BMJ No 7115 Volume 315
Papers Saturday 25 October 1997
Randomised controlled trial to evaluate early discharge scheme for patients with stroke
Anthony G Rudd, Charles D A Wolfe, Kate Tilling, Roger Beech
Abstract
Objective: To assess the clinical effectiveness of an early discharge policy for patients with stroke by using a community based rehabilitation team.Design: Randomised controlled trial to compare conventional care with an early discharge policy.
Setting: Two teaching hospitals in inner London.
Subjects: 331 medically stable patients with stroke (mean age 71) who lived alone and were able to transfer independently or who lived with a resident carer and were able to transfer with help.
Interventions: 167 patients received specialist community rehabilitation for up to 3 months after randomisation. 164 patients continued with conventional hospital and community care.
Main outcome measures: Barthel score at 12 months. Secondary outcomes measured impairment with motoricity index, minimental state examination, and Frenchay aphasia screening test; disability with the Rivermead activity of daily living scales, hospital anxiety and depression scale, and 5 m walk; handicap with the Nottingham health profile; carer stress with caregiver strain index and patient and carer satisfaction. The main process measure was length of stay after randomisation.
Results : One year after randomisation no significant differences in clinical outcomes were found apart from increased satisfaction with hospital care in the community therapy group. Length of stay after randomisation in the community therapy group was significantly reduced (12 v 18 days; P<0.0001). Patients with impairments were more likely to receive treatment in the community therapy group.
Conclusions: Early discharge with specialist community rehabilitation after stroke is feasible, as clinically effective as conventional care, and acceptable to patients. Considerable reductions in use of hospital beds are achievable.
Introduction
Stroke accounts for the use of a fifth of acute medical beds and a quarter of long term beds.(1) In the United Kingdom up to 70% of patients are admitted to hospital in inner city districts,(2) with the mean length of stay varying from 11.0 to 38.9 days around Europe.(3) The cost to the NHS of acute stroke care has been estimated at £4626 per discharge,(4) with a prediction that the total costs of stroke care will rise in real terms by 30% between 1991 and 2010 because of the effects of an aging population.(5) Hospital and community care is often fragmented and haphazard,(6) despite there being clear evidence of the benefits of coordinated care.(7)
It has been argued that a more balanced approach to stroke care between hospital and community should be adopted.(8) Care in the community may be more acceptable to patients and gives the opportunity to deal with psychosocial issues and handicap.(9) Earlier discharge from hospital with rehabilitation provided at home may enable these to be handled more effectively than with conventional care and may provide the opportunity for financial savings.
By using a rehabilitation team in the community we evaluated the effect of early discharge compared with conventional care on disability at one year in patients who had been admitted to hospital after a stroke. The resource implications of this policy were also investigated. The trial was designed to be applicable to most patients with stroke admitted to a district general hospital. No attempt was made to avoid initial hospital admission because it was considered impracticable to establish appropriate packages of health and social care quickly enough to prevent admission for patients who often live alone in poor quality housing in this inner city district.
Patients and methods
Case ascertainment and entry criteria
A hospital based stroke register was maintained at St
Thomas's and King's College Hospitals, London, between January 1993
and July 1995. The World Health Organisation's definition of stroke
was used.(10) Twice weekly checks of the wards were
undertaken by two dedicated research associates with nursing training.
Patients were managed in the normal way until the criteria were
fulfilled for entry into the trial. If patients lived alone they needed
to be able to perform functional independent transfer, and if they
lived with a willing carer they needed to be able to perform transfer
with assistance. The point at which these criteria were met was decided
after consultation with the hospital physiotherapist. All patients were
assessed within one working day of notification by a consultant
physician or at King's College Hospital by a medical registrar, who
obtained informed consent from the patient or next of kin when the
patient had cognitive impairment or dysphasia. Baseline data were
collected after randomisation by a researcher blinded to which arm of
the trial the patient was in. Ability to transfer did not necessarily
imply that discharge could take place immediately. The rehabilitation
team responsible for the patient was notified that the community team
was available to provide care after discharge and it was left to the
team to plan discharge whenever feasible. The study obtained approval
from the local ethics committees. Patients were excluded if they lived
too far away for the team to visit.
