, LIPA, SB, and sleep were derived from wrist-worn accelerometers. Genetic susceptibility of dementia was assessed by polygenic risk score (PRS) consisting of 82 single nucleotide polymorphisms. The isotemporal substitution model was applied to explore how reallocating time between movement behaviors was associated with incident dementia, mortality, and premature death. RESULTS: Of 94 086 included participants, 52 853 (56.2%) were female, and the mean (standard deviation, SD) age was 62.3 (7.8) years. Reallocating 1 h/day to MVPA from LIPA, SB, and sleep was associated with a 19%, 26%, and 18% lower risk of incident dementia (adjusted hazard ratios [HRs] and 95% confidence intervals [CIs]: 0.81 [0.68, 0.95], 0.74 [0.63, 0.87], and 0.82 [0.69, 0.96], respectively). A 22%, 30%, and 29% reduced risk of mortality were observed when reallocating 1 h/day from LIPA, SB, and sleep to MVPA (0.78 [0.72, 0.84], 0.70 [0.65, 0.75], and 0.71 [0.66, 0.77], respectively). Replacing 1 h/day of SB with MVPA, LIPA, and sleep was associated with a 26%, 8%, and 9% lower risk of incident dementia (0.74 [0.63, 0.87], 0.92 [0.87, 0.97], and 0.91 [0.85, 0.97], respectively), and reallocating 1 h/day from SB to LIPA (0.89 [0.87-0.92]) or MVPA (0.70 [0.65-0.75]) was associated with reduced risk of mortality. Similar results could be seen in premature death. Participants with high levels of MVPA and low genetic risk showed 72% lower risk of dementia comparing to participants with low levels of MVPA and high PRS (0.28 [0.17-0.50]). CONCLUSIONS: Reallocating time to MVPA from any behavior and substituting physical activity of any intensity for SB were associated with decreased risks of incident dementia, mortality, and premature death, suggesting the significance of maintaining a physically active lifestyle among old adults. Moreover, increasing MVPA level could partially attenuate the strength of association between genetic susceptibility and the risk of dementia.