Timing of Initiation of Rehabilitation After Stroke

Article Outline

• Abstract
• Methods
  • PSROP Facilities
  • PSROP Patient Selection Criteria
  • PSROP Study Variables
  • PSROP Data Collection
    • Point-of-care data
    • Disease-specific severity-of-illness data (signs and symptoms)
    • Additional patient, process, and outcome data
    • Patient sample
  • Processes and Interventions
    • Timing of rehabilitation after stroke
  • Outcomes
    • FIM subscores
  • Statistical Methods
• Results
  • Descriptive Statistics
  • Days From Stroke Symptom Onset to Rehabilitation Admission
• Discussion
• Conclusions
• Acknowledgments
• References
• Copyright

Abstract 

Maulden SA, Gassaway J, Horn SD, Smout RJ, DeJong G. Timing of initiation of rehabilitation after stroke.

Objective

To study associations between days from stroke symptom onset to rehabilitation admission and rehabilitation outcomes, controlling for a variety of confounding variables.

Design

Observational cohort study of 200 consecutive poststroke rehabilitation patients in each of 6 inpatient rehabilitation facilities.

Setting

Six U.S. inpatient rehabilitation hospitals.

Participants

Patients (N=969) with moderate or severe strokes who had days from stroke symptom onset to rehabilitation admission recorded in their medical records.

Interventions

Not applicable.

Main Outcome Measures

Discharge total FIM, discharge motor FIM, discharge activities of daily living (ADL) FIM, and discharge mobility FIM scores, as well as rehabilitation length of stay (LOS).

Results

Fewer days from stroke symptom onset to rehabilitation admission was associated significantly with better functional outcomes: higher total, motor, mobility, and ADL discharge FIM scores, controlling for confounding variables. For severely impaired patients with stroke in case-mix groups (CMGs) 108–114, the relation was strongest, with F statistics greater than 24.1 for each functional outcome. For patients with moderately severe stroke in CMGs 104–107, fewer days from stroke symptom onset to rehabilitation admission was associated significantly with shorter rehabilitation LOS.

Conclusions

Fewer days from stroke symptom onset to rehabilitation admission is associated with better functional outcomes at discharge and shorter LOS.

Key Words:  Cerebrovascular accident , Clinical practice variations , Rehabilitation , Stroke , Treatment outcomes

STROKE IS COMMON and is a leading cause of disability. Mortality rates from stroke have been declining,¹, ² resulting in more people living with residual disability.³, ⁴, ⁵ It is well documented in the literature that rehabilitation plays an important role in functional recovery of stroke survivors, providing quantifiable benefits beyond the natural recovery that occurs without any targeted therapy.³, ⁶, ⁷, ⁸

Unfortunately, existing studies of stroke rehabilitation outcomes are neither clear nor consistent,³ and debate continues over effectiveness of stroke rehabilitation programs.⁹, ¹⁰, ¹¹ Many rehabilitation providers argue that the acute care hospital payment system encourages acute care providers to discharge patients to rehabilitation before they are medically stable. However, early mobilization and more aggressive rehabilitation are components of stroke unit care thought to contribute to improved outcomes.¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷

Optimal timing of rehabilitation after stroke remains controversial. It is an important question to answer because it is modifiable, unlike other predictors of functional recovery after stroke (eg, age, premorbid function). Several studies provide evidence for the benefit of early rehabilitation compared with later intervention in patients with stroke.³, ⁶, ¹², ¹⁸, ¹⁹, ²⁰ However, interpretation of these studies is limited by heterogeneous definitions, study designs, and methods. For example, early rehabilitation may mean starting rehabilitation anywhere from 3 to 30 days after stroke.⁶ Johnston and Keister²¹ found that the positive correlation between early rehabilitation and improved functional outcomes disappeared when key patient characteristics, such as functional status on admission, were controlled for. The problem is compounded further by variation in type and severity of strokes, variation in rehabilitation procedures in different settings, and incomplete, vague, or ambiguous documentation of what constitutes each type of therapy.²², ²³

The question of how soon to start rehabilitation after stroke is relevant also in light of recent theories regarding neural responses to injury. Surrounding a cerebral infarct is a zone of cells that potentially are salvageable but are more vulnerable to injury poststroke (the ischemic penumbra). These cells may or may not recover, depending on a number of physiologic factors.²⁴ Based on animal studies and human imaging studies, there is evidence for neural reorganization thought to be dependent on some form of synaptic plasticity.²⁴, ²⁵, ²⁶ There may be increased potential for cortical plasticity in the 7 to 18 days after injury (animal literature),²⁷, ²⁸ suggesting a critical period to obtain the best recovery after stroke.²⁴

On the other hand, rehabilitation in the very early stages after stroke theoretically may harm vulnerable cells via oxidative and/or metabolic stress in concert with reperfusion injury. Many patients show extension of the infarct area by imaging within the first few days after stroke.²⁹ Even so, increases in infarct volume have not always been shown to correlate predictably with functional outcomes. Two studies in experimentally lesioned animals report a paradoxic exacerbation of brain damage with concomitant enhancement of recovery of function after early rehabilitative interventions.³⁰, ³¹

A detailed literature review substantiating the need to examine rehabilitation processes to improve outcomes for specific types of patients is presented elsewhere.³² The purpose of the analyses presented here was to study the associations between days from onset of stroke symptoms to rehabilitation admission and rehabilitation outcomes, controlling for a variety of confounding variables based on data from the Post-Stroke Rehabilitation Outcomes Project (PSROP).

