Association Between Physical Activity and Risk of Disabling Dementia in Japan

JAMA Netw Open. 2022 Mar 1;5(3):e224590. doi: 10.1001/jamanetworkopen.2022.4590.


Importance: The associations of daily total physical activity and total moderate to vigorous physical activity (MVPA) with dementia are still unclear.

Objective: To investigate the association between daily total physical activity and subsequent risk of disabling dementia in large-scale, extended follow-up prospective study.

Design, setting, and participants: This prospective cohort study used data from questionnaires collected between 2000 and 2003 from 8 areas from the Japan Public Health Center-based Prospective Disabling Dementia Study. Participants included adults aged 50 to 79 years in with available follow-up data on disabling dementia. Data analysis was performed from February 1, 2019, to July 31, 2021.

Exposures: Daily total physical activity, total MVPA, and leisure-time MVPA.

Main outcomes and measures: The main outcome was incidence of disabling dementia during the dementia ascertainment period between 2006 and 2016, based on the national long-term care insurance system. Risks of dementia in association with daily total physical activity, total MVPA, and leisure time MVPA were calculated using multivariable adjusted hazard ratios (aHRs).

Results: Among 43 896 participants (mean [SD] age, 61.0 [7.5] years; 23 659 [53.9%] women), 5010 participants were newly diagnosed with disabling dementia during a mean (SD) of 9.5 (2.8) years in the dementia ascertainment period. In the highest daily total physical activity group, compared with the lowest activity group, risk of dementia was lower in men (aHR, 0.75 [95% CI, 0.66-0.85]; P for trend < .001) and women (aHR, 0.75 [95% CI, 0.67-0.84]; P for trend < .001). Similar inverse associations were observed in men and women for total MVPA (men: aHR, 0.74 [95% CI, 0.65-0.84]; P for trend < .001; women: aHR, 0.74 [95% CI, 0.66-0.83]; P for trend < .001) and leisure-time MVPA (men: aHR, 0.59 [95% CI, 0.53-0.67]; P for trend < .001; women: aHR, 0.70 [95% CI, 0.63-0.78]; P for trend < .001). However, these inverse associations disappeared when participants diagnosed with disabling dementia within 7 years of the starting point were excluded in men (aHR, 0.93 [95%CI, 0.77-1.12]) and within 8 years were excluded in women (aHR, 0.86 [95%CI, 0.71-1.04]). The association remained significant among men in the highest vs lowest group of leisure-time MVPA, after excluding participants diagnosed within the first 9 years (aHR, 0.72 [95% CI, 0.56-0.92]; P for trend = .004).

Conclusions and relevance: This cohort study examined associations of daily total physical activity and total MVPA with risk of disabling dementia. The findings suggest that a high level of leisure-time MVPA was associated with decreased risk of disabling dementia in men.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Cohort Studies
  • Dementia* / epidemiology
  • Exercise*
  • Female
  • Humans
  • Japan / epidemiology
  • Male
  • Middle Aged
  • Prospective Studies