Association of social contact with dementia and cognition: 28-year follow-up of the Whitehall II cohort study.
Level 3 - non-randomized controlled study
Prospective longitudinal cohort study with 28-year follow-up.
PubMed 31374073 · doi:10.1371/journal.pmed.1002862
What was done
Researchers analyzed data from the Whitehall II prospective cohort study of London civil servants (aged 35–55 at baseline in 1985–1988, followed to 2017). Social contact with non-cohabiting friends and relatives was self-reported across six waves. Incident dementia was identified through three linked electronic clinical and mortality databases, and cognitive function (memory, fluency, reasoning) was assessed across five waves. Cox proportional hazards regression with inverse probability weighting evaluated the association between social contact at ages 50, 60, and 70 and subsequent dementia risk, adjusting for demographics, socioeconomic status, education, health behaviors, employment, and marital status. Linear mixed models examined midlife social contact (ages 45–55) and subsequent 14-year cognitive trajectories.
What was found
Among 10,228 participants (mean baseline age 44.9 years, SD 6.1; 33.1% female; 89.1% white), more frequent social contact at age 60 was associated with lower dementia risk (hazard ratio [HR] 0.88 per SD increase, 95% CI 0.79–0.98, p = 0.02). Associations at age 50 (HR 0.92, 95% CI 0.83–1.02, p = 0.13) and age 70 (HR 0.91, 95% CI 0.78–1.06, p = 0.23) were not statistically significant. The association was driven by contact with friends (HR 0.90, 95% CI 0.81–1.00, p = 0.05), with no association for contact with relatives. Highest versus lowest tertile of midlife social contact was associated with 0.07 SD higher global cognitive function (95% CI 0.03–0.11, p = 0.002), maintained over 14 years of follow-up.
Why it matters
With 28 years of follow-up reducing the risk of reverse causation bias, this study provides strong observational evidence that frequent social contact in late midlife—especially with friends—is linked to lower dementia risk and better cognitive trajectories.
Limits
Dementia diagnosis was ascertained from electronic health and mortality records rather than standardized in-person clinical examinations, likely missing milder cases. The cohort consisted of London civil servants who were predominantly male (66.9%) and white (89.1%), limiting generalizability. Social contact was self-reported.
Cited by
- supports Maintaining or increasing social relationships in older age is associated with a significantly reduced risk of developing dementia.