Back to Article Outline

Methods 

The clinical practice improvement (CPI) methodology was used in the PSROP because it captures in-depth, comprehensive information about patient characteristics (including clinical signs and symptoms), rehabilitation processes of care, and rehabilitation outcomes needed to characterize the process of care and ascertain the contribution of individual rehabilitation processes to outcomes.³³ An in-depth description of the study’s methods, including issues of validity and reliability, can be found in Gassaway et al³³ in this supplement. In this article, we provide only a summary of the study’s extensive methods to help interpret the findings reported here.

PSROP Facilities 

Six U.S. inpatient rehabilitation facilities (IRFs) participated in the PSROP and were selected based on geographic location and their willingness to participate; they are not a probability sample of IRFs in the United States. Each site contributed detailed data on about 200 consecutive poststroke patients, for a total of 1161 patients. Patients with stroke from these 6 IRFs constitute a convenience sample.

PSROP Patient Selection Criteria 

Each participating IRF obtained institutional review board approval for this observational study and enrolled consecutively admitted patients who met the following inclusion criteria:

PSROP Study Variables 

Three types of study data—(1) patient characteristics (eg, age, sex, race, payer, type of stroke, side of stroke, admission severity of illness, functional status measures), (2) process variables (eg, treatments, interventions), and (3) outcome variables (eg, discharge functional status, discharge severity of illness, discharge destination)—were obtained from multiple sources either at the point of care or from postdischarge chart review in the IRF.

PSROP Data Collection 

The study’s physicians, nurses, psychologists, social workers, and physical, occupational, recreational, and speech language pathology therapists each created a form to include the level of intervention intensity they thought was needed to capture a complete and accurate picture of the contribution made by that discipline to rehabilitation care (beyond what was already contained in traditional medical record documentation). Each rehabilitation discipline developed its own content and decided on the frequency with which its form should be completed. One therapy intervention documentation form was completed for each patient treatment session. Examples of forms used by physical therapists, occupational therapists, and speech-language pathologists are given elsewhere.³³

The Comprehensive Severity Index (CSI) is the study’s principal severity adjuster. The CSI is an exhaustive, disease-specific severity system that provides a consistent method of defining severity of illness levels using over 2200 individual patient historical factors, physiologic parameters, laboratory results, and physical findings. In the CSI, severity is defined as the physiologic and psychosocial complexity presented to medical personnel due to the extent and interactions of a patient’s disease(s).³³ The CSI was measured separately for admission to rehabilitation (first 24h), discharge from rehabilitation (discharge day), and maximum. (Maximum CSI covers the full rehabilitation stay, including admission and discharge period.) Often a patient is the sickest on admission, and thus the admission and maximum CSI scores will be the same. However, when iatrogenic conditions develop, the maximum CSI score becomes larger (more severe) than the admission score. Discharge CSI scores typically are the lowest, because patients have improved and stabilized throughout the rehabilitation stay.

In addition to disease-specific severity-of-illness information, the CSI software system allows for the collection of additional study-specific patient, process, and outcome data elements, identified and defined by the project clinical team into an instrument called the auxiliary data module (ADM). Most variables contain date and time fields so that they can be associated with other variables in time sequence. The PSROP ADM contained over 200 variables, many of which have numerous data entries. For example, some data related to vital signs, weight, and pain, among others, were collected for each day of the rehabilitation stay, so these single variables have as many entries as the length of stay (LOS). Outcome variables in the ADM included discharge FIM instrument scores, LOS, death, and discharge destination. Patient and process variables included living situation, ambulation, activities of daily living (ADLs), and employment before stroke; age; sex; payer source; admission FIM score; case-mix group (CMG) (used for Medicare payment purposes); rehabilitation LOS; and acute admission LOS.³³

The functional performance for each study patient on admission to and discharge from inpatient rehabilitation was obtained via retrospective chart review using the study site’s reporting of FIM scores. We assumed all clinicians providing FIM data within IRFs, as part of standard practice, were FIM credentialed; no additional documentation of FIM elements was performed for project purposes. We assigned each study patient to a CMG following stroke CMG definition rules based on motor FIM score, cognitive FIM score, and patient age.³³, ³⁵, ³⁶

From the 1161 PSROP patients, we excluded patients who died and outlier patients admitted to rehabilitation more than 200 days after stroke symptom onset. There were 1031 patients remaining for study from the 6 U.S. rehabilitation sites.

