Effects of Race and Poverty on the Process and Outcome of Inpatient Rehabilitation Services Among Stroke Patients
Background and Purpose— The greater mortality and residual physical impairments among black stroke patients may be attributable to differential utilization of rehabilitation services. This report examines, within an equal-access healthcare system, racial differences in time to initiation of stroke rehabilitation services and in the trajectory of physical function recovery.
Methods— This study was a secondary analysis of data from an inception cohort of 1073 stroke patients hospitalized between April 1995 and March 1997 and followed up for up to 1 year. Inpatient data came from medical record reviews; follow-up data came from telephone interviews at 1, 6, and 12 months after stroke. The study included consecutive acute ischemic or intracerebral hemorrhagic stroke patients from 9 VA medical centers. The main outcome measures were time to initiation of inpatient rehabilitation services and ability to perform activities of daily living.
Results— There were no racial differences in receipt of inpatient rehabilitation services (blacks, 76%; whites, 70%) or in the proportion of patients referred within 3 days of admission (blacks, 43.5%; whites, 42.0%). Among patients who experienced delay in initiation of rehabilitation, only low-income blacks experienced worse functional recovery over 12 months.
Conclusions— Low-income black stroke patients who experience delay in initiation of inpatient rehabilitation have a worse trajectory of functional recovery in the first year after stroke. Poverty-associated factors in the postdischarge setting may explain this phenomenon.
Blacks experience not only a higher stroke incidence than whites but also higher stroke mortality and, among stroke survivors, worse residual physical impairment.1,2 Recent evidence indicates that greater stroke severity cannot fully account for these racial differences in stroke outcomes.3 However, differential access to timely acute and postacute stroke services may play a role. Stroke patients who receive prompt care in stroke rehabilitation units, for example, experience lower mortality and better recovery of physical function in the immediate term and for at least 10 years after stroke onset.4 Because few studies have examined racial differences in utilization of stroke rehabilitation services,5–7 it is unclear whether differential utilization of services accounts for observed racial differences in stroke recovery. This report presents evidence regarding racial differences in the process and outcomes of stroke rehabilitation services within an equal-access healthcare system.
This is a secondary analysis of data from the VA Acute Stroke (VAST) study, a 9-site, nationwide prospective cohort study of 1073 patients with acute stroke who were hospitalized within the Veterans Health Administration between April 1, 1995, and March 31, 1997, for acute ischemic (ICD-9-CM 434 or 436) or intracerebral hemorrhagic (ICD-9-CM 431) stroke.8
Patients were followed from admission through 1 year after stroke onset. Data on acute care practices were obtained by medical record review. Data on health outcomes were obtained by both medical record review (for status at discharge) and telephone interviews (for status after discharge). To the extent possible, information on patients who were unable to communicate (eg, those who were demented, were comatose, or experienced significant receptive or expressive aphasia) was obtained from proxy respondents; a qualified proxy was someone who spent the majority of his or her day with the patient. The study protocols were reviewed and approved by the Institutional Review Board at each participating site.
For this report, the original cohort was restricted to the 738 patients (69% of the 1073 patients) who were either black or white and were referred to inpatient rehabilitation (inpatient cohort). Blacks were as likely as whites to be referred to rehabilitation after adjustment for stroke severity and cognitive impairment (adjusted odds ratio [OR], 1.28; P=0.13).
To assess the trajectory of physical function recovery, the inpatient cohort was further restricted to the 598 patients (81% of the 738 patients) who had at least 2 assessments of physical function: baseline (ie, discharge) and ≥1 follow-up periods (follow-up cohort). Attrition occurred for a number of reasons: we were unable to contact the patient before the interview date had passed; the patient was unable to communicate and had no proxy informant; or the patient died. Attrition was not significantly associated with the patient’s race.
For analysis of the inpatient rehabilitation process, the primary outcome was elapsed time in days between hospital admission and initial contact by rehabilitation service personnel (eg, physician, physical therapist, occupational therapist, speech therapist, kinesiotherapist, or recreational therapist).
