Among 7,267 stroke survivors in the UK Biobank followed for a median 10.2 years, social isolation independently raised all-cause mortality risk by 41% (HR 1.41), CVD mortality by 57% (HR 1.57), and depression by 35% (HR 1.35). Loneliness carried even steeper penalties for dementia — a 67% higher risk (HR 1.67) — alongside 23% and 45% elevated risks for all-cause and CVD mortality respectively. Critically, these associations were significantly more pronounced in stroke survivors than in the 387,041 stroke-free comparators, with interaction p-values of 0.006 and 0.042 for all-cause and CVD mortality. Health behaviours (smoking, physical inactivity, poor diet) mediated 12–55% of these associations, suggesting meaningful but incomplete explanatory pathways.
Stroke survivors represent a population where social connection is both uniquely threatened — by disability, communication difficulties, and role loss — and uniquely consequential. This large observational study adds specificity to the Surgeon General's 2023 loneliness advisory by quantifying excess risk in a vulnerable subgroup rather than the general population. The mediation finding is actionable: nearly half the mortality signal runs through modifiable behaviours, meaning social prescribing, peer-support programmes, and rehabilitation that targets lifestyle habits simultaneously may compound benefits. Limitations include the UK Biobank's predominantly White, relatively healthy cohort, self-reported loneliness measures, and residual confounding inherent to observational design. Still, the effect sizes are large enough — and consistent with prior mechanistic evidence linking isolation to cortisol dysregulation and inflammation — to treat social connection as a legitimate post-stroke clinical priority.