Preliminary analyses showed that admission CMG was a stronger predictor of functional outcome than admission FIM score alone. Therefore, we chose to analyze the 1031 patients within CMG groupings while controlling for admission FIM scores as independent variables in regression analyses. To maintain sample sizes large enough to detect small effect sizes, CMGs were combined into moderate (CMGs 104–107, n=486 patients) and severe (CMGs 108–114, n=483 patients) patient groups. We focused analyses on the 969 patients with moderate and severe stroke, because there were too few patients with mild stroke in CMGs 101 to 103 (n=62).

Processes and Interventions 

To determine the relation between days from onset of stroke symptoms to rehabilitation admission and functional outcomes, we defined the following variable: number of days from stroke onset (defined as first symptom onset) to admission to a dedicated rehabilitation unit. Other time intervals were defined as control variables. For example, days from symptom onset to acute care hospital admission was included in the regression models. In some cases, a patient experienced the onset of stroke symptoms during an acute care hospital stay (eg, for a cardiac procedure), yielding a negative number of days from symptom onset to acute care admission. In these cases, we counted the number of days from symptom onset to acute care hospital admission as zero. Similarly, if a patient was receiving therapy in the hospital before onset of stroke symptoms, the number of days between symptom onset and initiation of therapy was counted as zero.

Outcomes 

The FIM provides a useful global measure of functional independence. However, we found that patients with identical FIM scores actually differed markedly in distinct aspects of functioning that are targeted by particular therapies. Therefore, we used FIM subscores indicating impairment in specific domains—such as ADLs, motor function, or mobility—as outcome measures. To obtain the FIM ADL subscore, 6 individual FIM component scores for activities (bathing, eating, grooming, dressing upper body, dressing lower body, toileting) were added. To obtain the FIM mobility subscore, 3 individual FIM component scores for transfers (toilet; bed, chair, and wheelchair; tub and shower transfers) and 2 individual FIM component scores for locomotion (stairs, walk and wheelchair locomotion) were added together. The FIM motor subscore was the sum of the FIM mobility subscore, the FIM ADL subscore, and the FIM component scores for bladder and bowel control.

Statistical Methods 

Ordinary least squares regression was used to examine associations between days from symptom onset to rehabilitation admission with each resident functional outcome at discharge, controlling for age, sex, race, ambulation independence before admission, ADL independence before admission, rehabilitation LOS, side of brain affected by stroke, days from symptom onset to acute care admission of 4 or more, acute care LOS of 20 days or more, and admission to IRF after implementation of a new prospective payment system.

All potential predictor variables were checked for multicollinearity; no correlations were greater than .50. Stepwise R² selection procedure allowed independent variables to enter and leave each model. The importance of each predictor was determined by its F value. We created the most parsimonious model for each outcome by allowing only significant (P<.05) variables to remain in the model. Analyses were performed within subpopulations (CMG groups) of the sample. All analyses were performed with SAS statistical software.^a

Back to Article Outline

Results 

Descriptive Statistics 

Characteristics of the 969-patient sample (age, sex, race, payer, stroke characteristics, FIM scores and subscores, severity-of-illness scores, number of days from stroke onset to rehabilitation admission) are shown in table 1. In addition, rehabilitation LOS, discharge destination, and other functional and severity outcome data are presented for the 969-patient sample.

Table 1. Study Sample Patient, Process, and Outcome Characteristics

PSROP Variables	N⁎	Value
Patient Characteristics
Mean age (y)	969	66.6±14.4
Sex (% male)	969	52.0
Race (%)	969
White		54.9
Black		25.7
Other, including Hispanic		19.4
Payer (%)	969
Medicare		58.4
Medicaid		10.6
Commercial		27.7
Self-pay		2.6
Unknown/missing		0.7
Type of stroke (%)	969
Hemorrhagic		24.5
Ischemic		75.5
Side of stroke (%)	969
Right		44.3
Left		43.1
Bilateral		9.9
Unknown		2.7
Mean admission total FIM score	969	57.8±18.3
Mean admission motor FIM score	969	37.4±12.3
Mean admission cognitive FIM score	969	20.5±8.3
CMG (%)	969
Moderate (104–107)		50.2
Severe (108–114)		49.8
Mean rehab admission CSI score	969	21.6±14.0
Mean no. of days from stroke onset to rehab admission	958	13.8±18.7
Process variables
Mean LOS	969	19.6±10.1
Outcome variables
Mean maximum CSI score	969	32.3±20.9
Increase in severity (maximum − admission CSI scores)	969	10.7±11.2
Mean discharge CSI score	969	10.5±11.9
Mean gross medical improvement (maximum − discharge CSI scores)	969	21.7±15.0
Mean net medical improvement (admission − discharge CSI scores)	969	11.1±9.9
Mean discharge total FIM score	955	85.2±21.9
Mean increase in total FIM score (discharge − admission)	955	27.3±14.1
Mean discharge motor FIM score	958	60.2±16.8
Mean increase in motor FIM score (discharge − admission)	958	22.8±11.9
Mean discharge cognitive FIM score	964	24.9±7.5
Mean increase in cognitive FIM score (discharge − admission)	964	4.4±4.2
Discharge destination (%)	969
Community discharge including home		80.1
Home only		76.8
Inpatient institutional discharge		19.9