The primary outcome for analysis of physical function recovery was performance of activities of daily living (ADL) during the initial year after stroke onset.9,10 Performance of ADL was measured by the Rankin Scale score at the time of discharge and by the Barthel Index after discharge. The Rankin score was standardized to the Barthel percent metric (ie, recoded to a scale from 0 to 100 with 0=worst and 100=best) to allow assessment of relative change from discharge to the last point of follow-up.
Primary Independent Variable and Covariates
Self-reported racial group dichotomized as black or white was the primary independent variable. When self-reported race was missing, race as recorded in the medical record was used. Covariates included age, sex, marital status, living situation at discharge (alone, with others, or nursing home or institution), income level (with low-income status being US$10 000 or less in annual household income), stroke type11 and severity,3 cognitive function,12 and whether the patient’s care was directed primarily by a neurologist.13 We also included a variable indicating the time in days from stroke onset to admission because time to referral to rehabilitation may be affected by how soon after stroke onset the patient presents for care.
Ordinary least-squares regression analysis was used to test the baseline association between racial status and length of time to initial contact by rehabilitation service in the hospital, controlling for relevant covariates. Mixed-model analysis was used to assess the longitudinal effects of race, socioeconomic status, and delay in initiation of rehabilitation on functional recovery.14,15
Characteristics of patients in the inpatient and follow-up cohorts are shown in Table 1.
Race and Time to Initiation of Rehabilitation
The median time to initiation of rehabilitation was 3 days for patients regardless of race. The average time was approximately one-half day longer for blacks than whites (3.8 versus 4.4 days, respectively; P<0.05), a result of the racial difference in the proportion of patients referred to rehabilitation ≥10 days after admission (8% of black patients [n=17] and 4% of white patients [n=18]). There was no racial difference in total number of in-hospital contacts with rehabilitation personnel.
Delay in Rehabilitation and Functional Recovery
On average, black patients recovered physical function during the first year after stroke at a significantly slower rate than whites (Table 2, model 1). This effect was associated with time to initiation of inpatient rehabilitation so that delay had a greater negative affect on poststroke recovery among blacks than whites (Table 2, model 2). Examination of improvement in physical function, defined as a change of 25 percentage points, among patients who began rehabilitation within 3 days of admission showed that 26% of whites and blacks experienced substantial improvement (P=0.97). However, among those who began rehabilitation after 3 days, 35% of whites compared with 16% of blacks experienced substantial improvement (P=0.007). These patterns held after the exclusion of patients who had initiated rehabilitation ≥10 days after admission.
Modifying Effect of Low Income on Use and Outcome of Inpatient Rehabilitation
Low-income black patients had a greater likelihood of beginning inpatient rehabilitation beyond the recommended 3 days (47.0% versus 27.3% for low- versus higher-income black patients, respectively; OR, 2.36; P<0.04). However, no such association existed among white patients (41.8% versus 40.5%; OR, 0.95; P=0.80). Functional recovery was impaired only for low-income blacks experiencing delay in initiation of inpatient rehabilitation (Table 3). That is, among patients experiencing delay in initiation of rehabilitation, substantial improvement in physical function occurred for only 9.1% of low-income blacks compared with 35.7% of higher-income blacks (P=0.02). Again, these patterns held after the exclusion of patients who had initiated rehabilitation ≥10 days after admission.
This report is the first to document the existence of a significant racial difference in the association between the process of rehabilitative care and subsequent outcomes among stroke patients. Low-income black stroke patients—those who were delayed in the initiation of stroke rehabilitation—experienced relatively worse poststroke physical function recovery after adjustment for important clinical factors such as stroke severity and type. There was no racial difference in either the proportion of stroke patients who were referred to inpatient rehabilitation or the intensity of rehabilitation.
It is unlikely that delay in initiation of inpatient rehabilitation is directly responsible for the worse functional recovery among poor black stroke patients after discharge. If that were the case, relatively worse recovery should have occurred among all stroke patients or, at least, among poor whites who experienced delay. This did not occur. The trajectory of physical function recovery among the black patients who were higher income, although relatively few in number, and whites with low income was similar to that of whites with higher income. Moreover, the average level of function was similar among all race-income groups at discharge. Hence, it appears that circumstances in the posthospital environment differentially affect recovery from stroke for poor blacks vis-à-vis poor whites and patients of higher incomes.
