Clinical Rehabilitation 2004; 18: 580
/586

Acute stroke unit care combined with early supported discharge. Long-term effects on quality of life. A randomized controlled trial Hild Fjærtoft Department of Neuroscience and Motion, Faculty of Medicine, Norwegian University of Science and Technology, The Stroke Unit, Department of Medicine, University Hospital of Trondheim and Department of Community Medicine and General Practice, Faculty of Medicine, Norwegian University of Science and Technology, Bent Indredavik Department of Neuroscience and Motion, Faculty of Medicine, Norwegian University of Science and Technology and The Stroke Unit, Department of Medicine, University Hospital of Trondheim, Roar Johnsen Department of Community Medicine and General Practice, Faculty of Medicine, Norwegian University of Science and Technology and Stian Lydersen Unit for Applied Clinical Research, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Received 28th December 2003; returned for revisions 28th January 2004; revised manuscript accepted 26th March 2004.

Objectives: The aim of the present trial was to compare the effects of an extended stroke unit service (ESUS) with the effects of an ordinary stroke unit service (OSUS) on long-term quality of life (QoL). Design: One year follow-up of a randomized controlled trial with 320 acute stroke patients allocated either to OSUS (160 patients) or ESUS (160 patients) with early supported discharge and follow-up by a mobile team. The intervention was a mobile team and close co-operation with the primary health care service. All assessments were blinded. Main outcome measure: Primary outcome of QoL in this paper was measured by the Nottingham Health Profile (NHP) at 52 weeks. Secondary outcomes measured at 52 weeks were differences between the groups measured by the Frenchay Activity Index, Montgomery-A˚sberg Depression Scale, Mini-Mental State Score and the Caregivers Strain Index. Results: The ESUS group had a significantly better QoL (mean score 78.9) assessed by global NHP after one year than the OSUS group (mean score 75.2) (p /0.048). There were no significant differences between the groups in the secondary outcomes, but a trend in favour of ESUS. Caregivers Strain Index showed a mean score of 23.3 in the ESUS group and 22.6 in the OSUS group (p /0.089). Conclusion: It seems that stroke unit treatment combined with early supported discharge in addition to reducing the length of hospital stay can improve long-term QoL. However, similar trials are necessary to confirm the benefit of this type of service.

Address for correspondence: Hild Fjærtoft, Department of Community Medicine and General Practice, ISM, Medisinsk teknisk forskningssenter, NTNU, Ha˚kon Jarlsgt. 12, N-7489 Trondheim, Norway. e-mail: [email protected] # Arnold 2004

10.1191/0269215504cr773oa

Acute stroke unit care combined with early supported discharge

Introduction The effectiveness of treatment in stroke units for acute stroke patients is well-documented.1 The organization of the subsequent care is still being discussed. The long-term consequences after stroke in general are, however, incompletely described in the literature, and are mainly focused on strokerelated disabilities and subsequent recovery of motor function.2 A few randomized early supported discharge trials (ESD) have focused on quality of life (QoL) aspects for patients, and some of them have found better QoL for patients in the ESD group from a short-term perspective.3
5 However most of the trials have not found any differences between ESD patients and hospitalized patients over time.6
9 It has been difficult to agree on a uniform way to measure QoL for stroke patients, and no single method has gained general acceptance. Hence, it seems that a multidimensional approach is required.10,11 We have previously shown a significant positive effect for the ESD group after six months and 12 months on functional outcome measured by Rankin Scale (RS) and Barthel Index (BI).12,13 We have also described a significant difference in institutional length of stay in favour of an extended stroke unit service (ESUS) group (12 days), and explored analyses to identify which patients would benefit most from an ESD service.12,13 The main aim of the present analyses was to compare the effect of an ESUS with that of an evidence-based ordinary stroke unit service (OSUS) related to long-term quality of life.