Randomisation
After consent randomisation was restricted in permuted
blocks of 10 with random number tables provided in blank sealed opaque
envelopes. Patients were allocated to continue conventional treatment
or to receive community treatment. No stratification was
performed.
Interventions
Patients allocated to conventional care continued with their
treatment, discharge planning, and outpatient care in the normal way.
Stroke care at both hospitals is similar and well coordinated. About
half of the patients who are admitted receive treatment in a stroke
unit, with the remainder being treated in general medical or elderly
care wards. Outpatient resources available to them included a hospital
based stroke clinic, geriatric day hospital, generic domiciliary
physiotherapy and speech and language therapy, hospital outpatient
physiotherapy, and the usual community resources. The maximum level of
home care available in the study area to all patients was three one
hour visits daily by a home help for personal care, meals on wheels,
and community nurse visits for specific tasks.
Patients randomised to the community therapy team remained in hospital until the required package of social services care could be organised and any home adaptations undertaken. A store of commodes, high chairs, and toilet frames was kept by the team to expedite discharge. The patients were assessed for rehabilitation needs before discharge in conjunction with the hospital based therapists to set initial objectives and to ensure continuity of care. After discharge, patients were given a planned course of domiciliary physiotherapy, occupational therapy, and speech therapy, with visits as frequently as considered appropriate (maximum one daily visit from each therapist). Each patient had an individual care plan, which was reviewed at a weekly team meeting. Patients received care from the team for a maximum of three months. On discharge patients were referred to conventional services when appropriate. All other services apart from therapy were as described for the conventional group. There was no augmentation of social services resources.
The community therapy team comprised a senior physiotherapist grade 1 with neurological training, a senior occupational therapist grade 1, a half time speech and language therapist with adult neurological training, and a full time therapy aide. A consultant physician (AGR) coordinated the team and chaired the weekly clinical meeting.
Assessment of outcome
Baseline data on demography and severity of stroke were collected
within 48 hours of randomisation. Two research associates conducted all
the assessments at discharge and at 2, 4, and 6 months. A third
research associate conducted the assessment at 12 months by interview,
usually in the patient's home, and she was blind to the treatment
group, although occasionally the patient did make clear which group
they had been in. No evaluation of the efficacy of the blinding
procedure was performed.
A range of standardised outcome assessments was used to measure aspects of impairment, disability, and handicap and could be completed in most cases in 45 minutes.(11)
Impairments were assessed with the following tests.
Motoricity index, which is a measure of limb function with a maximum score of 100 for normal subjects. Severe paralysis is defined as a score of 0-32, moderate as 33-64, and mild as 65-99.
Minimental state examination - Severe cognitive impairment is defined as a score of 23 or less out of a total of 30.
Frenchay aphasia screening test is a screening instrument to detect aphasia in patients with stroke. A score of 13 or less out of 20 indicates aphasia.
Disability was measured with the following tests.
Modified Barthel score - To try to overcome the recognised differences between the intervals at the lower, middle, and upper ends of the scale, analysis was also performed by dividing the scale into 0-14 (severe disability), 15-19 (moderate disability), and 20.
Rivermead activities of daily living score has been validated for use in elderly patients with stroke. Scores range from 15-45, with 15 indicating dependence in all measured activities.
Hospital anxiety and depression scale is a 21 point scale,
with a cut off point of greater than 10 identifying those with a high
probability of the disorder and 8-10 being borderline. It is designed
for self assessment, c The 5 metre timed walk - In a trial of late physiotherapy
after stroke Wade et al used a 10 metre walk,(12) suggesting
gait speed offers a simple and s Nottingham health profile was used to assess subjective
health status across six domains: energy, pain, emotion, sleep, social,
and physical mobility. The maximum total score is 45, with a high score
indicating poor health status.
