A Comparison of Stroke Rehabilitation Practice and Outcomes Between New Zealand and United States Facilities

Article Outline

• Abstract
• Stroke Epidemiology and Services in NZ Versus the United States
• Methods
  • Analysis
• Results
• Discussion
• Conclusions
• Acknowledgments
• References
• Copyright

Abstract 

McNaughton H, DeJong G, Smout RJ, Melvin JL, Brandstater M. A comparison of stroke rehabilitation practice and outcomes between New Zealand and United States facilities.

Objective

To compare stroke rehabilitation practice and outcomes between New Zealand (NZ) and the United States.

Design

Prospective observational cohort study.

Setting

Seven inpatient rehabilitation facilities (IRFs) in the United States and NZ.

Participants

Consecutive convenience sample of 1161 patients in 6 U.S. IRFs and 130 in 1 NZ IRF (age, >18y) after acute stroke.

Interventions

Not applicable.

Main Outcome Measures

Change in FIM score and discharge destination.

Results

NZ participants were older than U.S. participants (mean: 74.1y vs 66.0y, respectively; P<.001). Measures of initial stroke severity were higher for U.S. participants. Mean rehabilitation length of stay (LOS) was shorter for U.S. participants (18.6d vs 30.0d, P<.001), but physical and occupational therapy time per patient was considerably higher despite the shorter LOS. U.S. therapists were involved in more active therapies for more of the time. Outcomes were better for U.S. participants, with fewer discharged to institutional care (13.2% vs 21.5%, P=.006) and larger changes in FIM scores.

Conclusions

U.S. participants with acute stroke who were selected for rehabilitation had better outcomes than NZ participants, despite shorter stays in the rehabilitation facility. U.S. participants had more intensive input from physiotherapists and occupational therapists, which may explain some of the larger increases in FIM scores. This suggests that further studies with tighter controls on case mix may add additional information on the effects of therapy intensity on patients with stroke.

Key Words:  Cerebrovascular accident , Health care systems , Outcome assessment (health care) , Rehabilitation

STROKE REHABILITATION is a complicated undertaking. Convincing evidence exists for the use of specialized stroke rehabilitation,¹ but little evidence currently exists to help stroke rehabilitation clinicians determine exactly how the details of stroke rehabilitation should be practiced. As a result, it is inevitable that different stroke rehabilitation teams will operate in different ways. These differences might be especially noticeable across different health systems, because stroke rehabilitation services react to various incentives and disincentives in an effort to provide the best care they can within resource constraints for the patients whom they manage. A comparison of the structures, processes, and outcomes of stroke rehabilitation services across different health systems might tell us much about what is important in stroke rehabilitation and give clinicians and funders in those systems information to guide future change.

Few attempts at international comparisons of stroke management have been published. The most ambitious project comparing stroke outcomes across international borders has been with the BIOMED studies in Europe.² These researchers attempted to relate marked differences in stroke mortality and dependency to differences in stroke practices across countries, combining acute and postacute phases of stroke care and making allowances for case mix as permitted by the quality of the data collected. Those studies looked only at broad differences in practice, generally in the acute phase of care.

A comparison of the UK National Health Service (NHS) and the United States (using data from Kaiser Permanente in California, Medicare California, and U.S. Medicare as a whole) was made with stroke care as 1 key diagnosis.³ This suggested that for people aged 65 years and older, admissions rates for stroke were broadly similar (NHS, 823/100,000; Kaiser, 788; Medicare California, 1155; U.S. Medicare, 1183). Mean lengths of stay (LOSs) were markedly different (NHS, 27.1d; Kaiser, 4.3d; Medicare California, 5.8d; U.S. Medicare, 6.5d). However, as pointed out by various commentators,⁴ the analysis was significantly flawed: for the UK, postacute (ie, rehabilitation) care was included, whereas for the U.S. data, it was not.