Methods The present paper is a report on a randomized controlled trial of 320 acute stroke patients. It is a long-term follow-up on quality of life at 12 months, and the methodology has been reported previously, as well as the short-term results12 and the improvement of long-term functional outcome.13 The primary outcome of the trial on which the sample size calculation was based was the Rankin Scale at 26 weeks.12

581

All patients admitted to the stroke unit at St Olav University Hospital from the City of Trondheim were assessed for inclusion. Inclusion criteria were as follows: Patients admitted to the stroke unit with signs and symptoms of an acute stroke12 were included within 72 hours of admission and less than seven days after the onset of symptoms; Scandinavian Stroke Scale (SSS) between 2 and 57 points; living independently before the onset of stroke; and not participating in other trials. Informed consent was obtained. After inclusion, the patients were randomized either to the ESUS group or to the OSUS group. Both groups received identical well-documented stroke unit care during the acute phase (e.g., the first 1
/2 week) with focus on early mobilization/ rehabilitation combined with a standardized medical programme. The primary health care system organized the follow-up for the OSUS group with further inpatient rehabilitation or a follow-up programme organized after discharge from the hospital,12 while a co-ordinating mobile team (MT) organized a comprehensive follow-up service for the ESUS group. The team tried to establish a service and a support system that allowed the patient to live at home as soon as possible after the stroke as well as to continue rehabilitation at home or in a day clinic. The team consisted of a nurse, an occupational therapist, a physiotherapist and a part time physician. The most important elements of the extended stroke unit service are shown in the ‘intervention-line’ in Figure 1, and have been described in detail previously.12,13 The Regional Committee on Medical Research approved the study protocol. The randomization was carried out in permuted blocks with random number tables provided in sealed opaque numbered envelopes. The size of the blocks was 6, 8 and 10. No stratification was used. Assessments Baseline characteristics were recorded before inclusion and have been presented earlier. All assessments were blinded as far as possible in such a trial and conducted by neutral physiotherapists specially trained in the assessment scales. Primary outcome of quality of life after 52 weeks in this paper was assessed by the Nottingham

582 H Fjærtoft et al.

Figure 1 The extended stroke unit service.

Health Profile (NHP).14 The NHP measures emotional, social and physical distress. It quantifies health status and has been used for the general evaluation of health. It emphasizes subjective aspects of health assessment. NHP part I contains 38 items in six sections: pain, emotional reactions, sleep, physical mobility, social isolation and energy level. Each item reflects departures from normal and the items are weighted to reflect their importance. Part I of the NHP forms a profile of six scores corresponding to the different sections of the questionnaire, and there is no single summary index. Each component is weighted to give a score 0
/100 (Thurstone method of paired comparison).15 Part II of the NHP relates to seven areas of task performance most affected by health. This section has no weights but a summary statistic. These data are of more limited use. We have also used a global NHP score.16 As secondary outcomes we measured the differences between the groups with the assessment scales Frenchay Activity Index,17 Montgomery˚ sberg Depression Scale18 and Mini-Mental State A Examination (MMSE).19 The Frenchay Activity Index has been developed specifically for measuring disability and handicap in stroke patients. It measures more complex physical activities and social functioning with a score for each of the 15 items (max score 60). The ˚ sberg Depression Scale is a rating Montgomery-A scale for mood disorders. It consists of 10 items (0
/6) and avoids emphasis on somatic symptoma-

tology. The MMSE is a simplified scored form of the cognitive mental status examination and includes 11 questions and a score of 30. The Caregivers Strain Index20 is a validated 13item strain index with questions to the caregivers. Statistical analyses Age and severity of stroke measured by Scandinavian Stroke Scale (SSS) and differences in Barthel Index (BI) and Rankin Scale (RS) at baseline were analysed by the Mann
/Whitney test. The other baseline characteristics were analysed by Pearson’s chi-squared test. To investigate any differences between the groups in the primary and secondary outcomes, we used the Mann
/ Whitney test. The missing cases are not quite the same for the different scales in the secondary outcome. The test population holds 252
/272 patients.

