Do Factors Associated with Discharge Location Following Traumatic Brain Injury Differ by Race in Older Adults?
(School of Medicine (UMB) Epidemiology and Human Genetics Doctoral Student)
OBJECTIVE: The highest rates of TBI related-hospitalization and death occur in older adults. Rehabilitation following TBI improves functional outcomes. Evidence suggests that rehabilitation following TBI differs by race among younger adults, but whether this is true among older adults is unknown. We aimed to identify factors associated with discharge location by race among older adults with TBI. APPROACH: We analyzed factors associated with discharge to home or acute care for individuals ≥65 treated at the R Adams Cowley Shock Trauma Center for TBI between 1996-2012 who survived to hospital discharge. Home comprised residential facility, or home with or without services. Acute care included acute care or inpatient rehabilitation. Race was categorized as white, black, other (not white or black). Covariates included age, gender, Abbreviated Injury Scale (AIS)-Head score, Injury Severity Score (ISS), insurance (Medicare and auto vs. other (includes self-pay)), Glasgow Coma Scale (GCS) score, and intensive care unit stay (ICU) stay. Logistic regression models were built separately by race and discharge location. FINDINGS: We identified a total of 3,856 older adults without missing data on race (n=6) or discharge location (n=71) of which there were 3,176 (82%) whites, 543 blacks (14%), and 136 (4%) other race individuals. The small number and heterogeneity of other race resulted in unstable estimates and they were excluded resulting in a sample size of 3,719. A total of 1,685 (45.31%) were discharged home and 1,635 (43.96%) were discharged to acute care (Table 1). In adjusted logistic regression, factors associated with discharge to home (age, ISS, GCS, and ICU stay) and acute care (ISS, GCS, and ICU stay) vs. all other discharge locations, respectively, did not differ between blacks and whites (Table 2). Among whites, age was significantly associated with discharge to acute care (odds ratio (OR) 1.04, 95% confidence interval (CI) 1.03-1.05). Blacks with other insurance were less likely to be discharged to acute care compared to blacks with Medicare/auto insurance (OR 0.50, 95% CI 0.31-0.79). CONCLUSIONS: In this group of Medicare-eligible older adults, insurance status was associated with discharge to acute care among blacks but not whites.