RESUMO
BACKGROUND: Stroke patients and family members should receive stroke education including recognition of stroke symptoms and prompt activation of emergency medical services (EMS). The impact of this education is unclear. We aimed to measure the associations between EMS use and timing of hospital arrival and first-ever and recurrent strokes as a proxy for stroke education. METHODS: The study analyzed data from validated strokes identified by the Brain Attack Surveillance in Corpus Christi (BASIC) project between 1/1/2000-1/1/2020. We analyzed 5,617 first-ever strokes, 259 instances of recurrent stroke within 1 year of the first (early recurrence), and 451 recurrent strokes over 1 year from the first (late recurrence). Following imputation, associations of both EMS arrival (available starting late 2011) and early arrival (< 3 hours) with first-ever versus recurrent stroke (early and late) were assessed with logistic models, accounting for the clustering of multiple strokes per participant with generalized estimating equations. Full model covariates included stroke type, initial stroke severity, marital status, race/ethnicity, gender, age, insurance, education, and EMS use (early arrival model only). RESULTS: Compared to first-ever stroke, there were significantly higher unadjusted odds of arrival by EMS for the late recurrence group (late recurrence OR = 1.54, 95% CI = 1.18-1.99; early arrival OR = 1.24, 95% CI = 0.87-1.76). The association for late recurrence remained significant after adjustment (aOR = 1.46, 95% CI = 1.09-1.95). The pre-2010 unadjusted odds of early arrival were non-significant for both early and late recurrence groups (late recurrence OR = 1.05, CI = 0.70-1.56; early recurrence OR = 0.85, CI = 0.54-1.33), while late recurrence was associated with early arrival after 2010 (OR = 1.32, 95% CI = 1.03-1.69). After full adjustment, it was no longer significant (aOR = 1.25, 95% CI = 0.96-1.62). Higher initial stroke severity, married status, and EMS use were associated with higher odds of early arrival, while African Americans (AAs) had lower odds than non-Hispanic Whites (NHWs). However, AAs did have higher odds of EMS use relative to NHWs. Those who were married and living together had borderline significant lower odds of EMS use compared to those who were not. CONCLUSIONS: Our study examines the association of repeat stroke on early arrival and EMS use as a surrogate for adequate stroke education. Recurrence at least one year after the first stroke was associated with higher EMS usage, but there was not enough evidence to establish a relationship with early arrival after accounting for EMS usage and possible confounders. By examining subsets, we can identify groups that would benefit from targeted education. For example, younger, non-AA patients with smaller strokes would benefit from more education on EMS use and African American patients would benefit from education related to faster recognition or urgency of presentation.