New Zealand (NZ) has a public health system modeled on that of the UK, and the Post-Stroke Rehabilitation Outcomes Project (PSROP) provides an opportunity to compare stroke rehabilitation resource use and outcomes, adjusted for case mix, that was not possible in the analysis of Ham et al.³ We know of no previous attempt to compare inpatient stroke rehabilitation practice between different countries. Previous comparisons of stroke rehabilitation practice across different hospitals but within the same health system have been at the level of retrospective audit of practice, comparing this with short-term outcomes.⁵, ⁶

The PSROP, which includes 6 U.S. sites and 1 NZ site, allows the opportunity to compare stroke rehabilitation practice between the 2 countries at a high level of detail with prospectively obtained data and standardization of data input and outcomes.

We aimed to identify differences in stroke rehabilitation practice between U.S. rehabilitation facilities and NZ hospitals and determine whether these differences affected outcomes at hospital discharge. The null hypothesis was that despite major differences in resource use between the 2 health systems for stroke rehabilitation, for people with stroke admitted to a rehabilitation facility, outcomes in terms of disability and institutionalization at hospital discharge would be similar once allowance was made for stroke case mix.

Back to Article Outline

Stroke Epidemiology and Services in NZ Versus the United States 

With what published information is available, it is possible to say that stroke prevalence is broadly similar between NZ and the United States. For people aged 65 years and older in the United States, stroke prevalence was approximately 45 cases per 1000, and in NZ it was 48 cases per 1000.⁷ Both populations have dominant white majorities with minority populations that have a higher prevalence of important stroke risk factors and stroke incidence. There is evidence from NZ that Maori and Pacific people who survive a stroke have worse case-mix adjusted outcomes at 12 months than for European New Zealanders.⁸ There is also evidence that a similar relation may hold for ethnic minorities in the United States.⁹ In both countries, hospital admissions for stroke have increased and stroke mortality has declined over the last 20 years.

In NZ, specialized stroke units for acute care and/or rehabilitation of stroke are rare,¹⁰ and the Wellington site involved in this study did not provide stroke unit care. Wellington is the capital city of NZ, and the health district provides services for about 250,000 people. There are 2 general hospitals, one of which is a university teaching hospital with tertiary services. About 90% of people with acute stroke in NZ are admitted to hospital, generally to a general medical ward.

Average LOS for an acute stroke admission in Wellington is about 7 days. During this acute stay, patients are assessed by members of a rehabilitation service and managed in 1 of 3 ways: likely to die in hospital (managed for whole admission in general medical ward), needs long-term institutional care and unlikely to benefit from short-term inpatient rehabilitation (discharged to institutional care), and may benefit from short-term inpatient rehabilitation (transferred to inpatient rehabilitation service). In Wellington, about 40% of all acute stroke admissions are transferred for inpatient rehabilitation. Depending on where a patient with stroke lives geographically, he/she is managed in 1 of 2 inpatient rehabilitation units, separated by 20km but managed by the same service along similar lines. Inpatient rehabilitation involves intense multidisciplinary team input, and patients are selected on the basis of their ability to participate in an active program of rehabilitation. However, there is no requirement that an arbitrary amount of clinician input (eg, 3h/d) be delivered.

All the hospital care (acute and rehabilitation) is provided free to patients as part of the NZ public health system. Outpatient rehabilitation is also free and provided for a limited time, generally less than 12 weeks.¹¹ Institutional care is provided with a cost to patients, although this is means tested, and a full government subsidy is provided for about 50% of all patients. Long-term care institutions provide a very limited amount of rehabilitation clinician input—for example, at most 1 hour of physiotherapy (PT) per week and usually no occupational therapy (OT), speech and language pathology (SLP), or social work input. Medical input is from family doctors. There is no equivalent of the skilled nursing facility (SNF) in NZ. This leads to a situation where patients, initially unable to tolerate intensive rehabilitation, might still be admitted to a rehabilitation facility before any consideration of long-term care as an option. In terms of acute and rehabilitation LOS, intensity of inpatient and outpatient rehabilitation, and general rehabilitation practice, Wellington is representative of what happens to people with stroke in most larger centers in NZ.¹⁰