NOTE. Values are mean ± standard deviation or as otherwise indicated.

Abbreviation: rehab, rehabilitation.

Days From Stroke Symptom Onset to Rehabilitation Admission 

We hypothesized that delayed time to rehabilitation admission would be associated with lower functional outcomes, as measured by FIM scores (or subscores) at discharge from rehabilitation. First, we examined the association of days from symptom onset to rehabilitation admission alone as a predictor of discharge outcomes in simple ordinary least squares regression analyses. These findings are presented in table 2. In both groups (CMGs 104–107, 108–114), more days from stroke symptom onset to rehabilitation admission was associated significantly with lower discharge total FIM, discharge motor FIM, discharge mobility FIM, and discharge ADL FIM scores. In moderately impaired patients (CMGs 104–107), days from stroke symptom onset to rehabilitation admission had a P value of .042 or less for all FIM subscores, and in severely impaired patients (CMGs 108–114), days from stroke symptom onset to rehabilitation admission had a P value of .008 or less for all FIM subscores. In the moderately impaired group, more days from stroke onset to rehabilitation admission also was associated significantly with longer rehabilitation LOS (P<.001). By comparison, the severely impaired group did not show an association between days from symptom onset to rehabilitation admission with LOS (P=.39).

Table 2. Days From Stoke Onset to Rehabilitation Admission: Associations With Discharge FIM Scores and Rehabilitation LOSs^⁎

Outcome Variable	Days From Stroke Onset to Rehabilitation Admission as a Single Independent Variable
	Coefficient†	P	R2‡
Moderate Stroke (CMGs 104–107)
Discharge total FIM score	−.17	<.001	.025
Discharge motor FIM score	−.12	.001	.023
Discharge mobility FIM score	−.07	<.001	.038
Discharge ADL FIM score	−.04	.042	.009
Rehab LOS	.09	<.001	.024
Severe Stroke (CMGs 108–114)
Discharge total FIM score	−.11	.008	.015
Discharge motor FIM score	−.10	.003	.019
Discharge mobility FIM score	−.04	<.001	.024
Discharge ADL FIM score	−.05	.001	.022
Rehab LOS	.02	.394	.002

Next we performed multiple regression analyses, allowing many additional patient and treatment characteristics to enter the models for the various discharge outcome variables. The findings are presented in Table 3, Table 4. In table 3 we examine patients with moderate stroke in CMGs 104 to 107. When many additional patient (including maximum CSI) and treatment variables were allowed to enter the model, we found that days from stroke onset to rehabilitation admission remained statistically significant and in the same directions as in the single-variable regression analyses. In table 4, we examine patients with severe stroke in CMGs 108 to 114. Again, the findings from table 2 remain when many additional patient (including maximum CSI) and treatment variables were allowed to enter the models. Patient and treatment variables that entered significantly in each of these equations were in the expected directions related to the outcomes.

Table 3. Regression Results for Outcomes of Discharge FIM Scores and Rehabilitation LOS for Patients With Moderate Stroke in CMGs 104 to 107 (n=475)

Independent Variables	Dependent Variables
	Discharge Total FIM Score			Discharge Motor FIM Score			Discharge Mobility FIM Score			Rehabilitation LOS (n=480)
	Coeff	F	P	Coeff	F	P	Coeff	F	P	Coeff	F	P
Days from stroke onset to rehab admission	−0.11	8.2	.004	−0.12	11.7	<.001	−0.08	24.8	<0.001	0.05	4.3	.038
Partial R2		.010			.013			.030			.01
Age	−0.13	14.4	<.001	−0.12	16.4	<.001	−0.05	15.1	<.001
Female							−0.69	3.9	.050
Admission motor FIM score	0.56	44.7	<.001	0.55	55.8	<.001	0.23	44.7	<.001	−0.33	36.6	<.001
Admission cognitive FIM score	0.70	105.4	<.001				−0.07	6.5	.011	−0.09	4.2	.042
Maximum CSI score	−0.12	11.7	<.001	−0.07	5.2	.023				0.09	16.8	<.001
Employed PTA	3.91	12.2	<.001	2.57	7.1	.008	1.10	5.7	.017	1.66	5.9	.015
Ambulate Independently PTA				2.25	4.3	.039	1.02	3.9	.049	1.67	3.8	.050
R2	.413			.279			.275			.191

NOTE. Variables allowed in regressions; age, female, admission motor FIM score, admission cognitive FIM score, ambulation independent before admission, ADLs independent before admission, employed before admission, maximum severity score (CSI), rehab LOS, stroke on right side of brain, stroke on left side of brain, bilateral stroke, stroke on unknown side of brain, race, days from symptom onset to acute admission ≥4, acute admission LOS ≥20 days, post PPS.