Results During a period from March 1995 to May 1997, 468 patients from the City of Trondheim were screened for inclusion in our trial. Three hundred and twenty patients were included after fulfilling the inclusion criteria, and 148 were excluded. The patients were randomly allocated to ESUS (160 patients) or to OSUS (160 patients). There were no significant differences in baseline characteristics,

Acute stroke unit care combined with early supported discharge

583

Figure 2 Flow diagram for the extended stroke unit service and the ordinary stroke unit service.

and the distribution of stroke diagnoses was similar in both groups.12 After one year, 139 participants in the ESUS group and 134 in the OSUS group were alive (Figure 2). Of these, 133 in the ESUS group and 125 in the OSUS group were assessed with the NHP. A total of 10 patients (6/4) could not be tested with this scale because of other severe disease, mental impairment or severe aphasia. One was terminally ill. Five patients in the control group dropped out of the trial for different personal reasons and did not want to be tested at one year. Table 1 shows the distribution of the NHP. There were no significant differences between the groups in any of the six dimensions: energy, emotional reactions, physical mobility,

social isolation, pain or sleep. For the global NHP part I, there was a significant better outcome in QoL for the ESUS group than for the OSUS group (p/0.048). For the secondary outcome, ˚ sberg Frenchay Activity Index, Montgomery-A Depression Scale, MMSE and Caregivers Strain Index, there were nonsignificant differences between the groups (p
/0.05) as presented in Table 2. The differences are in favour of the ESUS group.

Discussion The results of this trial indicate that stroke patients who received stroke unit treatment combined with

584 H Fjærtoft et al. Table 1 Global Nottingham Health Profile and the profile for the six dimensions in NHP at 52 weeks in the ESUS group and the OSUS group ESUS n /133)

OSUS (n /125)

p-value

Global NHP 1 Global NHP 2

81.6 (78.9) 84.3 (79.0)

78.9 (75.2) 79.4 (75.9)

0.048 0.073

Energy Emotional reactions Physical mobility Social isolation Pain Sleep

24.0 7.2 23.9 0.0 0.0 12.6

24.0 10.5 34.5 0.0 0.0 12.6

0.058 0.059 0.107 0.319 0.785 0.940

(33.6) (14.4) (31.6) (13.7) (12.5) (20.8)

(39.4) (19.7) (36.4) (15.6) (13.9) (19.5)

Values are median (mean). ESUS, extended stroke unit service; OSUS, ordinary stroke unit service. *p B/0.05.

early supported discharge reported a better quality of life measured by global NHP than patients who were offered ordinary stroke unit care (p/0.048). There was a nonsignificant trend in direction of better outcome for the ESUS group regarding the six items of the NHP, especially in the dimensions ‘Energy’ (p/0.058) and ‘Emotional reactions’ (p /0.059). For the secondary outcomes, Montgomery˚ sberg Depression Scale, Frenchay Activity Index A and MMSE, there were no significant differences between the groups. The Caregivers Strain Index showed a nonsignificant trend towards a reduced burden for the caregivers in the ESUS group (p /0.089).

The trial has a randomized controlled design, and appears to be one of the few trials enrolling an apparently unselected hospitalized stroke population.3 Furthermore, all the included patients have received the best available care in the acute phase (stroke unit treatment).1 The mortality was similar in both groups, and drop-outs were below 5%. We assume that the low number of drop-outs would probably not influence the results. Intention-totreat analysis is not feasible in this trial, because the NHP we have used requires communication with all the included patients. One of the challenges in this type of trial is the completeness of the blinded evaluation procedure, as the patients communicate with their assessors. We did not perform any evaluation of the efficacy of the blinding procedure. Another challenge is the lack of general agreement for standardized methods of QoL assessment in stroke patients. The NHP is often regarded as a measure of general perceived health status,21 but the profile has been used in several stroke trials as a measurement of QoL.14,22 It is also frequently used in clinical trials. Compared with other QoL instrument (e.g., Sickness Impact Profile), the NHP is relatively short and easy to complete, which is of great importance for stroke patients. The NHP assesses whether there are any health problems, but does not sufficiently identify particular problems. It may well be primarily measuring mood, and this may not be the best measure of QoL itself. The reliability and validity are sufficient high.15 More