Caregiver strain index measured carers' stress, the
index having 13 items. Overall stress score was obtained by summing the
subjects' ratings across all the items.
Patient and carer satisfaction was assessed by using stroke specific
questionnaires. Analysis was performed separately for questions
relating to hospital care, rehabilitation, and home care.
To evaluate the differences in utilisation of services, length of stay,
number of therapy sessions with one unit being defined as a 20 minute
session per therapist, hospital readmission, and living conditions on
discharge and at one year were documented.
Statistical analyses
Randomisation was between 27 January 1993 and 19 July 1995.
Of 660 patients with stroke admitted to St Thomas's Hospital, 300
were randomly allocated treatment (figure). Two refused to participate.
Of the 358 who were not randomly allocated treatment, 160 died in
hospital, 76 lived out of the area, 3 were of no fixed abode, and 119
failed to meet the entry criteria by never achieving independent
functional transfers. At King's College Hospital 31 patients were
randomly allocated treatment.
Details of randomisation and outcome.
Table 1 describes the baseline characteristics assessed on admission.
There were no significant differences between the two arms of the
trial. Table 2 shows outcome at discharge from hospital. For the
Nottingham health profile the reasons for failing to complete the score
were refusal (11 patients), cognitive impairment (11), aphasia (21),
inability to speak English and problems with literacy (9), and death
before assessment (5). The incidence of missing data was not
significantly different between the groups. Reasons for failing to
complete the hospital anxiety and depression scale and the Frenchay
aphasia screening test were similar.
Table 3 describes the outcomes at one year. No significant differences
between the two groups were found on any of the measures, including the
individual scores on the Nottingham health profile subcategory. Patient
satisfaction for the care given in hospital was greater in the
community therapy group
(P=0.032).
Table 4 gives the number of patients lost to follow up, survival to one
year, recurrence of stroke, and readmission rates. Five patients in the
community therapy group were lost to follow up due to refusal to
participate further and emigration. Four of the conventional treatment
group were lost to follow up at 1 year. No differences in mortality
between the groups were observed, and there were no differences in
readmission rates.
Table 5 details the principle services used during the study period.
While differences in the distribution of provision of inpatient and
outpatient therapy are shown, the total therapy provided did not differ
significantly between the groups.
To our knowledge, this is the first randomised controlled
trial of early discharge of patients with stroke to a team of
therapists in the community. The trial was conducted in an inner city
district, and the findings should be applicable to other cities. The
study sample represented about half of the patients with stroke
admitted over the study period and three quarters of the patients with
stroke who did not die in hospital. The pragmatic design of the trial
with simple entry criteria and recruitment from two hospitals with
patients on a range of types of ward makes it likely that the results
are generalisable to other units. Assessment of outcome at discharge
from hospital indicated no significant differences, although the length
of stay after randomisation to the community therapy arm was
significantly shorter.
Methodological considerations The overall survival rates were similar in both groups, and early
discharge did not result in greater rates of institutionalisation or
hospital readmission. Although the total therapy for each discipline
provided was not significantly different between the two groups for
those having therapy, provision of therapy for patients with an
impairment was better in the patients treated in the community,
possibly reflecting better assessment and team working.
Nine patients (2.7%) were lost to follow up at 1 year, mainly because
they had left the country. Completion rates for the outcome assessments
were low because of a combination of cognitive and language problems in
the study population but were similar to those achieved in a comparable
group of patients with stroke.(16) A considerable proportion
of the patients (26%) were living alone at 1 year and were without a
principal carer able to give information. Completion rates for the
scales and follow up rates were not significantly different between the
two groups and are unlikely to have affected the final results.
