|
Physical activity for the over-65s: could it be a cost-effective exercise for the NHS? |
Munro J, Brazier J, Davey R, Nicholl 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 Regular physical exercise for the elderly as primary prevention.
Economic study type Cost-effectiveness analysis.
Study population A hypothetical cohort of elderly people of both sexes, over the age of 65, with a wide range of abilities.
Setting Community. The study was carried out in Sheffield, UK.
Dates to which data relate The effectiveness data were based on a survey published in 1994. The resource use data and prices corresponded to 1993-1994.
Source of effectiveness data Effectiveness data were based on opinion.
Modelling A probabilistic model (form not described) was used in order to estimate annual costs and benefits associated with the programme for a cohort of 10,000. This was a steady state model, i.e. it assumed that the programme was running. Setting up costs could therefore be ignored and benefits could be assumed to occur in the same year as costs. The model incorporated data related to inpatient costs and programme implementation costs (providing for replacing the 20% of patients who drop out each year from the programme, thereby incurring recruitment costs). It used reductions in incidence rates for death and hospital admissions due to five indications: coronary heath disease, stroke, diabetes, femoral neck fractures, and mental disorders.
Methods used to derive estimates of effectiveness The authors assumed the effectiveness of the physical exercise programme based on data from a previously published review of observational studies. The effectiveness was expressed in terms of deathsand episodes (hospitalisations) avoided, calculated by applying figures for reduction in the annual incidence of events from five different diagnoses (coronary heart disease, stroke, diabetes, femoral neck fractures and mental disorders) to 1995 epidemiological data from the UK.
Estimates of effectiveness and key assumptions It was assumed that there is an average life expectancy of 10 years for over 65s. It was further assumed that 80% of those who begin the programme would still be in it at 12 months. Overall, the annual deaths avoided per 10,000 persons attending the programme were 76.53, and the corresponding figure for annual number of episodes avoided was 229.49. The proportion of incidence of disease avoided for each diagnosis was calculated based on the assumption of 80% of people over 65 years of age not exercising to the recommended levels and therefore eligible to undergo intervention.
Measure of benefits used in the economic analysis Deaths and events (hospitalisations) avoided per year were the measures of benefit.
Direct costs Some quantities of resource use were reported separately from the costs. The costs measured were, salaries and materials associated with the programme including hire of halls and sessional facilitator fees. The costs corresponded to actual figures from a study under way in Sheffield simultaneously with this study. In-patient care costs (used to provide an estimate of costs avoided) were estimated from the specialty cost per day (1993-4) and the average disease specific length of stay for the age group. The final cost estimates were calculated by using a probabilistic model based on epidemiological data from the UK and the cost figures corresponded to the period 1993-94. The costs omitted from the analysis were those associated with primary care, community health services, outpatient visits, and social care.
Indirect Costs Indirect costs were not included due to their likely absence in the population studied.
Sensitivity analysis One-way simple sensitivity analyses were performed by varying the following parameters:cost of intervention, incidence reduction, life expectancy at 65, adherence to the programme (with direct effects upon the recruitment costs), unmeasured savings associated with the omitted costs, and average health care costs.
Estimated benefits used in the economic analysis The annual deaths avoided per 10,000 persons attending the programme were 76.53, while the corresponding figure for annual number of inpatient episodes avoided was 229.49.
Cost results The incremental annual cost of the programme was 253,700 per 10,000 participants. The total cost of the exercise programme was 854,706 and the estimated cost of inpatient episodes avoided was 601,000.
Synthesis of costs and benefits The cost per year of life saved (based on the assumption of a 10-year life expectancy at 65 years) and the cost per avoided health event were the outcome measures used in the synthesis (1993-94 price data). The figures were, respectively, 330 and 1100. The sensitivity analysis yielded a range of variation (after substituting values corresponding to the evidence available from the literature) of 100 to 1,500 for the cost per life year gained by the programme.
Authors' conclusions A publicly funded programme of regular moderate exercise for over-65-year-olds could achieve important health benefits at relatively low cost. The estimates provided by this analysis should now be tested in a rigorous randomized trial, and health commissioners should begin to think of purchasing exercise programmes alongside other health-promoting measures.
CRD COMMENTARY - Selection of comparators The comparator was the 'do-nothing' option (no promotion of exercise and normal care).
Validity of estimate of measure of benefit The authors made clear mention of the high uncertainty in the data derived from the assumptions of reduction in the incidence rates in the diagnoses, based, as they were, on a review of observational studies with an inherent likelihood of selection biases. In fact the study was intended as a preliminary analysis for another study being carried out in Sheffield, which as well as analysing gains in survival, was intended to analyse the quality of life (and hence the possible adverse effects of exercise) in the population studied.
Validity of estimate of costs The scope of the cost analysis was limited by the exclusion of savings resulting from freed resources in the social sector, primary care, community and outpatient services. It also excluded those costs associated with extended life.
Other issues The conclusions reached by the authors were partly justified given the uncertainties in the effectiveness data. The results were comparable to those associated with other well-established preventive interventions reported in the paper. The issue of generalisability was not addressed but the results are applicable to the NHS.
Implications of the study Randomized controlled trials are warranted to analyse the cost-effectiveness of physical exercise as a preventive intervention in patients over 65 years of age.
Bibliographic details Munro J, Brazier J, Davey R, Nicholl J. Physical activity for the over-65s: could it be a cost-effective exercise for the NHS? Journal of Public Health Medicine 1997; 19(4): 397-402 Indexing Status Subject indexing assigned by NLM MeSH Aged; Cost-Benefit Analysis; Exercise; Exercise Therapy /economics; Female; Great Britain; Health Promotion; Humans; Male; State Medicine /economics; Value of Life AccessionNumber 21998008004 Date bibliographic record published 28/02/1999 Date abstract record published 28/02/1999 |
|
|
|