In the United States, the trajectory of stroke care is somewhat different. After a brief stay in an acute care hospital, patients with stroke typically will be triaged to 1 of several locations: home, a hospital-based rehabilitation center if they are medically stable and can tolerate at least 3 hours of therapy a day, or an SNF if they have not achieved medical stability and are unable to tolerate a full dose of rehabilitation therapy. Those who are discharged home may receive home-based rehabilitation from a home health agency or may receive rehabilitation therapy at a rehabilitation outpatient center. Finally, some will be discharged to a nursing home if they are severely impaired and believed to be unable to benefit from rehabilitation. The transfer to one of these postacute settings is not always systematic and may depend in part on the preferences of a health plan and the advocacy skills of family members. Overall, the setting for postacute stroke rehabilitation will vary with each patient’s needs, health plan, family preferences, and geographic location, because types of postacute facilities vary from one part of the United States to another for reasons related to history, degree of urbanization, and the vagaries of local health care markets. Because most stroke survivors are older, they are eligible for Medicare, which remains the dominant payer of stroke rehabilitation services in the United States.

Back to Article Outline

Methods 

The methodology governing the full PSROP, provided in this supplement by Gassaway et al,¹² provides a detailed description of the larger study’s participating facilities, patient selection criteria, data collection instruments including their validity and reliability, and a detailed description of the project’s final study group. The methodology is summarized in Maulden et al.¹³ The institutional review boards at Boston University and at each participating inpatient rehabilitation facility (IRF) approved the study. NZ participants, along with those from 1 U.S. site, gave written consent for a 6-month telephone follow-up.

Wellington was the only non-U.S. site in the PRSOP, comprising 2 rehabilitation facilities that are geographically separated by 20km but are managed by the same overall service along similar lines. All study documentation was identical for both NZ and U.S. participants, with any uncertainties about labeling of specific activities and interventions by different therapists and nurses resolved by discussion with members of the study committee. The data manager for the NZ site (responsible for data extraction from clinical files) was trained in the United States along with U.S. site data managers and remained in close contact with the project manager throughout the study period.

Analysis 

Categoric variables were analyzed using the Fisher exact test or chi-square test, and continuous variables were analyzed by t test where assumptions of normality were met.

Back to Article Outline

Results 

There were 130 participants from NZ and 1161 participants from U.S. centers for comparison. The NZ population was significantly older (74.1y vs 66.0y, P<.001), less ethnically diverse (83.1% vs 58.1% white, P<.001), and more dependent before the stroke (dependent for activities of daily living [ADLs], 14.6% vs 9%, P=.002; dependent for ambulation: 26.9% vs 15.9%, P=.005) (table 1). Patients with stroke in NZ were also more likely to be living alone before their strokes (32.3% vs 20.7%, P=.004). Important stroke risk factors were similar in the 2 populations except for diabetes, which was more common in U.S. participants (27.7% vs 20.1%, P=.02) and mean weight, which was higher in U.S. participants (77.0kg vs 71.8kg, P=.007).

Table 1. Demographic Characteristics and Prestroke Variables

Demographic Characteristics	NZ (n=130)	U.S. (n=1161)	P
Mean age ± SD (y)	74.1±12.6	66.0±14.6	<.001⁎
Race (%)			<.001†
White	83.1	58.1
African American	0.0	25.8
Hispanic	0.0	7.7
Maori/Pacific	10.8	0.9
Asian	5.4	5.9
American Indian	0.0	0.8
Uncertain	0.8	1.0
Men (%)	51.5	51.9	1.00†
Health and functional characteristics
Prior stroke (%)	28.5	27.9	.920†
Hypertension diagnosis (%)	74.6	78.6	.310†
Diabetes diagnosis (%)	20.1	30.8	.020†
Current smoker (%)	24.6	20.9	.190†
Mean weight ± SD (kg)	71.8±16.3	77.0±18.4	.007⁎
ADL independent before stroke (%)	85.4	91.0	.002†
Ambulant without assistance or device before stroke (%)	73.1	84.1	.005†
Lived alone before stroke (%)	32.3	20.7	.004†

Abbreviation: SD, standard deviation.