Comparison with other studies It has been argued that admission to hospital at all for patients with
stroke is unnecessary except for a small proportion of
patients.(20) This is not the view taken by a WHO consensus
meeting(21) and is clearly not believed by most of the
general practitioners who admitted up to 70% of their patients in
London.(2) The only non-randomised trial to provide
intensive domiciliary rehabilitation as an alternative to hospital
admission for stroke failed to reduce admission rates.(22)
Pound et al found that patients with stroke have important psychosocial
needs during the acute stage of stroke that are often met by hospital
admission.(23)
Conclusions
Elderly
Care Unit, Division of Public Health Sciences,
Correspondence to: Dr
Rudd
The members of the St Thomas's Stroke Rehabilitation Trial
Research Group are G Browning, F Bunn, H Cambell, T Colella, V
Churchouse, V James, S Mackay, A Moberley, P Pound, E Richardson,
P Trail, Y Toun, F Warburton, and C Watson. We acknowledge the help and
support of our colleagues at St Thomas's and King's College
Hospitals and Dr M Andlaw, general practitioner.
Funding: The Stroke Association, Lambeth, Southwark and
Lewisham Health Authority, the Special Trustees of St Thomas's
Hospital, the Nuffield Provincial Hospitals Trust, Wandsworth Health
Gain Fund.
(Accepted 26 August 1997)
References
1 Wade D T. Stroke (acute cerebrovascular disease). In:
Stevens A, Rafferty J, eds. Health care needs
assessment. Vol 1. Oxford: Radcliffe Medical Press,
1994:111-255.
2 Wolfe C D A, Taub, Woodrow E, Richardson E, Warburton F G, Burney
PGJ. Patterns of acute stroke care in three districts of southern
England. J Epidemiol Community Health
1993;47:144-8.
3 Beech R, Ratcliffe M, Tilling K, Wolfe C D A. Hospital services
for stroke care: a European perspective. Stroke
1996;27:1958-64.
4 Isard P A, Forbes J F. The cost of stroke to the National Health
Service in Scotland. Cerebrovascular Disease
1992;2:47-50.
5 Bergman L, van der Meulen J H P, Limburg M, Halbema T D F. Costs
of medical care after first-ever stroke in the Netherlands.
Stroke 1995;26:1830-6.
6 Lindley R I, Amayo E O, Marshall J, Sandercock P A G, Dennis M,
Warlow C P. Hospital services for patients with acute stroke in the
United Kingdom: the Stroke Association survey of consultant opinion.
Age Ageing 1995;24:525-32.
7 Stroke Unit Trialists' Collaboration. Collaborative
systematic review of the randomised trials of organised inpatient
(stroke unit) care after stroke. BMJ 1997;314:1151-8.
8 Young J. Is stroke better managed in the community?
BMJ 1994;309:1356-8.
9 Forster A, Young J. Stroke rehabilitation: can we do better?
BMJ 1992;305:1446-7.
10 Hatono S. Experience from a multicentre stroke register. A
preliminary report. Bull WHO 1976;54:541-53.
11 Stojcevic N, Wilkinson P, Wolfe C. Outcome measurement in
stroke patients. In: Wolfe C, Rudd T, Beech R, eds. Stroke
services and research. London: The Stroke Association, 1996.
12 Wade D T, Collen F M, Robb G F, Warlow C P. Physiotherapy
intervention late after stroke and mobility. BMJ
1992;304:609-13.
13 Brooten D, Roncoli M, Finkler S, Arnold L, Cohen A, Mennuti M.
A randomised trial of early hospital discharge and home follow-up of
women having caesarean birth. Obstet Gynecol
1994;84:832-8.
14 Adler M W, Waller J J, Kasap H S, King C, Thorne S C. A randomised
controlled trial of early discharge for inguinal hernia and varicose
veins: some problems of methodology. Med Care
1974;12:541-7.