Table 2 Secondary outcome Frenchay Activity Index, Montgomery-A˚sberg Depression Scale, MMSE and Caregivers Strain Index assessed at 52 weeks

Frenchay Activity Index Mean (SD) Median (range) Montgomery-A˚sberg Depression Scale Mean (SD) Median (range) Mini-Mental State Examination Mean (SD) Median (range) Caregivers Strain Index Mean (SD) Median (range)

ESUS

OSUS

(n /136) 32.2 (11.1) 31.0 (15
/54) (n /133) 5.6 (5.8) 3.0 (0
/24) (n /132) 26.0 (4.5) 27.0 (11
/30) (n /133) 23.3 (2.7) 24.0 (15
/26)

(n /128) 31.1 (11.1) 30.5 (15
/57) (n /126) 5.9 (6.3) 4.0 (0
/36) (n /125) 25.4 (5.7) 28.0 (3
/30) (n /125) 22.6 (3.1) 23.0 (14
/26)

p-value 0.435 0.757 0.498 0.089

ESUS, Extended Stroke Unit Service; OSUS, Ordinary Stroke Unit Service; SD, standard deviation. *p B/0.05.

Acute stroke unit care combined with early supported discharge detailed scales in that area may be difficult to carry out in some groups of stroke patients. The global score of NHP is not validated, and its accuracy can therefore be discussed. However, the score is frequently used and the summary is expected to give important information about the total life situation of the patient.16 Some other ESD trials have shown improved QoL for the ESD group over a short time period (three months).4,5 They used QoL scales other than the NHP. Other trials with long-term follow-up have not shown any positive effect in QoL.6
8 Two of the other trials used the NHP.7,9 Only two of the trials had initial stroke unit programmes for all the patients in the study group.8,9 It seems that the support services in follow-up, in addition to stroke unit treatment, are of great importance for successful rehabilitation in a community setting. A meta-analyses of four randomized ESD trials recently concluded that the degree of organization of the service and the teams was of great importance in determining its effectiveness.3 However, the trials show different ways to organize the ESD teams so far, and it is difficult to conclude which one are the most effective. One of the questions with this kind of trial is which components of the present service system cause it to be so successful. We believe that the results depend on the total follow-up service and not any specific part of it. It has been shown that acute stroke unit treatment alone can improve long-term quality of life.23 Furthermore, it has been shown that early discharge from the hospital without enhanced specialist stroke service leads to a significant increase in morbidity.24 Our service with a ‘security system’ during follow-up provided by the mobile team in addition to the focus on rehabilitation in familiar surroundings may allow for an ‘enriched environment’ for both the patients and their caregivers. Besides, the fact that the patients in the ESUS group also achieved a better functional outcome13 may contribute to perceived better QoL and lesser burden for the caregivers. It has been previously described that stroke patients in general have reduced life satisfaction compared with their premorbid state.25 The longterm QoL consequences are insufficiently described, and perhaps more attention should be

585

paid to achieving agreement on a uniform measure of QoL for stroke survivors. The different conclusions of the published trials in this area confirm this apparent need for more research and further definition. This trial is, to our knowledge, the first ESD trial that reduces hospital stay, improves long-term functional outcome and suggests a better longterm QoL for the patients. Similar trials are necessary to confirm the benefit of such a service. Additionally, the economic consequences and resource implications of this service have to be explored.

Clinical messages It seems that early supported discharge service can improve long-term quality of life if: . stroke unit care is the initial treatment and . the follow-up is co-ordinated by a mobile team.

Acknowledgements The work behind this publication has been financed from the Norwegian Foundation for Health and Rehabilitation, and the study has been supported from the Norwegian Department of Health. We want to thank all the staff and patients who participated in this study.

