Abstract P472: Matching Young Adult Strokes in a Claims Database to the Clinical Get With the Guidelines-Stroke Registry

Abstract

Background: Due to its lower incidence, strokes in young adults, studies have difficulty answering why there are more women than men less than 35 with incident strokes. The goal of this study is to link claims and registry data to create a new longitudinal and clinical cohort of strokes in young adults. Methods: Colorado mandates reporting of commercial and public insurance (Medicaid and Medicare) claims to the All Payer Claims Database (APCD). We identified all first-time stroke admissions using ICD-9 /10 codes between 2012 to 2018 in young adults (18-55). Get with the Guidelines (GWTG) Stroke registry has stroke cases from 41 hospitals in Colorado since 2010. Matching was based on exact birthdate, sex and +/- 1day in admission/discharge dates. Confirmation of match was based on comparing the admitting hospital between the two databases. Cardiovascular(CV) and non-CV risk factors occurring from enrollment until 1 day prior to stroke date in the APCD were included. Student’s T-test, and Z-tests were used to test differences between men and women. Results: We identified 3032 stroke cases in the APCD. Of these, 1455 (48.0%) matched to an admission in the GWTG-Stroke registry. Of these, 1303 (89.6%) also matched based on the admitting hospital (Table). The final cohort contained 51% women, 11.3% <35, 35.2% between 35-45, and 63.5% between 45-55. The most common CV risk factors were hypertension, hyperlipidemia, and tobacco use in men, and hypertension, tobacco use, and obesity in women. The most common non-CV risk factors were migraines, renal failure, thrombophilia in men and migraines, malignancy, and thrombophilia in women. The prevalence of cardiovascular risk factors was higher in men, while the prevalence of non-cardiovascular risk factors was higher in women. There was no difference in the severity of strokes between men and women. Conclusions: Clinical registries such as GWTG-Stroke can be used to supplement longitudinal administrative cohorts of young adults with stroke to get richer data on stroke etiology and severity.