15 Adler M W, Waller J J, Creese A, Thorne S C. Randomised controlled
trial of early discharge for inguinal hernia and varicose veins.
J Epidemiol Community Health 1978;32:136-42.
16 Topol E J, Burek K, O'Neill W W, Kewman D G, Kander N H, Shea M J,
et al. A randomised controlled trial of hospital discharge three days
after myocardial infarction in the era of reperfusion. N Engl J
Med 1988;318:1083-8.
17 Townsend J, Piper M, Frank A O, Dyer S, North W R S, Meade T W.
Reduction in hospital readmission stay of elderly patients by a
community based hospital discharge scheme: a randomised controlled
trial. BMJ 1988;297:544-7.
18 Martin F, Oyewole A, Maloney A. A randomised controlled trial
of a high support hospital discharge team for elderly people.
Age Ageing 1994;23:228-34.
19 Clarke A. Why are we trying to reduce length of stay?
Evaluation of the costs and benefits of reducing time in hospital must
start from the objectives that govern change. Quality in Health
Care 1996;5:172-9.
20 Wade D T, Langton Hewer R. Management of stroke: home or
hospital? Hosp Update 1986:427-9.
21 Aboderin I, Venables G, on behalf of Pan European Consensus
Meeting on Stroke Management in Europe. J Intern Med
1996;240:173-80.
22 Wade D T, Langton-Hewer R, Skilbeck C E, Bainton D, Burns-Cox C.
Controlled trial of a home care service for acute stroke patients.
Lancet 1985;i:323-6.
23 Pound P, Bury M, Gompertz P, Ebrahim S. Stroke patients' views
on their admission to hospital. BMJ 1995:311:18-22.
The associations between the normally distributed continuous
variables and treatment group were examined by using the unpaired
t test. When the continuous variables showed evidence of
a non-normal distribution, the Mann-Whitney test was used and the
confidence interval calculated for the difference between medians. The
chi2 test was used to investigate the associations between
the categorical variables and the treatment group. For ordered
categorical variables the chi2 test for trend was used. The
sample size of 130 per group at one year has a power of 80% at the 5%
level of detection for a 3.5 point difference in Barthel score based on
a standard deviation of 10.Results
Table 1 - Baseline comparisons between groups of
patients with stroke according to place of rehabilitation. Values are
numbers (percentages) of patients unless stated otherwise
Variable
Community therapy
(n=167)Conventional
(n=164)
Sex: Male 92
(55)93
(57) Female 75
(45)71
(43) Age
(years):Mean
(SD)70
(11)72
(12) Median
(range) 70
(27-96) 73
(42-103) Ethnic
group: White 124
(74)122
(74) Black Caribbean 26
(16)28
(17) Black African 10
(6)4
(2) Other 7
(4)10
(6) Recurrent stroke 2
(1)2
(1) Previous Barthel
score: 0-14 12
(7)8
(5) 15-19 29
(17)41
(25) 20 126
(75)115
(70) Length of stay to randomisation
(days):Mean
(SD)22
(25)25
(30)
Median
(range) 13
(1-127) 13
(0-30) Not
known 1 0 Living conditions before
stroke: Living
alone 51
(31)62
(38) Living with
another person 104
(62)87
(53)
Institution/other 12
(8)15
(9)
Barthel score at
randomisation: 0-14 70
(42)65
(40) 15-19 78
(47)78
(48) 20 17
(10)20
(12) Not
known 2 1 Dysphasia 68
(42)57
(35) Incontinence 47
(29)46
(29)
Table 2 - Outcome assessments at discharge from hospital
of patients with stroke according to place of rehabilitation
