BACKGROUND: Volunteering has been linked to reduced mortality in older adults, but the mechanisms explaining this effect remain unclear. This study investigated whether volunteering is associated with increased survival in participants of the English Longitudinal Study of Ageing and whether differences in survival are modified by functional disabilities. METHODS: A multivariate Cox Proportional Hazards model was used to estimate the association of volunteering with survival over a period of 10.9 years in 10 324 participants, while controlling for selected confounders. To investigate effect modification by disability, the analyses were repeated in participants with and without self-reported functional disabilities. RESULTS: Volunteering was associated with a reduced probability of death from all causes in univariate analyses (HR=0.65, CI 0.58 to 0.73, p<0.0001), but adjustment for covariates rendered this association non-significant (HR=0.90, CI 0.79 to 1.01, p=0.07). Able-bodied volunteers had significantly increased survival compared with able-bodied non-volunteers (HR=0.81, 95% CI 0.69 to 0.95, p=0.009). There was no significant survival advantage among disabled volunteers, compared with disabled non-volunteers (HR=1.06, CI 0.88 to 1.29, p=0.53). CONCLUSIONS: Volunteering is associated with reduced mortality in older adults in England, but this effect appears to be limited to volunteers who report no disabilities.