|
Occupational therapy compared with social work assessment for older people: an economic evaluation alongside the CAMELOT randomised controlled trial |
Flood C, Mugford M, Stewart S, Harvey I, Poland F, Lloyd-Smith W |
|
|
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 An occupational therapy-led assessment and a social worker-led assessment of older people were examined.
Study population The study population comprised patients older than 65 years who were living in their own homes, and their carers. Participants were excluded only when they required an urgent response. Individuals with dementia were eligible providing they had an informal carer able to give consent.
Setting The setting was likely to have been community care. The economic study was conducted in the UK.
Dates to which data relate The dates to which the effectiveness and resource use data referred were not reported in this paper. Further details were provided in the parent study (Stewart et al. 2005, see 'Other Publications of Related Interest' below for bibliographic details). The analysis used 2001 prices.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out prospectively alongside the clinical trial. With the exception of the drug costs, which were based on a sub-sample of one-third of the participants, the same patients provided the effectiveness data and the cost data,.
Study sample The method used to derive the study sample was not reported in the present study. The following information was derived from the parent clinical study, the CAMELOT trial (Stewart et al. 2005).
For 80% statistical power to show a clinically important difference of 8 points in the Community Dependency Index (CDI), based on a two-tailed test (alpha = 0.05), the trial aimed to include 300 participants (150 in each arm). Recruitment was higher on the assumption that 30% of the participants would be lost to follow-up.
A total of 881 patients were eligible for the study question and were contacted by letter between February 2000 and March 2001. Of these, 560 declined to enter the study and 321 were included in the analysis. These 321 participants were randomised to two groups, 160 (62.5% female) received the occupational therapist-led assessment (OT group) and 161 (65.2% female) received the social worker-led assessment (SW group). The mean age of the participants was 81.0 (+/- 7.0) years in the OT group and 81.6 (+/- 7.5) years in the SW group.
In addition, a total of 113 carers were included in the analysis. There were 49 carers (65.3% female) in the SW group and 64 (65.6% female) in the OT group. The mean age of the carers was 69.8 (+/- 12.1) years in the SW group and 68.3 (+/- 14.2) years in the OT group. Further details were given in the parent study (Stewart et al. 2005).
Study design The study was a community-based, single-blind, randomised controlled trial that was conducted over 2 years and 4 months. The randomisation sequence was generated using standard random number tables. A study researcher, who remained blind to group assignment, conducted the follow-up at 4 and 8 months. Thirty participants in the SW group and 27 participants in the OT group were lost to follow-up at 8 months.
Analysis of effectiveness The analysis was performed on the basis of treatment completers (at 8 months follow-up, there were 133 participants in the OT group and 131 in the SW group). The primary health outcomes assessed were:
the CDI score (range: 0 - 100; higher scores denote greater independence), and
the EQ-5D score, which provided a weighted health index (range: -0.59 - 1; higher scores denoting greater well-being).
The CDI score measured the individual's ability within their environment to carry out a range of daily living tasks relating to self-care. In the parent study, the groups were shown to be similar in terms of the demographic baseline characteristics.
Effectiveness results The EQ-5D and CDI scores in each group were not reported. Only the difference in EQ-5D and CDI scores between baseline and 8-month follow-up data were reported.
The difference in EQ-5D score was -0.06 in the SW group and -0.05 in the OT group.
The difference in CDI score was 3.70 in the SW group and 2.29 in the OT group.
For carers, the OT group had a significantly better quality of life scores (these results were not shown in the present study).
No statistical analysis on the health outcomes was reported.
Clinical conclusions There was no statistically significant difference between the SW and the OT interventions in terms of the clinical outcomes for participants.
Measure of benefits used in the economic analysis The outcome measure used was the quality-adjusted life-years (QALYs) gained associated with each strategy. The utility weights were derived from the effectiveness analysis.
Direct costs The direct costs consisted of both health service-related costs and patient costs. The health service costs concerned general practice, social services, community health services, equipment supplies agency and local authority housing department. The patient costs included the purchases of disposables and non-prescription medications, travel, and additional household costs for the participants and their carers. The value of carers' time (and lost productivity associated with caring) was not measured. The costs and the quantities were not reported separately. The resource use data were collected from clinical records at each of the agencies providing care. The unit costs were derived from routine sources, locally where possible, and from national sources representative of local costs. The participants and carer costs were as reported in follow-up interviews, which included a structured cost questionnaire. The capital costs were valued based on purchase price, expected lifetimes and straight-line depreciation. All the costs were adjusted to 2001 values using the medical care component of the Consumer Price Index. No discounting was carried out as the costs were incurred during less than one year. The price year was 2001.
Statistical analysis of costs The costs were expressed as the mean +/- standard deviation, and confidence intervals (CIs) were estimated using bootstrap estimation.
Indirect Costs The indirect costs were not included in the analysis.
Currency UK pounds sterling (). Conversions to US dollars ($) and Euros were also reported: 1 = $1.59 = Euros 1.47.
Sensitivity analysis Threshold analyses were undertaken on the cost-effectiveness using cost-effectiveness acceptability curves. The discounting of life expectancy was investigated in a sensitivity analysis.
Estimated benefits used in the economic analysis No statistical analysis on the health benefits was reported.
