Stroke Unit Care Combined With Early Supported Discharge
Long-Term Follow-Up of a Randomized Controlled Trial
Background and Purpose— Early supported discharge from a stroke unit reduces the length of hospital stay. Evidence of a benefit for the patients is still unknown. The aim of this trial was to evaluate the long-term effects of an extended stroke unit service (ESUS), characterized by early supported discharge. The short-term effects were published previously.
Methods— We performed a randomized controlled trial in which 320 acute stroke patients were allocated to either ordinary stroke unit service (OSUS) (160 patients) or stroke unit care with early supported discharge (160 patients). The ESUS consists of a mobile team that coordinates early supported discharge and further rehabilitation. Primary outcome was the proportion of patients who were independent as assessed by modified Rankin Scale (RS) (RS ≤2=global independence). Secondary outcomes measured at 52 weeks were performance on the Barthel Index (BI) (BI ≥95=independent in activities of daily living), differences in final residence, and analyses to identify patients who benefited most from an early supported discharge service. All assessments were blinded.
Results— We found that 56.3% of the patients in the ESUS versus 45.0% in the OSUS were independent (RS ≤2) (P=0.045). The number needed to treat to achieve 1 independent patient in ESUS versus OSUS was 9. The odds ratio for independence was 1.56 (95% CI, 1.01 to 2.44). There were no significant differences in BI score and final residence. Patients with moderate to severe stroke benefited most from the ESUS.
Conclusions— Stroke service based on treatment in a stroke unit combined with early supported discharge appears to improve the long-term clinical outcome compared with ordinary stroke unit care. Patients with moderate to severe stroke benefit most.
It is a challenge to organize a healthcare service that can provide effective management of patients who have suffered from a stroke. Many trials have shown that care of acute stroke patients in stroke units improves outcome.1–6 Hence, access to stroke unit care is recommended for all acute stroke patients and should be the first link in the chain of care.7 However, the manner in which subsequent care should be organized is still an open question. A few small randomized controlled trials have emphasized early supported discharge and rehabilitation at home or in day hospitals. These trials showed no significant improvement in functional outcome.8 A weakness of most of the trials is that optimal stroke unit care had not been provided in the acute phase. Hence, it has been difficult to draw definite conclusions about the benefit of such a service.8 The aim of the present trial was to compare the effect of an extended stroke unit service (ESUS) with the effect of evidence-based ordinary stroke unit service (OSUS). The essential elements of ESUS were initial treatment in a
See Editorial Comment, page 2691
stroke unit combined with early supported discharge coordinated by a mobile stroke team, cooperation with the primary healthcare system, and greater emphasis on rehabilitation at home.
Subjects and Methods
All acute stroke patients who were admitted to our stroke unit were assessed for inclusion in the trial. Inclusion criteria were as follows: patients from the city of Trondheim, Norway, admitted to the stroke unit with signs and symptoms of an acute stroke9; inclusion within 72 hours of admission and <7 days after the onset of symptoms; Scandinavian Stroke Scale (SSS) score between 2 and 57 points; ability to live independently before the onset of stroke; and no participation in other trials. Informed consent was obtained. The methodology of this study has been reported previously, as well as the short-term results.9
Patients who satisfied the criteria for inclusion were randomized to either a constructed ESUS or OSUS. Both groups received similar stroke unit care during the acute phase (eg, the first 1 to 2 weeks), with focus on early mobilization/rehabilitation combined with a standardized acute medical treatment program.9,10 This treatment has been shown to improve outcome compared with treatment provided in general medical wards.2,11,12 In the OSUS group, further follow-up was organized by the primary healthcare system.
Patients in the ESUS group were offered a comprehensive follow-up stroke service organized by a mobile team.9 This hospital-based team consisted of a physiotherapist, an occupational therapist, a nurse, and the consulting service of a stroke physician.
