Leisure activity participation and risk of dementia: An 18-year follow-up of the Whitehall II Study

Abstract

Objective: To test the hypothesis that leisure activity participation is associated with lower dementia risk, we examined the association between participation in leisure activities and incident dementia in a large longitudinal study with average 18-year follow-up.

Methods: We used data from 8,280 participants of the Whitehall II prospective cohort study. A 13-item scale assessed leisure activity participation in 1997-1999, 2002-2004, and 2007-2009, and incidence of dementia (n cases = 360, mean age at diagnosis 76.2 years, incidence rate 2.4 per 1,000 person-years) was ascertained from 3 comprehensive national registers with follow-up until March 2017. Primary analyses were based on complete cases (n = 6,050, n cases = 247) and sensitivity analyses used multiple imputation for missing data.

Results: Participation in leisure activities at mean age 55.8 (1997-1999 assessment), with 18.0-year follow-up, was not associated with dementia (hazard ratio [HR] 0.92 [95% confidence interval 0.79-1.06]), but those with higher participation at mean age 65.7 (2007-2009 assessment) were less likely to develop dementia with 8.3-year follow-up (HR 0.82 [0.69-0.98]). No specific type of leisure activity was consistently associated with dementia risk. Decline in participation between 1997-1999 and 2007-2009 was associated with subsequent dementia risk.

Conclusion: Our findings suggest that participation in leisure activities declines in the preclinical phase of dementia; there was no robust evidence for a protective association between leisure activity participation and dementia. Future research should investigate the sociobehavioral, cognitive, and neurobiological drivers of decline in leisure activity participation to determine potential approaches to improving social participation of those developing dementia.

Figure 1. Flow chart of participants in the study (n = 8,280)

Figure 2. Association of each leisure activity in 1997–1999 with subsequent incident dementia

CI = confidence interval; HES = hospital episode statistics; HR = hazard ratio.