Detail
Community therapy
(n=167)Conventional
(n=164)P
value
95% CI for difference between groups
Barthel activities of daily living Mean
(SD)15
(4)15
(4)Median
(range) 16
(1-20) 15
(4-20)0.94*
(-1 to 1) Not
known 2 0 Frenchay
aphasia Mean
(SD) 18
(8.4) 19
(8.3)0.92†
(-2.0 to 1.9) Median
(range) 21
(0-20) 21
(0-30)Not
assessed 20 23 Hospital
anxiety and depression scale No (%) with
anxiety: Normal 89
(70)106
(82)0.02‡ Borderline 18
(14)14
(11)Abnormal 20
(16)10
(8)Not
assessed 40 34 No
(%) with
depression:Normal 88
(70)93
(73)0.8‡ Borderline 25
(20)21
(16)Abnormal 13
(10)14
(11)Not
assessed 41 36 Minimental
state examination Mean
(SD) 21
(7.1) 21
(7.3)0.70†
(-1.2 to 2.1) Median
(range) 23
(0-30) 23
(0-30)Not
assessed 27 23 Motoricity Mean
(SD)83
(13)85
(12)0.15†
(-5.1 to 0.6) Median
(range) 87
(41-100) 86
(44-100)Not
assessed 15 10 Five
metre timed walk Mean
(SD)
(secs)15
(23)17
(35)Median
(range) 10
(4-180) 10
(5-360)0.08*
(-2 to 0) Not
assessed 65 51 Total
Nottingham health profile Mean
(SD)11
(7)10
(7)Median
(range) 11
(0-37) 9
(0-31)0.10*
(0 to 3) Not
assessed 32 33 No
(%) discharged from Acute medicine 53
(32)49
(30)0.65** Elderly
care 15
(9)12
(7)Stroke unit 78
(47)87
(53)Other 21
(13)16
(10)
*Mann-Whitney test.
†Unpaired t test.
‡chi2 test for trend.
Table 3 - Outcome assessments 1 year after stroke
according to place of rehabilitation
Detail
Community therapy
(n=136)Conventional
(n=126)P
value
95% CI for difference between two groups
Barthel activities of daily living Mean
(SD)16
(4)16
(4)Median
(range) 18
(2-20) 18
(3-20)0.30* 0 to 1 Not
assessed 1 0 No
(%) with grouped Barthel score:0-14 30
(22)35
(28)15-19 60
(44)56
(44)0.23† 20 45
(33)35
(28)Frenchay
aphasia Mean
(SD)22
(8)23
(7)0.99‡ -2.5 to 1.6 Median
(range) 25
(0-30) 26
(1-30)Not
assessed 28 24 Hospital
anxiety and depression scale No (%) with
anxiety: Normal 79
(69)79
(81)0.02† Borderline 16
(14)12
(12)Abnormal 20
(17)7
(7)Not
assessed 21 28 No
(%) with
depression: Normal 70
(61)59
(60)0.9† Borderline 20
(18)19
(19)Abnormal 24
(21)21
(21)Not
assessed 22 27 Minimental
state examination Mean
(SD)25
(6)25
(6)0.86‡ -1.5 to 1.7 Median
(range) 27
(8-30) 27
(5-30)Not
assessed 20 20 Motoricity Mean
(SD)88
(13)87
(14)0.80‡ -2.7 to 9 Median
(range) 91
(36-100) 90
(21-100)Not
assessed 5 1 Five
metre timed walk Mean
(SD)12
(6)12
(8)Median
(range) 10
(6-40) 9
(6-70)0.34* 0 to 1 Not
assessed 37 36 Total
Nottingham health profile Mean
(SD)14
(9)12
(8)Median
(range) 13
(0-42) 12
(0-36)0.11* 0 to 4 Not
assessed 18 21 Rivermead
activities of daily living scale Mean
(SD)27
(12)27
(11)Median
(range) 25
(15-45) 25
(15-45)0.93* -2 to 3 No (%) with
score 15 28
(21)28
(23)0.71** -12% to 8% No (%) with
score >15 104
(79)93
(77)Not
assessed 4 5 Caregiver
strain Mean
(SD)5
(4)4
(3)Median
(range) 5
(0-12) 3
(0-12)0.14* 0 to 2 Not
assesed 61 67 No
(%) of patients satisfied with Hospital care 78
(79)59
(65)0.032s 1% to
27% Therapy provision 56
(58)46
(51)0.29s -6% to 22% Community
support 44
(56)35
(50)0.44s -10% to 22%
General 58
(59)43
(48)0.14s -34% to 25%
No (%) of carers satisfied with Hospital
care 60
(74)41
(67)0.37s -8% to 22% Therapy
provision 40
(53)28
(46)0.39s -9% to 24% Community
support 28
(42)29
(51)0.35s -26% to 9%
General 68
(83)52
(83)0.95s 13%
*Mann-Whitney test.