This is a long-term follow-up of a previously published trial, and the elements of the ESUS have been described previously. The most important elements of the ESUS are shown in Figure 1 and are as follows: (1) After inclusion, a systematic assessment of the patient and the patient’s needs was performed. (2) A visit at home was planned, if possible together with the patient, as soon as the patient was medically stable (often within 3 to 5 days after admission to the stroke unit). Discharge planning, including further needs for help, support, and rehabilitation, was initiated at that point and in the most comprehensive manner possible. (3) A discharge meeting with the patient, relatives, physician, and a member of the mobile team was arranged before discharge from the hospital and after ongoing assessment of rehabilitation needs. (4) Follow-up visits, if necessary, were performed in cooperation with the primary healthcare system. A large part of the rehabilitation was provided in day hospitals and outpatient clinics, and further needs of rehabilitation and secondary prevention were outlined to the primary healthcare service. (5) A visit to the outpatient clinic at the hospital was arranged 3 to 5 weeks after discharge. An evaluation of the period after discharge from the stroke unit was made. (6) An educational meeting was arranged 3 months after discharge for the patients and their relatives.
After the visit at the outpatient clinic approximately 1 month after discharge, the primary healthcare system became responsible for further follow-up. They could consult the mobile team if problems occurred.
The Regional Committee on Medical Research Ethics approved the study protocol.
Method of Randomization
After consent, randomization was restricted in permuted blocks with random number tables provided in sealed opaque envelopes.
Baseline characteristics were recorded before inclusion. The primary outcome was a modified Rankin Scale (RS) score at 52 weeks after inclusion, with RS ≤2 classified as independence. Secondary outcomes at 52 weeks were performance on the Barthel Index (BI), with BI ≥95 characterized as independent, differences in final residence, and length of stay in an institution before discharge to final residence. A preplanned subgroup analysis of patients with a baseline score on the SSS ≤52 was also a secondary outcome. Finally, we performed analyses to identify patients who benefited most from an early supported discharge service.
All assessments were blinded as far as is possible in such a trial and conducted by neutral physiotherapists specially trained in use of the assessment scales.
Age and severity of stroke measured by SSS and differences in BI and RS score at baseline were analyzed by the Mann-Whitney test. The other baseline characteristics were analyzed by the Pearson χ2 test. In all the main analyses, the intention to treat population was used. We also used the Pearson χ2 test to investigate the difference between the groups in place of stay (home, institution, deceased). The significance level was set at 0.05. Logistic regression was performed with the use of RS score (≤2) as the dependent variable and treatment, SSS score, sex, age, and cohabiting status as independent variables. The number needed to treat (NNT) in the ESUS group versus the OSUS group to achieve 1 more independent patient was calculated for all patients, with additional stratification for severity of stroke, age, and sex. These analyses were also preplanned.
A total of 468 acute stroke patients from the city of Trondheim were admitted to our stroke study during a period of 24 months from March 1995. They were all screened for inclusion in the trial. Three hundred twenty of these patients fulfilled the inclusion criteria and were included and randomly allocated to ESUS (160 patients) or OSUS (160 patients). One hundred forty-eight patients were excluded. Details of the exclusion criteria were published earlier.
There were no significant differences in baseline characteristics between the groups.9
Table 1 shows that 56.3% of the patients treated in the ESUS and 45.0% of the patients treated in the OSUS were independent according to the primary outcome of modified RS score (P=0.044). NNT to achieve 1 more independent patient in the ESUS group versus the OSUS group was 9 (95% CI, 4.6 to 345). Odds ratio for independence was 1.56 (95% CI, 1.01 to 2.44; P=0.045). With adjustment for independent variables (treatment, SSS score, sex, age, and cohabiting status), the effect of the ESUS was even greater (odds ratio=1.93; 95% CI, 1.12 to 3.32; P=0.018) (Table 2). No interactions between the independent variables were significant.
The proportion of patients with a BI score ≥95 was not significantly higher in the ESUS group (52.5% versus 46.3%; P=0.264). The number of patients living at home was significantly higher in the ESUS group at discharge and 6 weeks9 but not at 52 weeks (Table 1). We performed a subgroup analysis for patients with baseline SSS score ≤52, and the results were even more favorable according to modified RS score: 47% of the patients treated in the ESUS and 28% treated in the OSUS were independent according to modified RS score (P=0.005). NNT to achieve 1 more independent patient in the ESUS group versus the OSUS group was 7 (95% CI, 3.6 to 27.3).
