Abstract Background and aims Socioeconomic inequalities in post-stroke outcomes persist, but causal mechanisms remain unclear.We used counterfactual mediation analysis to quantify these pathways. Methods We included all first-ever stroke patients from the South London Stroke Register (1995-2025), stratified by decade. Socioeconomic status (SES) was measured using Index of Multiple Deprivation quintiles(1-2 vs 3-5). Outcomes at 3-months, 1-year and 5-years included disability (Barthel Index), dependency (Frenchay Activities Index), quality of life (SF-12), and depression/anxiety (HADS). We estimated total effects using regression models adjusted for age and sex, then conducted counterfactual mediation analysis where inequalities existed. Mediators included cardiovascular risk factors (hypertension, diabetes, smoking, atrial fibrillation, alcohol, hypercholesterolaemia) at baseline and follow-up, stroke severity, stroke unit access, secondary-prevention-medications, and rehabilitation.Multiple imputation addressed missing data. Results Among 8,612 patients,socioeconomic inequalities in dependency and physical quality of life were present in 2005-2014 and 2015-2025, but not 1995-2004. Cardiovascular risk factors at follow-up emerged as dominant mediators, explaining 28.9%(95%CI:8.2-59.8%) of dependency inequalities and 30.1%(95%CI:8.8-101.8%) of physical health inequalities.Mental health inequalities appeared in 2015-2025,with cardiovascular risk factors mediating 13.8%(95%CI:0.4-37.5%) and 24.5%(95%CI:7.5-62.4%) of depression at 3-months and 1-year respecrtively and 19.7%(95%CI:4.7-64.7%) of anxiety inequalities at 1-year.No significant mediation effects were found for stroke severity, stroke unit access, secondary prevention medications, or rehabilitation. Conclusions Socioeconomic inequalities in stroke outcomes are primarily mediated by cardiovascular risk factors during follow-up.Interventions targeting risk factor control in deprived populations may reduce these inequalities. Conflict of interest Camila Pantoja-Ruiz: Nothing to disclose
Building similarity graph...
Analyzing shared references across papers
Loading...
C Pantoja-Ruiz
King's College London
Anita Lindmark
Umeå University
Eva Emmett
King's College London
European Stroke Journal
King's College London
Umeå University
Building similarity graph...
Analyzing shared references across papers
Loading...
Pantoja-Ruiz et al. (Fri,) studied this question.
synapsesocial.com/papers/69fd7e23bfa21ec5bbf06521 — DOI: https://doi.org/10.1093/esj/aakag023.878