Measures of stroke severity (table 2) at the time of maximum extent were somewhat similar between the 2 populations, including proportion with hemorrhages, proportion with aphasia, complete hemiplegia, and inability to walk. There was a surprisingly big difference in the diagnosis of depression (NZ, 0.8% vs U.S., 12.5%; P<.001) and “any mental health disorder” (NZ, 9.2% vs U.S., 54.9%; P<.001) between the 2 groups, which may reflect a combination of different thresholds for the diagnosis of a mental health disorder and/or a different level of likelihood that such a diagnosis is documented.

Table 2. A Comparison of Variables Describing the Extent and/or Consequences of Stroke

Stroke Variable	NZ (n=130)	U.S. (n=1161)	P
Hemorrhage (%)	20.0	23.3	.44§
Nonambulatory at max extent of stroke (%)	62.8	54.1	.09§
Complete hemiplegia (%)	7.7	12.8	.12§
Aphasia (%)	20.0	21.8	.74§
Depression in acute or rehab admission (%)	0.8	12.5	<.001§
Any mental health disorder (%)	9.2	54.9	<.001§
Mean admission to rehab FIM score ± SD
Total	65.6±28.6	61.0±20.3	.09‡
Motor	43.3±21.1	40.1±14.7	.11‡
Cognitive	22.3±10.7	21.0±8.3	.17‡
CMGs at rehabilitation admission (%)⁎			<.001§
Mild (CMG 101–103)	32.5	10.0
Moderate (CMG 104–107)	20.3	44.8
Severe (CMG 108–114)	47.2	45.2
Mean discrete CSI ± SD†	1.28±0.70	1.45±0.63	.005‡
Mean rehab admission continuous CSI ± SD†	15.6±10.5	20.7±13.7	<.001‡
Mean increase in severity during admission (max – admission CSI) ± SD†	10.4±13.0	10.7±11.7	.80‡

Abbreviations: CMG, case-mix group; CSI, Comprehensive Severity Index; rehab, rehabilitation.

At the time of admission to the rehabilitation facility, U.S. participants had nonsignificantly lower mean FIM scores (U.S., 61.0 vs NZ, 65.6) and significantly higher (worse) Comprehensive Severity Index (CSI) scores (U.S., 20.7 vs NZ, 15.6; P<.001) based on the continuous rather than the discrete 4-point version of the CSI.

Regarding the practice of rehabilitation in the different populations, there was a significantly shorter mean delay from stroke onset to rehabilitation admission for NZ participants (11.5d vs 13.8d, P=.011), although this was affected in part by a bimodal distribution for the U.S. participants, where the majority had a short delay (U.S. median delay, 7d vs NZ median delay, 9.5d) but others had a substantial delay (eg, entering the rehabilitation facility after a period in an SNF) (table 3).

Table 3. Comparison of Process Variables for Inpatient Stay of NZ and U.S. Participants

There was evidence of more intervention in the rehabilitation stay for U.S. participants for feeding (tube feeding: U.S., 16.9% vs NZ, 7.7%; P=.005) and oxygen (U.S., 16.5% vs NZ, 5.4%; P<.001). The mean rehabilitation LOS was significantly shorter for U.S. participants (18.6d vs 30.0d, P<.001), but during that time, more time was spent with a physiotherapist (U.S. mean, 800min vs NZ mean, 460.1min; P<.001) and occupational therapist (U.S. mean, 715.0min vs NZ mean, 208.4min; P<.001). U.S. participants were seen by a physiotherapist and occupational therapist on a larger proportion of the days that they spent in the rehabilitation facility (mean PT days/mean days in rehabilitation: U.S., 13.5/18.6d vs NZ, 13.3/30d; mean OT days/mean days in rehabilitation: U.S., 11.7/18.6d vs NZ, 5.8/30d).

