Racial and Ethnic Disparities in Mortality in US Veterans Following an Incident Subarachnoid Hemorrhage

Abstract

Background Blacks and Hispanics have been shown to have higher stroke mortality than Whites. However, few studies have investigated health disparities in post-stroke mortality for specific stroke types, such as subarachnoid hemorrhage (SAH). We analyzed risk of mortality by race/ethnicity after incident SAH in a national sample of stroke patients from the Veteran's Health Administration (VHA). Methods We included 742 patients (98% male, mean age 63±13 years) with a first occurrence of an inpatient SAH in a VHA hospital between 2002-2007. SAH was determined by ICD-9 code 430. We categorized patients as non-Hispanic White, non-Hispanic Black, or Hispanic. Veterans with a prior definite stroke type (including TIA) were excluded. We constructed a piecewise multivariate survival model for post event all-cause mortality, separated according to 1 year post admission follow-up intervals. We then looked at 30-day mortality for vascular and stroke specific deaths. Results The proportion of White, Black, and Hispanic patients was 66%, 23%, and 11% respectively. Total mortality at 30 days, 90 days, 1 year, and end of follow up was 22%, 27%, 33% and 54% respectively. Compared to Whites, Hispanics had higher mortality in the first 30 days post stroke (HR=1.67 95% CI=1.08-2.30), while Blacks and Whites experienced similar mortality (HR=1.02 95% CI=0.68, 1.54). Both Blacks and Hispanics had worse vascular (Black vs White HR=1.59 95% CI=0.95-2.66; Hispanic vs White HR=2.04 95% CI=1.13-3.66) and stroke-specific (Black vs White HR=1.47 95% CI=0.81-2.67; Hispanic vs White HR=2.12 95% CI=1.16-4.22) mortality (“case-fatality”) than Whites in the first 30 days. Conclusion Survivorship following subarachnoid hemorrhage varies in Veterans according to race and ethnicity. Further research into the severity of the stroke as well as risk factor control before and after stroke could help to improve our understanding of these differences.