†chi2 test for trend.
‡Unpaired t test.
**chi2 test.
Table 4 - Numbers (percentages) of patients with stroke
who were eligible and followed up, according to place of rehabilitation
Detail
Community therapy
(n=167)Conventional
(n=164)P value
(chi2 test)95% CI for % difference
between groups Dead by 12
months 26
(16)34
(21)0.22
(-13 to 3) Followed up
at 12 months alive 136
(81)126
(77)0.83
(-2 to 16) Lost to follow
up 5
(3)4
(2)0.83
(-3 to 4) Stroke
recurrence 10
(6)5
(3)0.20
(-2 to 7) Readmitted to
hospital 44
(26)42
(26)0.20
(-9 to 10)
Table 5 - Resources used by patients with stroke over 12
month trial period according to place of rehabilitation
Detail
Community therapy
(n=167)Conventional
(n=164)P
value
95% CI for difference between groups†
Length of stay from randomisation Mean
(SD)12
(19)18
(24)Median
(range) 6
(0-149) 12
(0-236)0.0001*
(-6 to -2) Not
known 2 1 Total
No of bed days (from
randomisation) 1,956 2,854 Physiotherapy
units† No (%) with motor deficit 151
(90)144
(88)No (%)
having inpatient physiotherapy 121
(80)111
(77)0.52‡
(-6% to 12%) Mean
units
(SD)10.8
(14.3)17.3
(23.8)Median units
(range) 6
(1-107) 10
(1-153)0.01*
(-4 to 0) Total
inpatient
units** 1,309 1,923 No
(%) having outpatient physiotherapy114
(75)28
(19)0.001‡
(45% to 67%) Mean
units
(SD)21
(24)18
(18)Median
(range) 11
(1-140) 10
(1-74)0.998*
(-4 to 5) Total
outpatient
units** 2,387 496 Median
total therapy
(range) 14
(1-156) 12
(1-153)0.07*
(0 to 6)
Occupational therapy units† No (%) with
functional deficit 151
(90)144
(88)No (%) having inpatient
occupational therapy 97
(64)104
(72)0.14‡
(-19% to 3%) Mean
(SD)9
(10)22
(21)Median
(range) 8
(1-74) 17
(1-105)0.0001*
(-12 to -5) Total
inpatient
unitss 916 2,292 No
(%) having outpatient occupational therapy125
(83)15
(10)0.001
(61% to
84%) Mean units
(SD)22
(24)33
(35)Median
(range) 12
(2-156) 15
(3-119)0.19*
(-18 to 2) Total
outpatient
units 2,740 498 Median
total therapy
(range) 17
(2-156) 18
(1-223)0.43*
(-2 to 5)
Speech therapy units† No (%) with functional
deficits
(speech/swallowing)119
(71)113
(69)No (%) having inpatient speech
therapy 42
(35)35
(31)0.46‡
(-8% to 16%) Mean
(SD)8
(8)14
(13)Median
(range) 5
(1-47) 9
(1-64)0.009*
(-7 to -1) Total
inpatient
unitss 319 491 No
(%) having outpatient speech therapy64
(54)8
(7)0.001
(35% to
59%) Mean units
(SD)21
(22)17
(29)Median
(range) 13
(1-90) 6
(2-89)0.29*
(-3 to 15) Total
outpatient
units 1,327 139 Median
total therapy
(range) 15
(1-100) 9
(1-89)0.11*
(-1 to 9) No
(%) according to living conditions at 1 year No of
patients 135 126 0.27‡ Alone 37
(27)31
(25)With
others 84
(62)70
(56)Sheltered
accomodation 6
(4)8
(6)Institution 8
(6)15
(12)In
hospital 0 2
(2)No (%) using home
help Yes 46
(34)47
(37)0.45‡
(-15% to 8%) Not
known 2 0 No
(%) using meals on wheelsYes 17
(12)14
(12)0.79‡
(-7% to 9%) Not
known 2 4
*Mann-Whitney test.