Because therapists recorded what they did while working with participants, it is possible to make some comments about the actual components of rehabilitation practice within disciplines in the 2 systems.

NZ physiotherapists spent a greater proportion of their time than their U.S. counterparts (table 4) with participants engaged in assessment activities and lower-level mobility activities (bed mobility, sitting balance, sit to stand), whereas U.S. therapists spent a greater proportion of their time in higher-level mobility activities (transfers, pregait, gait, advanced gait).

Table 4. Comparison of the Components of Rehabilitation Practice: PT, OT, and SLP

Activity	Percent of Total Time Spent in Each Class of Activity		P
	NZ	U.S.
PT
Assessment	15.7	9.6	.001⁎
Movement activities before transfers	32.4	24.3	<.001⁎
Movement activities, transfers, and walking	38.7	54.0	<.001⁎
Nonfunctional time	23.3	21.5	.400⁎
OT
Assessments	49.4	10.7	<.001⁎
ADLs	30.5	31.0	.870⁎
Mobility activities including transfers and wheelchair	13.0	20.1	.016⁎
Working with upper limb	0.9	24.6	<.001⁎
Home management, leisure, and community integration	9.5	12.9	.147⁎
Nonfunctional activities	18.8	10.6	.014⁎
SLP
Assessment	26.9	19.3	.100⁎
Swallowing	50.7	19.3	<.050⁎
Voice and expression activities	30.8	31.2	.940⁎
Comprehension activities	6.4	14.4	<.001⁎
Memory, problem solving, and other cognitive activities	2.4	33.3	<.001⁎
Nonfunctional activities	8.5	1.1	<.050⁎

NZ occupational therapists spent a large proportion (NZ, 49.4% vs U.S., 10.7%; P<.001) of their time in assessments both in the facility and home. In NZ, a home visit before discharge is virtually routine and usually is conducted by the occupational therapists. Occupational therapists are also responsible for much of the cognitive testing, because psychologists are rarely available or used. On the other hand, U.S. occupational therapists spent a considerable portion of time with participants working with the upper limb, usually the domain of the physiotherapists in NZ.

NZ speech-language therapists spent most of their time with participants involved in assessment of or activities around swallowing (NZ, 50.7% vs U.S., 19.3, P<.05), whereas U.S. speech-language therapists spent most of their time in activities around expression, comprehension, and cognitive activities.

Overall, NZ therapists spent more time in assessment and nonfunctional activities than their U.S. counterparts. Nonfunctional activities are activities not directly related to the functional enhancement of a patient or time spent on a patient’s behalf but not in direct contact with the patient (eg, time selecting and ordering a wheelchair or splint).

Outcomes at hospital discharge were better for U.S. than NZ participants (table 5). For U.S. participants, fewer participants were discharged to institutional care (U.S., 13.2% vs NZ, 21.5%; P=.006), there was a bigger increase in FIM score during admission (U.S., 26.2 vs NZ, 20.6; P<.001), and the change in CSI score was greater (U.S., 10.2 vs NZ, 5.6; P<.001). It is possible that the criteria for admission to institutional care in the 2 countries may be different. However, it is possible to say that levels of disability at rehabilitation discharge were very similar, with mean FIM scores within 2 points of each other (U.S., 87.2 vs NZ, 85.6; P=.57).

