Title:
Disparities in Time to Treatment and Outcomes in Patients Hospitalized with Ischemic Stroke
Author:
Sofia Oluwole
Date:
2019
Executive Summary:
Racial/ethnic minorities and women are disproportionately affected by stroke. Non-Hispanic (NH) Blacks and Hispanics have a greater risk of stroke compared to NH Whites. Similarly, women have a greater lifetime risk of stroke compared to men and experience more strokes annually. The vast majority of these strokes are ischemic and are typically characterized by a clot obstructing blood flow to the brain. Previous studies have shown that race/ethnic minorities and women who are hospitalized with ischemic stroke are less likely to receive many evidence-based care processes defined by the American Heart Association/American Stroke Association (AHA/ASA). NH Blacks and women receive intravenous tissue plasminogen activator (tPA), the standard thrombolytic therapy for acute ischemic stroke, less frequently than NH Whites and men. Furthermore, NH Blacks and women experience greater disability and overall lower quality of life in the months and years following an ischemic stroke. This dissertation sought to further assess racial/ethnic and sex disparities in the acute phase of stroke, with a primary focus on time to treatment with tPA (door-to-needle time), admission blood pressure, and acute outcomes in patients hospitalized with ischemic stroke. Specifically, this dissertation sought to a) determine whether there are racial/ethnic and sex disparities in achievement of the AHA/ASA target door-to-needle time <60 minutes and door-to needle time <45 minutes; b) determine whether there are racial/ethnic and sex differences in severely elevated admission blood pressure, an absolute contraindication to tPA which must be treated before tPA can be administered; and c) determine whether there are racial/ethnic and sex disparities in 3 clinical outcomes: in-hospital death, independent ambulation at discharge, and favorable modified Rankin Score at discharge (a measure of disability/functional outcome). Each aim for this dissertation was completed using data from the Florida-Puerto Rico Collaboration to Reduce Stroke Disparities (CReSD) Study.
Multivariable logistic regression modeling was performed to complete each aim. The regression models were fit using generalized estimating equations, GEE, to account for the clustering effect at the hospital level. After adjustment for multiple patient and hospital factors including stroke onset-to-hospital arrival time, stroke severity, academic status of hospital, and hospital size, women were less likely to be treated with tPA within 60 minutes and 45 minutes of hospital arrival. Likelihood of severely elevated admission blood pressure was greater in NH Blacks and women compared to NH Whites and men, respectively. NH Blacks had lower odds of in-hospital death, lower odds of independent ambulation at discharge, and lower odds of favorable modified Rankin score at discharge compared to NH Whites. Similarly, women had lower odds of in-hospital death, lower odds of independent ambulation at discharge, and lower odds of favorable modified Rankin score compared to men. Significant racial/ethnic and sex disparities were found among hospitalized ischemic stroke patients in CReSD. These findings highlight the need to improve blood pressure management and increase pre-hospital interventions to control blood pressure in incoming stroke patients, particularly in NH Blacks and women. It is also crucial to identify additional patient and process-of-care factors that contribute to slower treatment times in women and greater disability at discharge in NH Blacks and women.