†One unit of therapy=20 minutes of
therapist's time. ‡chi2 test.
**Inpatient therapy
units=therapy given between randomisation and discharge. Discussion
One year outcome measures support the notion that early discharge
to a rehabilitation team in the community has no adverse effect on
outcome. The Barthel score is the most widely used measure in stroke
rehabilitation trials and was used in this study because of the
importance of longer term disability. It could be argued that the
intervention was more likely to affect the Rivermead at one year. A
post hoc power calculation for the Rivermead activities of daily living
score, with about a quarter of patients scoring the minimum score (15)
at 1 year, showed a power of 80% to detect more than 41% or less than
12% of the early discharge group scoring 15. Satisfaction with care
was not significantly different between the two groups from the
perspectives of carers or patients, although patients in the community
therapy group were more satisfied with the care provided in the
hospital. The longer patients stay in hospital the more chance they
have of being exposed to aspects of care that dissatisfy them.
Alternatively the higher dissatisfaction among the patients receiving
conventional care reflected the fact that they were exposed to a
rehabilitation philosophy within the context of an acute hospital.
Patients may expect to be cared for by nurses and not to be encouraged
to independence, while at home the need for independence may be more
evident.
Early discharge schemes have been developed for surgical patients
after caesarean section,(13) varicose vein and hernia
operations,(14-15) and myocardial
infarction.(16) The effects on outcome in these studies were
small. Augmented support hospital discharge teams(17-18)
reduce readmissions to hospital of elderly patients, thereby saving
hospital resources. The overall conclusion of Clarke's literature
review was that with the very limited evidence reductions in length of
stay have not had a major effect on health outcome,(19)
although the outcomes measured may not have been relevant to the
intervention.
Provision of a specialist team of therapists led by a consultant
provides a model of care for patients with stroke which is as effective
as conventional hospital based care, with a significant reduction in
hospital stay. As the numbers of patients with stroke increase with the
aging population the only alternative to building more acute hospitals
may be to develop alternative community based specialist
care.
Key Messages
Early discharge from hospital after stroke with specialist
rehabilitation at home is feasible without an increase in readmission
rates or stress to carers This randomised controlled trial shows this method to be as
effective as conventional care when assessed with a range of measures
of impairment, disability, handicap, carer stress, and patient and
carer satisfaction at 1 year Significant reductions in bed usage can be achieved by the
provision of a community rehabilitation team with no significant
increase in rehabilitation services
United Medical and Dental Schools of Guy's and St
Thomas's Hospitals,
St Thomas's Hospital,
London SE1
7EH
Anthony G Rudd, consultant
physician
United Medical
and Dental Schools of Guy's and St Thomas's Hospitals,
St
Thomas's Hospital,
London SE1 7EH
Charles D A
Wolfe, senior lecturer
Kate
Tilling, lecturer
Roger
Beech, senior research
fellow
Conflict of interest: None.
Home | Current contents | Past issues |
Classified ads | Career Focus | Feedback
Collections |
About this site | About the BMJ | BMA | Medline