Abbreviations: Coeff, coefficient; PPS, prospective payment system; PTA, prior to admission.

Table 4. Regression Results for Outcomes of Discharge FIM Scores and Rehabilitation LOS for Patients With Severe Stroke in CMGs 108 to 114 (n=469)

Independent Variables	Dependent Variables Score
	Discharge Total FIM Score			Discharge Motor FIM Score			Discharge Mobility FIM Score			Discharge ADL FIM Score			Rehab LOS (n=478)
	Coeff	F	P	Coeff	F	P	Coeff	F	P	Coeff	F	P	Coeff	F	P
Days stroke onset to rehab adm	−0.15	26.7	<.001	−0.14	28.2	<.001	−0.05	24.1	<.001	−0.06	24.5	<.001
Partial R2		.022			.027			.032			.024
Age	−0.22	15.3	<.001	−0.20	16.5	<.001	−0.08	14.1	<.001	−0.07	9.5	.002	–0.13	14.0	<.001
Black	−3.75	5.1	.024	−2.85	4.2	.041
Stroke on right side of brain				−2.36	4.0	.046	−1.22	5.6	.018
Stroke on left side of brain										1.47	6.8	.009
Admission motor FIM score	1.22	104.0	<.001	1.18	157.0	<.001	0.44	143.8	<.001	0.47	101.4	<.001	–0.32	18.0	<.001
Admission cognitive FIM score	0.86	63.9	<.001							0.10	5.8	.017	0.14	4.0	.046
Maximum CSI score	−0.16	18.0	<.001	−0.12	17.6	<.001				−0.06	18.7	<.001	0.11	21.9	<.001
Employed PTA	4.87	7.0	.008	4.33	7.8	.006	1.41	4.3	.039	1.79	6.2	.013	2.40	4.2	.042
ADLs independent PTA				3.77	4.5	.034	1.69	4.8	.029				4.91	14.0	<.001
Rehab LOS	0.45	39.6	<.001	0.30	23.9	<.001	0.08	9.1	.003	0.16	32.3	<.001
Post PPS													−1.91	4.1	.043
R2	.540			.443			.322			.428			.197

NOTE. Variables allowed in regressions: age, female, admission motor FIM, admission cognitive FIM, ambulation independent prior to admission, ADLs independent prior to admission, employed prior to admission, maximum severity score (CSI), rehab LOS, stroke on right side of brain, stroke on left side of brain, bilateral stroke, stroke on unknown side of brain, race, days from symptom onset to acute admission ≥4, acute admission LOS ≥20 days, post PPS.

Abbreviation: adm, admission.

Back to Article Outline

Discussion 

The purpose of the multicenter PSROP was to open the “black box” of rehabilitation and determine, as precisely as possible, how specific elements of the rehabilitation process contribute to clinical outcomes. Timing of initiation of rehabilitation is one of those elements. Consistently, we found that fewer days from onset of stroke symptoms to rehabilitation admission was associated significantly with better functional outcomes: higher total, motor, mobility, and ADL discharge FIM scores. For severely impaired patients with stroke in CMGs 108 to 114, the relation was strongest, with F statistics greater than 24.1 for each functional outcome.

Also, for patients with moderately severe stroke in CMGs 104 to 107, fewer days from onset of stroke symptoms to rehabilitation admission was associated significantly with a shorter rehabilitation LOS (P=.038). The most intuitive reason for this relation is that those patients with less severe strokes and/or medical comorbidities naturally would be able to begin rehabilitation sooner and would also progress faster through rehabilitation, resulting in a shorter LOS. However, this relation was true after controlling for medical comorbidity and complications with the maximum CSI score. Perhaps earlier rehabilitation efforts provide patients with more practice opportunities to maximize functional gains, giving them a head start on entering rehabilitation. Or it could be that the additional stimulation of early rehabilitation enhances blood flow to injured areas and/or the ischemic penumbra, speeding clearance of toxic waste products such as free radicals and enhancing the healing process rather than inhibiting it. The optimal time window for increased synaptic plasticity (as mentioned elsewhere) may also occur early in the poststroke period, allowing for greater gains if rehabilitation is carried out during this critical interval. Additional research in laboratory animals could be done to confirm or refute this hypothesis, but the generalizability of the results in humans would still be uncertain.

