|
Stroke rehabilitation after hospital discharge: a randomized trial comparing domiciliary and day-hospital care |
Roderick P, Low J, Day R, Peasgood T, Mullee M A, Turnbull J C, Villar T, Raftery J |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The use of a new domiciliary rehabilitation service for elderly stroke patients, compared with geriatric day-hospital care.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients with a newly identified stroke, who were admitted to Poole Hospital National Health Service Trust or one of its associated community hospitals, and those with recent strokes directly referred from the community for day-hospital rehabilitation. Patients were excluded if they had terminal illnesses or needed day-hospital for social or medical reasons.
Setting The setting was an institution (day-hospital care) and the community (domiciliary care). The economic study was carried out in the Poole area, East Dorset, which is a mixed urban/rural area on the south coast of England.
Dates to which data relate The effectiveness evidence was gathered from October 1995 to June 1997. The costs were based on 1996 to 1997 prices.
Source of effectiveness data The evidence for the effectiveness data was derived from a single study.
Link between effectiveness and cost data The costing was undertaken prospectively on the same patient sample as that used in the effectiveness study.
Study sample The estimated study size was 128 (64 in each group). The authors deemed this sufficient to demonstrate a two-point difference in activities of daily living (thought to be clinically meaningful) at 5% significance and 80% power, given the distribution of Barthel score (mean 16, standard deviation 1.9) in a prior sample of 27 stroke patients discharged from the Poole day hospital.
From October 1995 to June 1997, 397 patients with stroke were admitted to Poole NHS Trust, of whom 165 (42%) were eligible for the study. A total of 125 (76%) consented and were randomised, plus 15 patients that were recruited from community referrals. The main reasons for refusal were patient preferences for therapy or unwillingness to take part in the trial. There were no differences in the age and gender distribution between those patients who participated and those who refused.
A total of 232 (58%) patients were ineligible for the study. The reasons for this were death (78 patients, 34%), virtual resolution of stroke (36 patients, 16%), disability too severe to benefit from rehabilitation (14 patients, 6%), severe dysphasia (13 patients, 5.5%), frailty (9 patients, 4%), advanced dementia (8 patients, 3%), too confused (2 patients, 1%), non-resident (22 patients, 9%), social referral to day-hospital (5 patients, 2%), age less than 55 years (3 patients, 1%), self-discharge (1 patients, 0.5%), other (4 patients, 2%), and unknown (37 patients, 16%).
Study design This was a randomised controlled trial. The participants were randomised before discharge by calling a central office where closed lists of computed-generated randomisation schedules were kept. Randomisation was stratified by gender, age, disability level (Barthel index <10, 10 to 14 or >= to 15) and day-hospital catchment. This was a single-centre study with associated centres. The duration of follow-up was 6 months. The outcome data were reported for 54 (84%) of the domiciliary patients and 58 (78%) of those in the day-hospital group. There were a few crossovers, 5 from the domiciliary group to day-hospital and 2 from day-hospital to domiciliary care. Most losses to follow-up (16 out of 26) were due to death, a recurrent stroke or the patient moving away. There were no difference in age or initial Barthel index between the participants and those lost to follow-up.
Analysis of effectiveness The analysis of the clinical study was conducted on an intention to treat basis. The primary outcome was functional status, as measured by the change in the Barthel index. The secondary outcomes were mobility, mental state, social activity and generic quality of life. Sociodemographic and clinical information were collected at the study entry. The Barthel index, Rivermead Mobility index, and Philadelphia Geriatric Center Morale Scale (PGCMS) were recorded at entry and at 6 months' follow-up. The Frenchay activities index and perceived quality of life (SF-36) were recorded at 6 months' follow-up. Cognitive status was assessed by the Abbreviate Mental Test.
Effectiveness results The Barthel index and Rivermead Mobility index improved in both groups, but non significant differences were found in favour of the domiciliary group. A subgroup analysis showed that this was confined to those with less disability.
Scores for the Frenchay activities index (low in both groups, indicating poor social function) and SF-36 physical component were consistent with greater physical function in the domiciliary group. The PGCMS fell in both groups, but less so in the domiciliary group. The groups showed no difference in either the SF-36 mental component score or in a composite poor outcome of death, recurrent stroke and a 6-month Barthel score of <14.
Clinical conclusions There were no significant differences in the patient outcomes, although there were non significant improvements in measures of physical function and social activity in the domiciliary group. Neither service influenced the patient's mental state, and their social activity remained low.
Measure of benefits used in the economic analysis No summary benefit measure was used in the economic analysis. A cost-consequences approach was therefore adopted (see the 'Effectiveness Results' section).
Direct costs The costs and the quantities were reported separately. The resources used during the 6-month period were for rehabilitation services, general practitioner visits, district nurse visits, hospital readmissions, hospital outpatient visits, and social services. The quantities were estimated using actual data, which were obtained from the study and collected during the 6-month follow-up period. No discounting was carried out. This was appropriate given that the study period was less than a year (6 months). The NHS trusts costs were based on financial returns, which were supplemented by national unit costs estimates for other services. The authors obtained the unit costs of each of the several day-hospitals within the community trust and averaged them, weighting by activity. Five days of hospital costs were involved and care was co-ordinated by multidisciplinary teams. The domiciliary stroke teams consisted of one whole-time-equivalent senior physiotherapist (grade 1) and 0.5 whole-time-equivalent senior occupational therapist (grade 1). They met daily to plan activity and fortnightly with a consultant geriatrician to review patients, using a goal setting approach. In both situations therapy was provided until maximum potential for recovery was reached. 1996 to 1997 prices were used
Statistical analysis of costs The costs were treated stochastically. The Mann-Whitney U-test was used as the cost data were skewed.