Table 5. Comparison of Outcomes for NZ and U.S. Participants

Outcome Variables	NZ (n=130)	U.S. (n=1161)	P
Discharge destination, n (%)			.006⁎
Home	92(70.8)	906(78.0)
Community assisted living	3(2.3)	34(2.9)
Institutional	28(21.5)	153(13.2)
Hospital	1(0.8)	47(4.1)
Other acute rehab	3(2.3)	15(1.3)
Died	3(2.3)	6(0.5)
Mean discharge FIM score ± SD	85.6±30.7	87.2±22.5	.570†
Mean increase in FIM score ± SD	20.6±15.2	26.2±14.0	<.001†
Mean net medical improvement (admission – discharge CSI) ± SD‡	5.6±14.5	10.2±10.7	<.001†
Mean rehab discharge CSI continuous scores ± SD‡	10.0±16.0	10.5±12.6	.600†

Back to Article Outline

Discussion 

Our results show that significant differences exist for stroke rehabilitation practice and outcomes for participants in NZ and U.S. rehabilitation facilities. NZ participants tended to be older, frailer, and more likely to live alone before stroke, but U.S. participants scored somewhat worse on measures of disability and comorbidity at the beginning of stroke rehabilitation. U.S. participants stayed a much shorter time in the rehabilitation facility but had much higher input from PT and OT in that time, both in terms of the proportion of days on which they were seen and the total number of minutes of time.

U.S. participants had better outcomes, with more rapid change in disability scores and a lower chance of discharge to institutional care. These differences occurred despite the increased severity of U.S. participants’ disabilities at the time of their rehabilitation admissions.

The components of rehabilitation practice for different types of therapists were surprisingly different between NZ and U.S. facilities. U.S. therapists of all types spent a smaller proportion of their time in assessment and nonfunctional activities and proportionately more time in active management of participants. This was particularly so for occupational therapists. U.S. occupational therapists spent almost a quarter of their time involved in activities with the upper limb, an activity rarely performed by NZ occupational therapists and more often performed by NZ physiotherapists, as noted earlier. Speech-language therapists in NZ spent more of their time involved in swallowing activities than the more traditional speech and language activities, whereas U.S. speech-language therapists provided significant input into cognitive activities.

There are some major questions that might affect the interpretation of these results. First, how representative of NZ and U.S. practice are the facilities studied? Certainly, the NZ facility falls somewhere in the middle of NZ rehabilitation facilities for efficiency (rehabilitation LOS), and the staffing is broadly similar with other units in the country. The 6 U.S. facilities in the PSROP are a geographically diverse group of IRFs, and based on comparisons with a more nationally representative group of IRFs,¹⁴ these 6 facilities serve a somewhat nationally representative sample of stroke rehabilitation patients served in IRFs.

Second, how much difference does the age disparity between the U.S. and NZ study groups make on outcome, particularly institutionalization? One could argue that the older population studied in NZ was at higher risk of poststroke rehabilitation institutionalization. For study participants as a whole, increasing age was only very weakly associated with institutionalization. In an earlier NZ study, age was not a significant independent variable in rate of change of disability in hospital for people with stroke,¹⁴ suggesting that the age difference cannot be the sole explanation for the differences in practice patterns and outcomes reported here.

The cause of the age disparity needs to be considered, because this may suggest important unmeasured covariates in outcome. With the availability of additional postacute rehabilitation venues such as SNFs in the United States, it is possible that older patients with stroke in the United States are more likely to be managed in an alternative postacute setting than those in NZ. One consequence of this is that in NZ, a significant proportion of elderly people with stroke will be “given a go” in a rehabilitation facility, with a fairly high expectation of the need for eventual institutional care rather than for discharge direct from an acute hospital to the institutional setting. NZ lacks SNFs either as an alternative to an IRF or as an intermediate step on the way to an IRF. It could be argued that such facilities might inadvertently help improve the effectiveness and efficiency of rehabilitation facilities by enabling IRFs to work with patients who are more likely to succeed with IRF-level care. Thus, in the U.S., hospital-based rehabilitation facilities may be less likely to admit patients with any risk of not being discharged to home. This would have the effect of reducing the age of the population admitted to an IRF and providing a small advantage in favor of discharge home compared with NZ participants. Published U.S. data from 1999 show a mean age for patients with stroke admitted to IRFs of 70 years,¹⁵ whereas that for patients with stroke admitted to subacute rehabilitation facilities (mainly SNFs) was 76 years,¹⁶ suggesting some sort of selection process related to age. Nevertheless, age aside, the severity indicators in this study, which included various comorbid conditions, favored the NZ participants.