Having microlevel data provided the ability to focus on the individual patient level to explore reasons for our findings and whether the findings would disappear when other variables were controlled for. The CSI enabled us to go beyond controlling only for stroke severity: it allowed us to control for many complex comorbidities common to patients with stroke, reflecting more accurately the realities of clinical practice. The maximum CSI score predicted outcomes as expected: a higher score (sicker patient) was associated with a lower discharge FIM score and its component scores and also with a longer rehabilitation LOS. These data support the hypothesis that early inpatient rehabilitation for patients with moderate and severe stroke and more days of acute inpatient rehabilitation for patients with severely impaired stroke are associated with better functional outcomes, after controlling for severity of illness. The findings also are consistent with prior literature regarding the importance of such variables as age, sex, race, severity of illness, baseline level of function, and employment before stroke.

It should be noted that the standard deviation of the time interval from stroke onset to initiation of rehabilitation was quite large (see table 1). This variability is at least in part due to the wide range of practices in delivery of poststroke rehabilitation care that exist between different regions of the country, hospital systems, and provider groups. In some areas, patients sometimes were discharged to skilled nursing facilities before entering inpatient rehabilitation. Hospital bed availability, staffing shortages, insurance coverage, and socioeconomic status of patients all potentially contribute to the variation in the time interval between stroke onset and rehabilitation. Efforts to reduce this variability by prioritizing timely rehabilitation of patients with stroke could result in substantial improvements in patient functional status, reduced LOS, and decreased health care costs.

As mentioned earlier, some have theorized that early aggressive rehabilitation potentially might be harmful to patients with stroke because of increased oxidative and/or metabolic stress on vulnerable cells. It is of interest, therefore, that the most severely impaired group had the strongest association between earlier rehabilitation and better functional outcomes. Presumably, if this theory were true, these severely affected patients would be most vulnerable to any harmful effects of early rehabilitation. In fact, it is this group that apparently gains the most functional benefits from early rehabilitation. These results concur with our findings in subgroup analyses of physical therapy, occupational therapy, and speech-language pathology therapy.³⁷, ³⁸, ³⁹

This is encouraging news. Although the evidence is not conclusive, the findings from this multicenter study regarding timing of initiation of rehabilitation after stroke suggest that we need not fear implementing early, aggressive interventions for severely affected patients with stroke; on the contrary, it appears to be the best thing we can do to maximize return of function. In addition, moderately affected patients with stroke apparently benefit from early rehabilitation, both in terms of functional outcomes and shorter LOS. Hence, greater efforts to initiate rehabilitation as soon as feasible and to transfer patients to dedicated rehabilitation facilities in a timely manner should result in greater rehabilitation efficiency and improved functional outcomes in most patients with stroke. Lack of a defined time point for measurement of functioning after stroke was a limitation of the study. Ideally, the FIM score would be measured at some predetermined endpoint (eg, 30d after stroke onset) for all patients. Use of rehabilitation discharge FIM scores is less satisfactory, because discharge itself is affected by institutional policies, patient preferences, socioeconomic concerns, insurance coverage, rate of recovery, and other variables. Despite these limitations, these analyses offer opportunities to uncover new insights and to confirm or reject original hypotheses. The significance of the time period between stroke onset and rehabilitation admission provides opportunities to alter care processes to achieve best possible outcomes.

As the population ages and the health care system faces challenges of providing high quality care with limited resources, it is increasingly important to identify the most effective and efficient means of delivering rehabilitation services to patients with stroke. There are enormous challenges inherent in studying a topic as complex as stroke rehabilitation. Strokes and patients with stroke are heterogeneous. Rehabilitation interventions vary in content, process, duration, intensity, and purpose. Functional outcome measures are difficult to define and interpret. The CPI methodology applied in this project enabled a comprehensive and detailed approach to this topic. The findings provide evidence for the importance of early rehabilitation in patients with stroke routinely encountered in clinical practice, and they could help lay the groundwork for future research and possibly randomized controlled trials.

Back to Article Outline

Conclusions 

For moderately and severely impaired patients with stroke, fewer days between onset of stroke symptoms and admission to inpatient rehabilitation is associated with better functional outcomes at discharge. For moderately impaired patients with stroke, fewer days between onset of stroke symptoms and admission to acute inpatient rehabilitation also is associated with shorter rehabilitation LOS. Providers should strive to transfer patients with stroke as soon as possible from an acute care hospital into acute rehabilitation to improve functional outcomes.

