Abstract WP447: Race, Socioeconomic Status and Care Decisions After Intracerebral Hemorrhage

Abstract

Introduction: Spontaneous intracerebral hemorrhage (ICH) is a leading cause of disability. Incidence of dysphagia is similar across racial subgroups, but studies have shown increased odds of percutaneous endoscopic gastrostomy (PEG) placement among minorities, Medicaid enrollees, and those with low household income. Socioeconomic status (SES) indicators such as education or occupation have not been fully investigated as predictors of maximal care after ICH among minorities. Hypothesis: SES is associated with maximal care post-ICH in non-white patients. Methods: Patients with spontaneous ICH without underlying pathology were enrolled into a single center, observational cohort study. SES was imputed using neighborhood socioeconomic characteristics by patient’s zip code, according to an established method which calculates an index from six SES indicators: median household income, median home value, percent >25 years old who completed high school, percent >25 who completed college, percent with income from passive sources (investments, rents), and percent working in management, executive, or professional careers. Neighborhood SES data was obtained from the United States Census Bureau’s 2012-2016 American Community Survey. We used PEG placement as a surrogate for maximal care. Based on a single regression model with an interaction term for SES and race, models were created by racial-ethnic subgroups. Results: Among 607 patients comprised of 245 blacks, 56 Hispanics and 306 non-Hispanic whites, the SES index differed across subgroups (p < 0.001). White patients had the highest index and black patients had the lowest. After adjustment for age, gender, premorbid modified Rankin Scale score, and ICH Score, no association was observed between SES index and PEG placement in black (odds ratio 0.99, 95% confidence interval [0.90, 1.09], p= 0.87) or white patients (OR 0.94, 95%CI [0.84, 1.04], p=0.24). However, higher SES index was associated with reduced likelihood of PEG in Hispanic patients (OR 0.69, 95%CI [0.45, 0.95], p=0.041). Conclusions: A reduced likelihood of maximal care as measured by PEG placement was observed only among Hispanic patients with higher SES index. Further study into culturally-bound decisional factors after ICH is warranted.