The mean QALYs gained at 8 months were -0.38 in the SW group and -0.33 in the OT group. This resulted in a mean difference of 0.05 QALYs between the OT and SW groups.
Cost results The mean total costs of care per participant over the 8-month period were 4,379 (+/- 4,173) for the OT arm and 3,837 (+/- 4,736) for the SW arm. The difference in the mean cost per case was 542 (95% CI: -434 - 1,519). This difference was not statistically significant.
The wide CI reflected the high variability in cost between the trial participants.
The value of equipment and adaptations services was significantly higher in the OT arm by 581 (95% CI: 178 - 984).
The mean primary and community care costs were equivalent for both groups.
The mean secondary health service costs were higher per participant overall in the OT arm by 103 (95% CI: -508 - 714). This difference was not statistically significant.
The differences in costs were significantly different between the two groups for carers, (p=0.02, t-test) but not for the participants.
Synthesis of costs and benefits Since the outcomes for the patients were not different, a cost-minimisation approach should have been considered. However, the authors reported that even a cost-minimisation analysis was inconclusive, since the trial was not powered to demonstrate cost-differences. Therefore, they used cost-effectiveness acceptability curves to explore uncertainty around the cost-effectiveness estimates.
The authors found a less than 50% probability that early OT intervention would be more cost-effective than the alternative in the local circumstances and, at best, at a cost of 14,000 per QALY gained.
The discounting of life expectancy did not alter the results.
Authors' conclusions "From a policy perspective, the lack of difference in clinical and cost-effectiveness means that either a social work or an occupational therapy service is successful in making care assessments that enable an older person to remain in their own home."
CRD COMMENTARY - Selection of comparators The authors did not clearly justify the choice of the comparators they used. You should consider whether the comparators are relevant in your setting, or whether other comparators could have been relevant as well.
Validity of estimate of measure of effectiveness The effectiveness evidence was based on data derived from a prospective single-blind randomised controlled trial, which was appropriate given the study question. However, it was not possible to comment on the internal validity of the trial, as the present study did not provide details on power calculations (to determine sample size), methods of sampling, blinding and loss to follow-up. These were retrieved from the parent study (see Stewart et al. 2005). Considering the methods reported in the parent study, the analysis of effectiveness was handled credibly. The study sample appears to have been representative of the study population since no exclusion criteria were reported.
Validity of estimate of measure of benefit The measure of health benefits used in the economic analysis was the QALYs. This was appropriate for comparing the results of this study with those of different interventions. It is worth noting that future benefits were not discounted in the base-case analysis, but in the sensitivity analysis. However, the discount rate used was not reported. Details of life expectancy (data and sources) were reported.
Validity of estimate of costs The study adopted the perspectives of the health service and patients. While it appears that the majority of relevant costs have been included, the cost of the volunteer carers' time was omitted from the analysis. Further, the authors acknowledged that costs for the participants and carers were an approximate and limited estimate of out-of-pocket costs. This means that the cost-effectiveness of the interventions might have been overestimated. The costs and the quantities were not analysed separately and this hinders the generalisability of the results. No sensitivity analysis of the costs was undertaken, which may limit the interpretation of the study findings. Discounting was appropriately not undertaken as the costs were incurred during less than one year. The year to which the prices referred was reported, and this increases the reproducibility of the results.
Other issues The authors drew comparisons with other studies that found difference results. They suggested that differences in populations and severity or illness, or in the measure of quality of life, might explain the differences in findings. The issue of generalisability to other settings was not addressed. Although the results of the clinical study were not reported in full, the authors' conclusions reflected the scope of the analysis.
The authors reported several limitations to their analysis. First, analyses from the carer perspective were difficult because the number and follow-up of carers in the trial provided less than full data. Second, the assessment of out-of-pocket costs was limited. Third, some costs were omitted from the analysis. Fourth, the estimated costs of services may not be representative of future practice. Finally, as most installations of home adaptations had not been achieved before the trial follow-up, the analysis might have produced a pessimistic view of cost-effectiveness.
Implications of the study The authors suggested "further research on cost-effectiveness of OT assessment should consider how to achieve the earlier installation of equipment, and/or include longer follow-up than was possible in this study".
Source of funding Funded by the Department of Health.
Bibliographic details Flood C, Mugford M, Stewart S, Harvey I, Poland F, Lloyd-Smith W. Occupational therapy compared with social work assessment for older people: an economic evaluation alongside the CAMELOT randomised controlled trial. Age and Ageing 2005; 34(1): 47-52 Other publications of related interest Stewart S, Harvey I, Poland F, et al. Are occupational therapists more effective than social workers when assessing frail older people? Results of CAMELOT, a randomised controlled trial. Age Ageing 2005;34:41-6.
Hay J, LaBree L, Luo R, et al. Cost-effectiveness of preventive occupational therapy for independent-living older adults. J Am Geriatr Soc 2002;50:1381-8.
Indexing Status Subject indexing assigned by NLM MeSH Aged; Cost-Benefit Analysis; Delivery of Health Care; Dependency (Psychology); Frail Elderly; Geriatric Assessment; Great Britain; Health Care Costs; Health Resources /economics /utilization; Humans; Occupational Therapy /economics; Quality of Life; Quality-Adjusted Life Years; Social Work /economics AccessionNumber 22005000199 Date bibliographic record published 28/02/2006 Date abstract record published 28/02/2006 |
|
|
|