The higher proportion of NZ patients living alone before their strokes may have influenced more to be discharged to institutions. The NZ mean discharge FIM score of 85.6 indicates that a large proportion of those discharged required continued assistance. The relatively similar discharge FIM and CSI scores at discharge for the 2 patient groups would suggest that the differences were not due to clinical factors.

There is an increasing body of evidence that increasing the intensity of stroke rehabilitation improves outcomes,¹⁷, ¹⁸ and this study supports that notion. Although the case mix of these compared populations had differences, an analysis of these differences tends to support the conclusion that increased therapy intensity results in more rapid functional improvement in patients early after stroke. The difference in the rate of change of FIM scores between the groups was substantial. As discussed earlier, previous studies found that age had little effect on these rates of change. It seems possible that medical severity could delay functional recovery, yet it was greater in the U.S. population, which showed the fastest and greatest improvement. The discharge FIM scores of the 2 groups were not significantly different, so the prestroke differences in function do not seem to have had substantive effects. The closeness of the admission FIM scores suggests that the ceiling effects of the FIM had little influence. Living alone before admission should not influence functional capacity. Depression and mental health disorders would be more likely to slow rather than increase improvement rates. Although further studies on more closely matched populations may be needed to ultimately clarify the impact of more intense therapy, the evidence strongly suggests that more intensity can result in greater and more rapid gains in appropriately selected patients with stroke.

Back to Article Outline

Conclusions 

This study shows that it is not only the total hours of therapy that are important but what happens during the therapy session. Rehabilitation services that manage people with stroke should consider the level of intensity of therapy input and concentrate on active therapy. For NZ services, an overemphasis on assessment may contribute to delays in initiating active therapy, leading in turn to longer-than-necessary stays in a hospital. Given the therapy intensity for U.S. patients observed in this study, there may be ample opportunity to increase therapist input for NZ patients on more days during a rehabilitation stay. The NZ public health system has tended to focus too much on overall costs without examining the components of those costs that make a difference. Practices to promote efficiency, or better outcomes if they involve new spending (eg, such as more staff or staff working on 6 days rather than 5), have been difficult to implement. The results reported here provide some focus for a change in mindset that should benefit patients with stroke. It should be noted, however, that the costs of inpatient rehabilitation in the 2 countries are massively different—for NZ, the per-day cost is around US$320, whereas U.S. rehabilitation facilities charge around US$1050/d. Even with a much shorter LOS, the mean total cost of a rehabilitation stay in a U.S. facility is almost double that in NZ. Depending on one’s perspective (eg, patient, clinician, funder), the difference in outcomes reported here may or may not represent good value.

The other lesson from this study is that there is much to be learned from rehabilitation practitioners in different countries and different parts of the same country if robust study methods and analysis can be adopted. We note with interest a European study hosted at the Free University of Brussels with design elements similar to this study. That study, known as the Collaborative Evaluation of Rehabilitation in Stroke Across Europe, is investigating stroke rehabilitation practice in 4 European countries. Incorporating the best elements from many service delivery models may provide a rapid way to achieve better outcomes for people with stroke.

Back to Article Outline

Acknowledgments 

We acknowledge the role and contributions of the 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 (Rehabilitation Department, 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 and Training Center on Medical Rehabilitation Outcomes, Boston University, Boston, MA).