An explanation for this phenomenon may be found in the fact that patients were clinically managed in the VA healthcare system, which, in principle, provides equal access to care for its eligible patient population. This suggests that ability to pay for outpatient rehabilitative care is probably not a major factor. The pace of recovery from stroke is more likely inhibited by other aspects of poverty, such as the absence of supportive social resources (eg, transportation to outpatient rehabilitation services or access to supplemental in-home rehabilitation services) that may help sustain or enhance the benefits of therapy over time. This hypothesis derives from work suggesting that early supported discharge with an emphasis on rehabilitation in the home yields improved functional recovery after discharge.16 If the hypothesis holds, race must be associated with an accentuation of these poverty-related factors that impede stroke recovery so that being black and having low income combine to yield a more negative health outcome than might be associated with low income alone.
Further research is needed to determine specifically the rehabilitation services and processes experienced by patients of different races or ethnicities and incomes. It is important to explore the postdischarge context to identify those factors associated with further improvement (or impediment) in achieved therapeutic benefits at discharge.
This work was supported in part by a grant from the VA Health Services Research and Development Service (SDR 93-003) and a grant from the VA Cooperative Studies/Epidemiologic Research and Information Center Programs (VA CSP/ERIC 602).
Dr Horner is now with the National Institute for Neurological Disorders and Stroke.
The views expressed in this report are those of the authors and do not necessarily reflect the views of either the Department of Veterans Affairs or the United States Government.
- Received September 24, 2002.
- Accepted October 10, 2002.
Horner RD, Matchar DB, Devine G, Feussner JR. Racial variations in ischemic stroke related physical and functional impairments. Stroke. 1991; 22: 1497–1501.
Broderick J, Brott T, Kothari R, Miller R, Khoury J, Pancioli A, Gebel J, Mills D, Minneci L, Shukla R, for the Greater Cincinnati/Northern Kentucky Stroke Study. Preliminary first-ever and total incidence rates of stroke among blacks. Stroke. 1998; 29: 415–421.
Jones MR, Horner RD, Edwards LJ, Hoff J, Armstrong B, Smith-Hammond CA, Matchar DB, Oddone EZ. Racial variation in initial stroke severity. Stroke. 2000; 31: 563–567.
Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Stroke unit treatment: 10-year follow-up. Stroke. 1999; 30: 1524–1527.
Kuhlemeier KV, Steins SA. Rehabilitation after stroke: gender and racial disparities? Arch Phys Med Rehabil. 1991; 72: 840–841.
McElligott JM, Dolezal J, Hamilton J. Stroke, racial variations, and effective inpatient rehabilitation in a rural setting. Arch Phys Med Rehabil. 1993; 74: 1278.
Horner RD, Hoenig H, Sloane R, Rubenstein LV, Kahn KL. Racial differences in the utilization of inpatient rehabilitation services among elderly stroke patients. Stroke. 1997; 28: 19–25.
Bosworth HB, Horner RD, Edwards LJ, Matchar DB, for the VAST Study Team. Depression and other determinants of stroke patient’s utility for current health state: evidence from the VA Acute Stroke (VAST) Study. Stroke. 2000; 31: 2603–2609.
Wolfe CDA, Taub NA, Woodrow EJ, Burney PGJ. Assessment of scales of disability and handicap for stroke patients. Stroke. 1991; 22: 1242–1244.
Goldstein LB, Jones MR, Matchar DB, Edwards LJ, Hoff J, Chilukuri V, Armstrong SB, Horner RD. Improving the reliability of stroke subgroup classification using the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria. Stroke. 2001; 32: 1091–1095.
Diggle PJ, Liang K-L, Zeger SL. Analysis of Longitudinal Data. Oxford, England: Oxford University Press; 1994.
Indredavik B, Fjaertoft H, Ekeberg G, Loge AD, Morch B. Benefit of an extended stroke unit service with early supported discharge: a randomized, controlled trial. Stroke. 2000; 31: 2989–2994.