Social relationships and mortality risk: a meta-analytic review.
Level 3 - non-randomized controlled study
Meta-analysis of observational prospective cohort studies.
PubMed 20668659 · doi:10.1371/journal.pmed.1000316
What was done
A meta-analytic review of 148 studies encompassing 308,849 participants to determine the extent to which social relationships influence mortality risk, identify key predictive dimensions, and evaluate moderators. Extracted data included participant characteristics (cause of mortality, initial health status, pre-existing conditions) and study variables (length of follow-up, assessment method of social relationships). Random-effects models were used to calculate weighted average effect sizes.
What was found
Stronger social relationships were associated with a 50% increased likelihood of survival (random effects weighted average OR = 1.50, 95% CI 1.42 to 1.59). The association remained consistent across age, sex, initial health status, cause of death, and follow-up duration. Effect sizes differed significantly by measurement type (p < 0.001): complex measures of social integration showed the strongest association (OR = 1.91, 95% CI 1.63 to 2.23), whereas binary indicators of living alone versus with others showed the weakest association (OR = 1.19, 95% CI 0.99 to 1.44).
Why it matters
This meta-analysis demonstrates that social integration is robustly associated with reduced mortality risk across diverse populations, yielding effect sizes comparable to standard clinical risk factors.
Limits
The included studies are observational, precluding definitive causal inference. The abstract does not provide numerical heterogeneity statistics or details on control for specific residual confounders such as baseline socioeconomic status. Binary residential status had a confidence interval crossing the null.