Supplier

Back to Article Outline

Acknowledgments 

We acknowledge the role and contributions of their collaborators at each of the clinical sites represented in the Post-Stroke Rehabilitation Outcomes Project: Brendan Conroy, MD (Stroke Recovery Program, National Rehabilitation Hospital, Washington, DC); Richard Zorowitz, MD (Department of Rehabilitation Medicine, University of Pennsylvania Medical Center, Philadelphia, PA); David Ryser, MD (Neuro Specialty Rehabilitation Unit, LDS Hospital, Salt Lake City, UT); Jeffrey Teraoka, MD (Division of Physical Medicine and Rehabilitation, Stanford University, Palo Alto, CA); Frank Wong, MD, and LeeAnn Sims, RN (Rehabilitation Institute of Oregon, Legacy Health Systems, Portland, OR); Murray Brandstater, MD (Loma Linda University Medical Center, Loma Linda, CA); and Harry McNaughton, MD (Wellington and Kenepuru Hospitals, Wellington, NZ). We also acknowledge the role of Alan Jette, PhD (Rehabilitation Research & Training Center on Medical Rehabilitation Outcomes, Boston University, Boston, MA).

Back to Article Outline

References 

1. American Heart Association . Heart disease and stroke statistics—2005 update . Dallas: AHA; 2004; Available at: http://www.americanheart.org/presenter.jhtml?identifier=1928. Accessed August 31, 2005
   • View In Article
2. Howard G , Howard VJ , Katholi C , Oli MK , Huston S . Decline in US stroke mortality (an analysis of temporal patterns by sex, race, and geographic region) . Stroke . 2001;32:2213–2220
   • View In Article
   • CrossRef
3. Ottenbacher KJ , Jannell S . The results of clinical trials in stroke rehabilitation research . Arch Neurol . 1993;50:37–44
   • View In Article
   • MEDLINE
4. Gresham GE . Past achievement and new directions in stroke outcome research . Stroke . 1990;21(Suppl II):II1–II2
   • View In Article
   • MEDLINE
5. Brandstater ME . An overview of stroke rehabilitation . Stroke . 1990;213(Suppl II):II40–II42
   • View In Article
6. Cifu DX , Steward DG . Factors affecting functional outcome after stroke (a critical review of rehabilitation interventions) . Arch Phys Med Rehabil . 1999;80(5 Suppl 1):S35–S39
   • View In Article
   • Abstract
   • Full-Text PDF (669 KB)
   • CrossRef
7. Reding MJ , McDowell FH . Focused stroke rehabilitation programs improve outcome . Arch Neurol . 1989;46:701–703
   • View In Article
   • MEDLINE
8. Gresham GE . Stroke outcome research . Stroke . 1986;17:358–360
   • View In Article
   • MEDLINE
   • CrossRef
9. Dombovy ML , Sandok BA , Basford J . Rehabilitation for stroke (a review) . Stroke . 1986;17:363–369
   • View In Article
   • MEDLINE
   • CrossRef
10. Lind K . A synthesis of studies on stroke rehabilitation . J Chronic Dis . 1982;35:133–149
    • View In Article
    • MEDLINE
    • CrossRef
11. Dobkin BH . Focused stroke rehabilitation programs do not improve outcome . Arch Neurol . 1989;46:701–703
    • View In Article
    • MEDLINE
12. Paolucci S , Antonucci G , Grasso MG , et al.   Early versus delayed inpatient stroke rehabilitation (a matched comparison conducted in Italy) . Arch Phys Med Rehabil . 2000;81:695–700
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (182 KB)
    • CrossRef
13. McNaughton H , Weatherall M , Taylor W , McPherson K . Factors influencing rate of Barthel Index change in hospital following stroke . Clin Rehabil . 2001;15:422–427
    • View In Article
    • MEDLINE
    • CrossRef
14. Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke . Stroke Unit Trialists’ Collaboration . BMJ . 1997;314:1151–1159
    • View In Article
    • MEDLINE
    • CrossRef
15. Langhorne P , Pollock A . What are the components of effective stroke unit care? . Age Ageing . 2002;31:365–371
    • View In Article
    • CrossRef
16. Assembling evidence about stroke units . In:  Langhorne P ,  Dennis M editor. Stroke units (an evidence based approach) . London: BMJ Books; 1998;p. 20–32
    • View In Article
17. Ronning OM , Guldvog B . Stroke unit versus general medical wards. II: Neurological deficits and activities of daily living . Stroke . 1998;29:586–590
    • View In Article
    • MEDLINE
    • CrossRef
18. Novack TA , Satterfield WT , Lyons K , Kolski G , Hackmeyer L , Connor M . Stroke onset and rehabilitation (time lag as a factor in treatment outcome) . Arch Phys Med Rehabil . 1984;65:316–319
    • View In Article
    • MEDLINE
19. Smith ME , Garraway WM , Smith DL , Akhtar AJ . Therapy impact on functional outcome in a controlled trial of stroke rehabilitation . Arch Phys Med Rehabil . 1982;63:21–24
    • View In Article
    • MEDLINE