The total average length of inpatient stay was 18.6 days in the ESUS group and 31.1 days in the OSUS group (P=0.0324). The average stroke unit stay was similar in both groups (11 days).
Using the fitted logistic regression model, we analyzed the relation between the severity of stroke and the NNT in the ESUS group versus the OSUS group to achieve 1 more independent patient (Figure 2). The figure shows the whole group and the patients divided into 2 age groups: patients aged <75 years and patients aged ≥80 years. As shown for the whole group, a baseline SSS score between 35 and 54 corresponds to a NNT <10. For younger patients, the greatest benefit was seen for mild and moderate stroke with SSS score between 28 and 50. For older patients, the curves show the greatest benefit with SSS score >41.
The results of this trial show for the first time that stroke patients who are offered stroke unit treatment combined with early supported discharge have a better long-term functional outcome after 1 year of follow-up than patients offered ordinary stroke unit care.
The modified RS documents significantly better functional outcome in the ESUS group (P=0.044). The NNT to avoid dependency or death of 1 patient was 9; in the subgroup analysis for patients with a moderate to severe stroke, NNT was approximately 7. This effect is comparable to that achieved with treatment with alteplase within 3 hours of stroke onset.13 The early supported discharge treatment seems to be highly effective.
The present trial is a randomized controlled trial and appears to be one of the few trials in which the included patients have been an unselected hospitalized stroke population.8 Another strength of the present trial is that all patients received optimal treatment in a stroke unit in the acute phase.2 We have previously shown a positive effect of ESUS up to 6 months,9 but the results now indicate that this service has important long-term effects as well. Furthermore, we have created a model to calculate the benefit of early supported discharge care for patients with different levels of stroke severity (Figure 2). In a study of this size with a limited number of patients, we should be careful to draw definite conclusions. However, the results indicate which patients benefit most and which patients might be given priority for such a service.
A limitation of this kind of trial is that we do not know exactly why the present service system works so well. The strength of the blinded evaluation procedure is also unclear, and no evaluation of the efficacy of this blinding procedure was performed. The trial lacked sufficient power to detect moderate differences in the secondary outcome BI. Alternatively, the BI may not be sensitive enough in long-term follow-up because of a “ceiling effect.”
Evidence is accumulating that supports the development of services that allow stroke patients to be sent home from the hospital earlier than usual, with appropriate levels of support. A few studies have already shown that early supported discharge reduces the use of inpatient rehabilitation.14–19 A meta-analysis of 7 published early supported discharge trials showed recently that the mean difference in inpatient rehabilitation was 13 days between early supported discharge care and traditional care (95% CI, 7 to 19).20 This corresponds to the results of our trial. However, our trial also shows a significantly better outcome at 1 year. Characteristics of this trial compared with most of the others8 are as follows: standard care in a stroke unit for both groups; inclusion soon after the onset of stroke; a relatively large number of participants; and an unselected patient population with greater functional impairment. These factors may explain some of the favorable results.
Our service with a mobile team coordinating the rehabilitation efforts with an emphasis on rehabilitation in familiar surroundings may allow for an “enriched environment” for the patients and their relatives. In contrast, it appears that early discharge from the hospital without enhanced stroke services may lead to an increase in morbidity.21 Community rehabilitation may offer a significant advantage concerning the long-term consequences of stroke.22 We also believe that continuity in the chain of care is of great importance for the efficacy of the rehabilitation process. Compared with the other studies, we likely have more links in the chain of care in our ESUS. This includes various combinations of home service and inpatient rehabilitation.
In our trial, ESUS, which combines stroke unit care with early supported discharge, shows not only a reduction in hospital stay, similar to other early supported discharge trials, but also a significant improvement of long-term functional outcome compared with OSUS. This is the first long-term follow-up of an early supported discharge service combined with stroke unit care, and other similar trials are necessary to confirm the benefits of such a service. In addition, the effect on quality of life, resource implications, and economic consequences need further research.
This study was supported by the Norwegian Department of Health and the Norwegian Foundation for Health and Rehabilitation. We thank all the staff and patients who participated in this study. We also thank Gitta Rohweder for linguistic revision.