Back to Article Outline

References 

1. Stroke Unit Trialists’ Collaboration . Organised inpatient (stroke unit) care after stroke . Cochrane Database Syst Rev . 2002;1:CD000197
   • View In Article
2. Wolfe CD , Giroud M , Kolominsky-Rabas P , et al.  European Registries of Stroke (EROS) Collaboration   Variations in stroke incidence and survival in 3 areas of Europe . Stroke . 2000;31:2074–2079
   • View In Article
   • CrossRef
3. Ham C , York N , Sutch S , Shaw R . Hospital bed utilisation in the NHS, Kaiser Permanente, and the US Medicare programme (analysis of routine data) . BMJ . 2003;327:1257
   • View In Article
   • CrossRef
4. Ebrahim S , Frankel S , Davey Smith G . Hospital bed utilisation in the NHS and Kaiser Permanente (quality of care, length of stay, and readmissions need to be considered [letter]) . BMJ . 2004;328:583–584
   • View In Article
   • CrossRef
5. McNaughton H , McPherson K , Taylor W , Weatherall M . Relationship between process and outcome in stroke care . Stroke . 2003;34:713–717
   • View In Article
   • CrossRef
6. Weir N , Dennis MS . Towards a national system for monitoring the quality of hospital-based stroke services . Stroke . 2001;32:1415–1421
   • View In Article
   • CrossRef
7. Feigin VL , Lawes CM , Bennett DA , Anderson CS . Stroke epidemiology (a review of population-based studies of incidence, prevalence, and case-fatality in the late 20th century) . Lancet Neurol . 2003;2:43–53
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (247 KB)
   • CrossRef
8. McNaughton H , Weatherall M , McPherson K , Taylor W , Harwood M . The comparability of community outcomes for European and non-European survivors of stroke in New Zealand . N Z Med J . 2002;115:98–100
   • View In Article
   • MEDLINE
9. Stansbury JP , Jia H , Williams LS , Vogel WB , Duncan PW . Ethnic disparities in stroke (epidemiology, acute care, and postacute outcomes) . Stroke . 2005;36:374–386
   • View In Article
   • CrossRef
10. Gommans J , Barber A , McNaughton H , et al.   Stroke rehabilitation services in New Zealand . N Z Med J . 2003;116:U435
    • View In Article
11. McNaughton H , Weatherall M , McPherson K , Taylor W , Harwood M . The comparability of resource utilisation for Europeans and non-Europeans following stroke in New Zealand . N Z Med J . 2002;115:101–103
    • View In Article
    • MEDLINE
12. 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
13. Maulden SA , Gassaway J , Horn SD , Smout RJ , DeJong G . Timing of initiation of rehabilitation after stroke . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S34–S40
    • View In Article
    • MEDLINE
14. 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
15. Deutsch A , Fiedler RC , Granger CV , Russell CF . The Uniform Data System for Medical Rehabilitation report of patients discharged from comprehensive medical rehabilitation programs in 1999 . Am J Phys Med Rehabil . 2002;81:133–142
    • View In Article
    • MEDLINE
    • CrossRef
16. Deutsch A , Fiedler RC , Iwanenko W , Granger CV , Russell CF . Uniform Data System for Medical Rehabilitation report (patients discharged from subacute rehabilitation programs in 1999) . Am J Phys Med Rehabil . 2003;82:703–711
    • View In Article
    • MEDLINE
    • CrossRef
17. Kwakkel G , Wagenaar RC . Effect of duration of upper- and lower-extremity rehabilitation sessions and walking speed on recovery of interlimb coordination in hemiplegic gait . Phys Ther . 2002;82:432–448
    • View In Article
    • MEDLINE
18. Kwakkel G , van Peppen R , Wagenaar RC , et al.   Effects of augmented exercise therapy time after stroke (a meta-analysis) . Stroke . 2004;35:2529–2539
    • View In Article
    • CrossRef