20. Speach DP , Dombovy ML . Recovery from stroke . Ballieres Clin Neurol . 1995;4:317–318
    • View In Article
21. Johnston MV , Keister M . Early rehabilitation for stroke patients (a new look) . Arch Phys Med Rehabil . 1984;65:437–441
    • View In Article
    • MEDLINE
22. Alexander H , Bugge C , Hagen S . What is the association between the different components of stroke rehabilitation and health outcomes? . Clin Rehabil . 2001;15:207–215
    • View In Article
    • MEDLINE
    • CrossRef
23. Ballinger C , Ashburn A , Low J , Roderick P . Unpacking the black box of therapy—a pilot study to describe occupational therapy and physiotherapy interventions for people with stroke . Clin Rehabil . 1999;13:301–309
    • View In Article
    • MEDLINE
    • CrossRef
24. Turton A , Pomeroy V . When should upper limb function be trained after stroke? Evidence for and against early intervention . NeuroRehabilitation . 2002;17:215–224
    • View In Article
    • MEDLINE
25. Stroemer RP , Kent TA , Hulsebosch CE . Neocortical neural sprouting, synaptogenesis, and behavioral recovery after neocortical infarction in rats . Stroke . 1995;26:2135–2144
    • View In Article
    • MEDLINE
    • CrossRef
26. Liepert J , Bauder H , Miltner WH , Taub E , Weiller C . Treatment-induced cortical reorganization after stroke in humans . Stroke . 2000;31:1210–1216
    • View In Article
    • MEDLINE
    • CrossRef
27. Hagemann G , Redecker C , Neumann-Haefelin T , Freund HJ , Witte O . Increased long-term potentiation in the surround of experimentally induced focal cortical infarction . Ann Neurol . 1998;44:255–258
    • View In Article
    • MEDLINE
    • CrossRef
28. Jones TA , Shallert T . Overgrowth and pruning of dendrites in adult rats recovering from neocortical damage . Brain Res . 1992;581:156–160
    • View In Article
    • MEDLINE
    • CrossRef
29. Baird AE , Benfield A , Schlaug G , et al.   Enlargement of human cerebral ischemic lesion volumes measured by diffusion-weighted magnetic resonance imaging . Ann Neurol . 1997;41:581–589
    • View In Article
    • MEDLINE
    • CrossRef
30. Risedal A , Zeng J , Johansson BB . Early training may exacerbate brain damage after focal brain ischemia in the rat . J Cereb Blood Flow Metab . 1999;19:997–1003
    • View In Article
    • MEDLINE
31. Farrell R , Evans S , Corbett D . Environmental enrichment enhances recovery of function but exacerbates ischemic cell death . Neuroscience . 2001;107:585–592
    • View In Article
    • MEDLINE
    • CrossRef
32. DeJong G , Horn SD , Conroy B , Nichols D , Healton EB . Opening the black box of post-stroke rehabilitation (stroke rehabilitation patients, processes, and outcomes) . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S1–S7
    • View In Article
    • MEDLINE
33. Gassaway J , Horn SD , DeJong G , Smout RJ , Clark C , James R . Applying the clinical practice improvement approach to stroke rehabilitation (methods used and baseline results) . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S16–S33
    • View In Article
    • MEDLINE
34. In:  Hart AC ,  Schmidt KM ,  Aaron WS editor. ICD-9-CM code book . Reston: St. Anthony’s Publishing; 1999;
    • View In Article
35. Carter G , Relles D , Buchanan J , et al.   A classification system for inpatient medical rehabilitation patients (a review and proposed revisions to the functional independence measure−function-related groups) . Santa Monica: RAND, UCLA, Harvard Center for Health Care Financing Policy Research; 1997; HCFA Publication No. PM-682-HCFA
    • View In Article
36. Carter GM , Beeuwkes-Buntin M , Hayden O . Analysis for the initial implementation of the inpatient rehabilitation facility prospective payment system . Santa Monica: RAND; 2002; CMS Publication No. 99-154
    • View In Article
37. Latham NK , Jette DU , Slavin M , et al.   Physical therapy during stroke rehabilitation for people with different walking abilities . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S41–S50
    • View In Article
    • MEDLINE
38. Richards LG , Latham NK , Jette DU , Rosenberg L , Smout RJ , DeJong G . Characterizing occupational therapy practice in stroke rehabilitation . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S51–S60
    • View In Article
    • MEDLINE
39. Hatfield B , Millet D , Coles J , Gassaway J , Conroy B , Smout RJ . Characterizing speech and language pathology outcomes in stroke rehabilitation . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S61–S72
    • View In Article
    • MEDLINE

• a Release 8.2; SAS Institute Inc, 100 SAS Campus Dr, Cary NC 27513.