Association Between Insurance Status on Risk of Stroke Recurrence with Incident Cryptogenic Stroke: A Secondary...

Economic and social disparities in health care contribute to differential outcomes among patients with acute stroke. A patient's insurance status reflects access to care and serves as a surrogate of social determinants of health, which can be targeted for intervention. We examined the relationship between insurance status and risk of recurrent stroke, major bleeding or death following an embolic stroke of undetermined source (ESUS).

This is a secondary analysis of a United States multicenter (n=27) retrospective observational cohort of patients with ESUS, which examined the association between insurance status and the primary composite outcome of recurrent stroke, major bleeding, or death. Insurance providers were categorized by Private/Veterans Affairs (VA), Medicare, Other, and Medicaid/pending/none, as surrogate exposures for economic and social determinants of health. Unadjusted and adjusted Cox proportional hazards models were built to estimate the association between insurance carrier and the primary and secondary outcomes, adjusting for age, stroke severity, pre-stroke disability, atherosclerotic risk factors, and potential embolic sources.

Of the 2130 patients included, 882 were Medicare beneficiaries (41.4%), 744 had private/VA insurance (34.9%), 368 had Medicaid/pending/none (17.3%) and 136 patients had other insurance (6.4%). Compared to Private/VA beneficiaries, patients with Medicaid/pending/no insurance were younger, more frequently Black, with more severe baseline stroke severity and higher prevalence of atherosclerotic risk factors. Patients with Medicaid/pending/no insurance had a higher probability of the primary outcome in unadjusted and adjusted regression (adjusted hazard ratio [aHR] 1.37, 95% confidence interval [CI] 1.08-1.74, adjusted α=0.03) when compared to Private/VA beneficiaries. Although not statistically significant (adjusted α=0.08), the risk of recurrent stroke trended toward being greater among patients with Medicaid/pending/no insurance (aHR 1.40, 95% CI, 1.04-1.87). Among subgroups, patients with Medicaid/pending/no insurance had a higher risk of recurrent stroke among patients with mild stroke, patients younger than 50 years of age, and male sex (pinteraction <0.01 for all).

Patients with Medicaid/pending/no insurance remain at a significantly higher risk of the composite endpoint in this diverse cohort of patients with ESUS independent of vascular risk factor presence. Optimizing access to care and follow-up may attenuate the ongoing, significant disparities in ESUS patients.