Heikkilä · Age and ageing 2024 · prospective multi-cohort study · n=33263

Stimulating leisure-time activities and the risk of dementia: a multi-cohort study.

Cited 10 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective multi-cohort observational study

PubMed 39003234 · doi:10.1093/ageing/afae141 · record verified 2026-08-26

What was done

Data were pooled from five prospective cohort studies in Finland and Sweden comprising 33,263 dementia-free adults aged 50 and older at baseline (mean follow-up 7.0 years, extending up to two decades). Mental, social, outdoor, consumptive, and physical leisure-time activities were assessed via self-report. Incident dementia was identified through clinical diagnoses or healthcare and death registers. Cox regression models evaluated hazard ratios (HRs) and 95% confidence intervals (CIs) for dementia risk stratified by follow-up duration (<10 years vs. ≥10 years).

What was found

A total of 1,408 individuals developed incident dementia. In follow-up under 10 years, active participation was associated with lower dementia risk across mental (HR: 0.52, 95% CI: 0.41 to 0.65), social (HR: 0.56, 95% CI: 0.46 to 0.72), outdoor (HR: 0.70, 95% CI: 0.58 to 0.85), consumptive (HR: 0.67, 95% CI: 0.53 to 0.94), and physical activity (HR: 0.62, 95% CI: 0.51 to 0.75), as well as activity variety (HR: 0.54, 95% CI: 0.43 to 0.68) and overall frequency (HR: 0.41, 95% CI: 0.34 to 0.49). In follow-up of 10 years or longer, all associations attenuated toward the null.

Why it matters

This study suggests that apparent protective associations between leisure activities and dementia are largely confined to the short term, likely reflecting reverse causation (preclinical cognitive decline reducing participation) rather than causal neuroprotection.

Limits

Leisure activities were self-reported, introducing potential recall error. The observational design cannot rule out residual confounding or exposure dilution due to unmeasured changes in activity levels over decades. The study population was restricted to Scandinavian cohorts, which may limit generalizability to other populations.

Cited by