Back

Social connections are differentially related to perceived and physiological age acceleration amongst older adults

Fancourt, D.; Bloomberg, M.; Steptoe, A.

2025-02-05 epidemiology
10.1101/2025.02.03.25320261 medRxiv
Show abstract

Human social connections are complex ecosystems formed of structural, functional and quality components. Weak social connections are associated with adverse age-related health outcomes, but we know little about the ageing-related processes underlying this. Using data from 7,047 adults aged 50+ in the English Longitudinal Study of Ageing, we explored associations between diverse aspects of social connections and both older subjective age and accelerated physiological age, which provide complementary psycho-behavioural and biological mechanistic explanations. We created and validated a novel physiological ageing index using clinical indicators pertaining to the cardiovascular, respiratory, haematologic, metabolic and cognitive systems using principal component analysis. Doubly-robust estimations using inverse-probability-weighted regression adjustment estimators showed that living alone, low social integration and high social isolation were risk factors for physiological age acceleration, with those who lived alone on average 1.4 years older than those who lived with others (95% CI 0.7-2.2 years older; 50% greater age acceleration than people who live with others). However, weak social connections were not related to older subjective age. Analyses were robust to multiple sensitivity analyses and maintained four years later. We use the findings to generate a novel hypothesis that accelerated physiological ageing may be a mechanism underpinning the relationship between weak social connections and age-related morbitidy and mortality outcomes.

Published in Nature Communications (predicted rank #7) · training set

Matching journals

The top 13 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.