Indirect Costs The indirect costs were not included in the study. The authors stated that "as most patients and their carers were elderly, employment issues were not stated".
Sensitivity analysis The authors performed sensitivity analyses to take account of uncertainty in the estimates of the therapists' workload, domiciliary staffing costs, therapists' travel, domiciliary overheads, and day-hospital and ambulance costs. Full details are available on request from the authors.
Estimated benefits used in the economic analysis No summary benefit measure was used in the economic analysis and, therefore, a cost-consequences approach was adopted. See the 'Effectiveness Results' section.
Cost results There were no significant differences (incremental) in the mean costs per patient for services used to 6 months between the domiciliary and day-hospital groups. The median costs per patient were lower (not significant) in the domiciliary group. The 95% confidence intervals for the difference (of the median) were -298 to 306 for rehabilitation, (p=0.93), -466 to 388 for health service, (p=0.72), and -429 to 975 for health and social services, (p=0.71).
Synthesis of costs and benefits The costs and benefits were not combined. The sensitivity analysis of the rehabilitation inputs showed that the domiciliary arm would have the largest cost advantage if its therapy caseload was increased by one-third, from about three to four patient visits per day. Even for scenarios most favourable to day-hospital, the median health service cost per patient in the 6-month follow-up was still lower in the domiciliary group. The type of sensitivity analysis was not stated in the text
Authors' conclusions No significant differences were detected in the effectiveness of the two services. Neither service influenced the patients' mental state, and their social activity remained low. The total costs were similar. A mixed model of day-hospital and domiciliary care may be the most cost-effective approach for community stroke rehabilitation, although this requires further research.
CRD COMMENTARY - Selection of comparators The comparator used (continue rehabilitation after stroke in a day-hospital) was justified on the grounds that it was current practice by the time of the study. You should decide if this is a widely used health technology in your own setting.
Validity of estimate of measure of effectiveness The analysis used a randomised controlled trial, which was appropriate for the study question. The study sample was representative of the study population. However, as the authors acknowledged, the sample size was too small conclusively to identify differences in rare adverse health events. The patient groups were shown to be comparable at analysis. The authors also stated that the study proved to be under-powered for the diversity of outcomes, especially after the losses to follow-up.
Validity of estimate of measure of benefit The authors did not derive a summary measure of health benefit. The analysis was therefore categorised as a cost-consequences analysis.
Validity of estimate of costs The perspective of the economic analysis was the health service and local authority social service. All the categories of costs relevant to this perspective appear to have been included in the analysis. The resources were reported, but the prices were not. A statistical analysis of the quantities was performed to take into account the skewed distribution of the cost data. Both the mean and median per patient costs were reported.
Other issues The authors made appropriate comparisons of their findings with those from other studies. The issue of generalisability was partially addressed in these comparisons. The study groups were older than those in the other studies. Consequently, there was evidence of increased social service costs with respect to other studies. The authors reported a number of limitations of their study. First, the sample size was too small to identify conclusive differences in rare adverse health events. Second, the nature of the therapy differed, there being more carer involvement in the domiciliary group and more group therapy in the day-hospital. Third, the study proved under-powered for the diversity of outcomes, especially after losses to follow-up. Finally, the trial could not address the marginal cost differences arising from switching from a day-hospital to domiciliary care model.
Implications of the study A mixed model of post-discharge rehabilitation for elderly stroke patients may be appropriate. Such models need to be evaluated. The optimum duration of post-discharge rehabilitation and the relative contributions of different therapists remain outstanding research questions. Effective interventions to improve the mental health and social activity of patients are still required.
Source of funding Funded by the South and West Research and Development Directorate.
Bibliographic details Roderick P, Low J, Day R, Peasgood T, Mullee M A, Turnbull J C, Villar T, Raftery J. Stroke rehabilitation after hospital discharge: a randomized trial comparing domiciliary and day-hospital care. Age and Ageing 2001; 30(4): 303-310 Other publications of related interest Colin C, Wade DT, Davies S, Horne V. The Barthel index: a reliability study. International Disability Studies 1988;10:61-3.
Collen FM, Wade DT, Robb GF, Bradshaw CM. The Rivermead Mobility Index: a further development of the Rivermead Motor Assessment. International Disability Studies 1991;13:50-4.
The Royal College of Physicians of London, British Geriatrics Society. Standardised assessment scales for elderly people. London: the Royal College and the Society; 1992.
Wade DT, Leigh-Smith J, Langton Hewer R. Social activities after stroke: measurement and natural history using Frenchay Activities Index. International Rehabilitation Medicine 1985;7:176-81.
Brazier JE, Harper R, Jones NM, et al. Validating the SF-36 health survey questionnaire: new outcome measure for primary care. BMJ 1992;305:160-4.
Indexing Status Subject indexing assigned by NLM MeSH Aged; Aged, 80 and over; Day Care, Medical; Female; Follow-Up Studies; Health Services for the Aged /statistics & Home Care Services; Humans; Male; Middle Aged; Patient Discharge; Social Work /statistics & Stroke /rehabilitation; Treatment Outcome; numerical data; numerical data AccessionNumber 22001007833 Date bibliographic record published 30/06/2003 Date abstract record published 30/06/2003 |
|
|
|