Abstract WP202: Race/ethnicity Matters: Differences in Stroke Inpatient Medical Rehabilitation Outcomes for Communities of Color

Abstract

Introduction: Stroke is a leading cause of disability and death in the United States. Racial/ethnic people with stroke demonstrate lower functional independence measure (FIM) scores before and after inpatient medical rehabilitation compared to non-Hispanic Whites. Racial/ethnic differences in FIM scores have implications for discharge. Method: This is a cross-sectional and retrospective analysis of data extracted and pooled across years 2002-2018. Inclusion criteria were: 1) stroke as a primary diagnosis and 2) non-Hispanic White (NHW), Hispanic, non-Hispanic Black (NHB), and non-Hispanic Asian (NHA) patients. Per systematic review evidence, ICD codes 430-438/I60-I69 were used to identify stroke diagnoses. Dependent variables were: length of stay (LOS), admission FIM Cognitive and Motor Total, FIM efficiency, FIM change per day, discharge FIM Cognitive and Motor Total, and discharge to home. NHWs were used as the reference group. Results: There were 3,911 patients across the study period. Age, gender, and admission year were used as covariates. There were racial/ethnic differences in LOS, F (3, 3853) = 2.658, p = .047, partial η2 = .002, with NHAs having longer LOS. Racial/ethnic differences in admission FIM scores were found, F (6, 7750) = 6.727, p < .001, Wilks' λ = .989, partial η2 = .006, with NHAs and Hispanics showing lower cognitive and motor scores. There were no racial/ethnic differences in cognitive or motor efficiency. Racial/ethnic differences in FIM change per day emerged, F (3, 3854) = 4240, p = .005, partial η2 = .003, with Hispanics and NHBs having lower change per day. Racial/ethnic differences in discharge FIM scores were noted, F (6, 7704) = 4.865, p < .001, Wilks' λ = .992, partial η2 = .004, with NHAs and Hispanics having lower total cognitive scores but only Hispanics had lower motor scores. Lastly, Hispanics had higher odds of being discharged home compared to other racial/ethnic groups ( OR = 1.246, 95% CI = 1.046-1.485, p = .014). Discussion: Controlling for covariates, there were significant racial/ethnic differences in stroke inpatient medical rehabilitation outcomes, particularly for Hispanics. Factors involved in creating/maintaining post-stroke rehabilitation disparities for Hispanics (and others) are discussed.