References 1

2 3

4

Stroke Unit Trialists’ Collaboration. A collaborative systematic review of the randomized controlled trial of organized inpatient (stroke unit) care after stroke. BMJ 1997; 314: 1151
/59. Indredavik B, Bakke F, Slørdahl SA, Rokseth R, Ha˚heim LL. Stroke unit treatment: 10-year followup. Stroke 1999; 30: 1524
/27. Early Supported Discharge Trialists. Services for reducing duration of hospital care for acute stroke patients (Cochrane review). The Cochrane Library, Issue 3. Oxford: Update Software 2002. Mayo N, Wood-Dauphinee S, Cote R et al . There’s no place like home. An evaluation of early supported discharge for stroke. Stroke 2000; 31 : 1016
/23.

586 H Fjærtoft et al. 5 Bautz-Holter E, Sveen U, Rygh J, Rodgers H, Wyller TB. Early supported discharge of patients with acute stroke: Disabil Rehabil 2002; 24: 348
/55 6 Rodgers H, Soutter J, Kaiser W et al . Early supported hospital discharge following acute stroke: pilot study results. Clin Rehabil 1997; 11: 280
/87. 7 Rudd AG, Wolfe CDA, Tilling K, Beech R. Randomized controlled trial to evaluate early discharge scheme for patients with stroke. BMJ 1997; 315: 1039
/44. 8 Widen Holmqvist L, Von Koch L, Kostulas V et al . A randomized controlled trial of rehabilitation at home after stroke in southwest Stockholm. Stroke 1998; 29: 591
/97. 9 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
/31. 10 de Haan R, Aaronsen N, Limburg M Langton Hewer R, Van C. Measuring quality of life in stroke. Stroke 1993; 24: 320
/27 11 Aaronsen NK. Quality of life: What is it? How should it be measured? Oncology 1988; 2: 69
/74. 12 Indredavik B, Fjartoft H, Ekeberg G, Løge A, Mørch B. Benefit of an extended stroke unit service with early supported discharge. A randomized, controlled trial. Stroke 2000; 31: 2989
/94 13 Fjærtoft H, Indredavik B, Johnsen R, Lydersen S. Stroke unit care combined with early supported discharge. Long term follow-up of a randomized controlled trial. Stroke 2003; 34: 2687
/92. 14 Ebrahim S, Barer D, Nouri F. Use of Nottingham Health Profile with patients after stroke. J Epidemiol Community Health 1986; 40: 166
/69.

15 16

17 18 19

20 21 22

23

24 25

Thurstone LL: The measurement of values. Chicago, IL: University Chicago Press, 1959. O’Brian BJ, Buxton MJ, Ferguson BA. Measuring the effective heart transplantation programmes: quality of life data and the relationship to survival analyses. J Chronic Dis 1987; 40 (suppl 1): 137s
/ 153s. Holbrook M, Skilbeck CE. An activities index for use with stroke patients. Age Ageing 1983; 12: 166
/ 170. ˚ sberg M, Mongomery SA. A new depression scale A designed to be sensitive to change. Br J Psychiatry 1979; 134: 382
/89. Folstein MF, Folstein SE, McHugh PR. ‘Minimental state’: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189
/98. Robinson BC. Validation of caregivers strain index. J Gerontol 1983; 38: 344
/48. Walker S, Rosser R. Quality of life assessment: key issues in the 1990s. Dordrecht: Kluwer Academic, 1993. Juby LC, Lincoln NB, Berman P. The effect of a stroke rehabilitation unit on functional and psychological outcome. A randomized controlled trial. Cerebrovasc Dis 1996; 6: 106
/10. Indredavik B, Bakke, F, Slørdahl, S, Rokseth R, Ha˚heim L. Stroke unit treatment improves longterm quality of life. A randomized controlled trial. Stroke 1998; 29: 895
/99. Ronning OM, Guldvog B. Outcome of subacute stroke rehabilitation. A randomized controlled trial. Stroke 1998; 29: 779
/84. Suenkeler I, Nowak M, Misselwitz B. Timecourse of health-related quality of life as determined 3,6,12 months after stroke. J Neurol 2002; 249: 1160
/67.