Sustained social participation and dementia: evidence from a Japanese longitudinal cohort study with a time-varying...

Social participation is linked to a reduced risk of cognitive impairment and dementia. However, few studies have considered the bidirectional relationship between social participation and cognitive decline over time. We aimed to estimate the effect of sustained social participation on dementia risk reduction among older Japanese adults, accounting for the bias induced by the bidirectional relationship. A longitudinal study was conducted using data from the Japan Gerontological Evaluation Study. A baseline survey was administered in 2013, with dementia onset information up to 2022 obtained from the municipality registry (n = 47,698; median follow-up 9.2 years). Additional questionnaire surveys in 2016 and 2019 collected data on social participation and time-varying confounders. The average treatment effect (ATE) of sustained social participation on dementia onset was estimated using doubly robust targeted minimum loss-based estimation. Of the participants, 17.2 % got dementia. Participants with baseline social participation had a lower incidence of dementia (14.7 %) compared to those without social participation (17.8 %). Sustained social participation in any group at least once per week from 2013 to 2019 was associated with a 3.2 percentage point increase in dementia-free survival probability (95 % confidence interval, CI: 1.9, 4.5) compared to never participated. Among specific group types, participation in sports groups (ATE = 4.2; 95 % CI: 2.0, 6.4) and hobby groups (ATE = 5.3; 95 % CI: 2.5, 8.0) was significantly associated with increased dementia-free survival probability. Sustained social participation was associated with reduced dementia risk among older Japanese adults.