Pre-stroke emergency department utilization in the Brain Attack Surveillance in Corpus Christi (BASIC) project

Abstract

Background

Pre-stroke ED visits may represent opportunities for stroke prevention. We evaluated the prevalence and predictors of ED visits in the 90-days preceding a stroke, and the frequency of visits for “high-risk” diagnoses known to be associated with short-term stroke risk, including neurologic symptoms, atrial fibrillation, falls, and hypertensive disorders.

Methods

Within an ongoing population-based stroke surveillance study in South Texas, we identified all physician-validated ischemic and hemorrhagic strokes from April 2003 to December 2020. Linked Medicare claims data were used to identify ED visits in the 90 days before each stroke. Logistic regression was used to evaluate clinical and sociodemographic factors associated with pre-stroke ED utilization. High risk diagnoses were identified by manual review.

Results

A total of 2498 validated stroke cases were matched to Medicare claims data. Patients were 57 % female and 47 % Mexican American with a mean age of 77 (SD 11). A total of 209 patients (8.4 %) had an ED visit in the 90-days before their stroke, including 102 (4.1 %) with multiple visits. Medicaid insurance (OR 1.54, 95 %CI 1.05–2.26) and diabetes (OR 1.73, 95 %CI 1.21–2.47) were associated with greater odds of a pre-stroke ED visit. Of 430 total ED visits, 87 (20.0 %) were for a high-risk diagnosis, including 70 (16.3 %) for a neurologic diagnosis, 12 (2.8 %) for a hypertensive disorder, and 5 (1.2 %) for atrial fibrillation.

Conclusions

Pre-stroke ED visits were common, particularly among patients with diabetes or Medicaid insurance, and were often associated with high-risk primary diagnoses, potentially representing opportunities for stroke prevention or early treatment.