ABSTRACT Objective: To analyze the association between time of arrival at a reference hospital and mortality of people with ischemic stroke. Method: Descriptive and inferential statistics were used. Modifying and confounding variables between time of arrival and mortality were observed in the multivariate analysis. The Akaike Information Criterion was used to choose the model. Statistical significance of 5% and risk correction using the Poisson Model were adopted. Results: Most participants arrived within 4.5 hours of symptom onset or wake up stroke to the referral hospital and 19.4% died. The score of the National Institute of Health Stroke Scale was a modifier. In the multivariate model stratified by scale score ≥14, arrival time >4.5h was associated with lower mortality; and age ≥60 years and having Atrial Fibrillation, to higher mortality. In the model stratified by score ≤13, previous Rankin ≥3, and presence of atrial fibrillation were predictors of mortality. Conclusion: The relationship between time of arrival and mortality up to 90 days was modified by the National Institute of Health Stroke Scale. Prior Rankin ≥3, atrial fibrillation, time to arrival ≤4.5h, and age ≥60 years contributed to higher mortality.

RESUMEN Objetivo: Analizar la asociación entre el tiempo de llegada a un hospital de referencia y la mortalidad de las personas con accidente cerebrovascular isquémico. Método: Fueron usadas estadísticas descriptivas e inferenciales. En el análisis multivariado se verificaron variables modificadoras y confusoras entre el tiempo de llegada y la mortalidad. Para la elección del modelo se utilizó el Criterio de Información de Akaike. Se adoptó la significación estadística del 5% y la corrección del riesgo mediante el Modelo de Poisson. Resultados: La mayoría de los participantes llegaron dentro de las 4,5 horas posteriores al inicio de los síntomas o wake up stroke al hospital de referencia y el 19,4% fallecieron. La puntuación de la escala National Institute of Health Stroke Scale fue un modificador. En el modelo multivariante estratificado por puntaje de escala ≥14, el tiempo de llegada >4,5 h se asoció con una menor mortalidad; y edad ≥60 años y con Fibrilación Auricular a mayor mortalidad. En el modelo estratificado por puntaje ≤13, Rankin previo ≥3 y tener fibrilación auricular fueron predictores de mortalidad. Conclusión: La relación entre tiempo de llegada y mortalidad hasta los 90 días fue modificada por puntaje de la escala National Institute of Health Stroke Scale. Rankin previo ≥3, fibrilación auricular, tiempo hasta la llegada ≤4,5 h y la edad ≥60 años contribuyeron a una mayor mortalidad.

RESUMO Objetivo: Analisar a associação entre tempo de chegada a hospital de referência e mortalidade de pessoas com Acidente Vascular Cerebral isquêmico. Método: Foram utilizadas estatísticas descritivas e inferenciais. Variáveis modificadoras e confundidoras entre tempo de chegada e mortalidade foram verificadas na análise multivariada. O Critério de Informação de Akaike foi utilizado para escolha do modelo. Adotou-se significância estatística de 5% e correção de risco pelo Modelo de Poisson. Resultados: A maioria dos participantes chegou em até 4,5 horas do início dos sintomas ou wake up stroke ao hospital de referência e 19,4% morreram. A pontuação da National Institute of Health Stroke Scale foi um modificador. No modelo multivariado estratificado por pontuação da escala ≥14, o tempo de chegada >4,5h estava associado a menor mortalidade; e idade ≥60 anos e ter Fibrilação Atrial, a maior mortalidade. No modelo estratificado por pontuação ≤13, Rankin prévio ≥3 e ter fibrilação atrial foram preditores de mortalidade. Conclusão: A relação entre tempo de chegada e mortalidade até 90 dias foi modificada pela pontuação da National Institute of Health Stroke Scale. Rankin prévio ≥3, fibrilação atrial, tempo de chegada ≤4,5h e idade ≥60 anos contribuíram para maior mortalidade.

ABSTRACT BACKGROUND: Stroke is a major cause of death and functional disability worldwide. Knowledge of the associated factors is essential for defining education, management, and healthcare strategies. OBJECTIVE: To analyze the association between arrival time at a neurology referral hospital (ATRH) and functional disability in patients with ischemic stroke 90 days after the event. DESIGN AND SETTING: Prospective cohort study conducted at a public institution of higher education in Brazil. METHODS: This study included 241 people aged ≥ 18 years who presented ischemic stroke. The exclusion criteria were death, inability to communicate without companions who could answer the research questions, and > 10 days since ictus. Disability was assessed using the Rankin score (mR). Variables for which associations showed a P value ≤ 0.20 in bivariate analysis were tested as modifiers between ATRH and disability. Significant interaction terms were used for multivariate analysis. Multivariate logistic regression analysis was performed with all variables, arriving at the complete model and adjusted beta measures. The confounding variables were included in the robust logistic regression model, and Akaike’s Information Criterion was adopted to choose the final model. The Poisson model assumes a statistical significance of 5% and risk correction. RESULTS: Most participants (56.0%) arrived at the hospital within 4.5 hours of symptom onset, and 51.7% presented with mRs of 3 to 5 after 90 days of ictus. In the multivariate model, ATRH ≥ 4.5 hours and females were associated with more significant disability. CONCLUSIONS: Arrival at the referral hospital 4.5 hours after the onset of symptoms or wake-up stroke was an independent predictor of a high degree of functional disability.