- Received April 14, 2003.
- Revision received June 8, 2003.
- Accepted July 8, 2003.
Stroke Unit Trialists’ Collaboration. A collaborative systematic review of the randomised controlled trial of organised inpatient (stroke unit) care after stroke. BMJ. 1997; 314: 1151–1159.
Indredavik B, Bakke F, Solberg R, Rokseth R, Håheim LL, Holme I. Benefit of a stroke unit: a randomized controlled trial. Stroke. 1991; 22: 1026–1031.
Kaste M, Palmomaki H, Sarna S. Where and how should elderly stroke patients be treated? A randomized trial. Stroke. 1995; 26: 249–253.
Strand T, Asplund K, Eriksson S, Hegg E, Lithner F, Wester PO. A nonintensive stroke unit reduces functional disability and the need for long-term hospitalization. Stroke. 1985; 16: 29–34.
Garraway WM, Akthar AJ, Hockey L, Presscott RJ. Management of acute stroke in the elderly: preliminary results of a controlled trial. BMJ. 1980; 280: 1040–1044.
Indredavik B, Bakke F, Slørdahl SA, Rokseth R, Håheim LL. Stroke unit treatment: 10-year follow-up. Stroke. 1999; 30: 1524–1527.
World Health Organization. Prevention, Treatment, and Rehabilitation. Geneva, Switzerland: World Health Organization; 1971.WHO Technical Report No. 469.
Early Supported Discharge Trialists. Services for reducing duration of hospital care for acute stroke patients. In: The Cochrane Library, issue 3, 2002. Oxford, UK: Update Software.
Indredavik B, Fjærtoft H, Ekeberg G, Løge AD, Mørch B. Benefit of an extended stroke unit service with early supported discharge: a randomized, controlled trial. Stroke. 2000: 31: 2989–2994.
Indredavik B, Bakke F, Slørdahl SA, Rokseth R, Håheim LL. Treatment in a combined acute and rehabilitation stroke unit: which aspects are most important? Stroke. 1999; 30: 917–923.
Indredavik B, Bakke F, Slørdahl SA, Rokseth R, Håheim LL. Stroke unit treatment improves long-term quality of life: a randomized controlled trial. Stroke. 1998; 29: 895–899.
Indredavik B, Slørdahl SA, Bakke F, Rokseth R, Håheim LL. Stroke unit treatment: long-term effects. Stroke. 1997; 28: 1861–1866.
Warlaw JM, del Zoppo G, Yamaguchi T. Thrombolysis for acute ischemic stroke (Cochrane review). In: The Cochrane Library, issue 3, 2001. Oxford, UK: Update Software.
Rudd AG, Wolfe CDA, Tilling K, Beech R. Randomised controlled trial to evaluate early discharge scheme for patients with stroke. BMJ. 1997; 315: 1039–1044.
Anderson C, Rubenach S, Mhurchu CN, Clark M, Spencer C, Winsor A. Hospital or home for stroke rehabilitation? Results of a randomized controlled trial, I: health outcomes at 6 months. Stroke. 2000; 31: 1024–1031.
Mayo N, Wood-Dauphinee S, Cote R, Gayton D, Carlton J, Buttery J, Tamblyn R. There’s no place like home: an evaluation of early supported discharge for stroke. Stroke. 2000; 31: 1016–1023.
Rodgers H, Soutter J, Kaiser W, Pearson P, Dobsn R, Skilbeck C, Bond J. Early supported hospital discharge following acute stroke: pilot study results. Clin Rehabil. 1997; 11: 280–287.
Widen Holmqvist L, von Koch L, Kostulas V, Holm M, Widsell G, Tegler H, Johansson K, Almazan J, de Pedro-Cuesta J. A randomized controlled trial of rehabilitation at home after stroke in southwest Stockholm. Stroke. 1998; 29: 591–597.
Ronning OM, Guldvog B. Outcome of subacute stroke rehabilitation: a randomized controlled trial. Stroke. 1998 29: 779–784.
Rodgers H. Rehabilitation: Hospital or Home? An Overview. Newcastle upon Tyne, UK: University of Newcastle upon Tyne; 1999.