|
Screening for abdominal aortic aneurysm: a computer assisted cost-utility analysis |
St Leger A S, Spencely M, McCollum C N, Mossa M |
|
|
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 Ultrasound screening to detect asymptomatic aortic aneurysms (AAA). Following screening, patients at low risk (AAA smaller than 6cm) would be screened annually and patients at high risk (AAA 6cm or greater) would be offered elective surgery.
Type of intervention Prevention and treatment.
Economic study type Cost-effectiveness analysis.
Study population Hypothetical cohort of 2,500 males between the ages of 68 and 72 with possible asymptomatic AAAs.
Setting The setting was secondary care. The economic analysis was conducted in Manchester, UK.
Dates to which data relate Effectiveness data were taken, in part, from studies published between 1988 and 1991. Cost data were taken from observation at a hospital during 1992/93, and from a 1988 publication. No base price year was used.
Source of effectiveness data The evidence/estimates for final outcomes were derived from the literature and from authors' assumptions.
Modelling A decision analysis model was constructed to combine estimates of the prevalence of AAAs, their growth rate and risk of rupture, with data on the costs of a screening programme, offering elective surgery for detected AAAs greater than 6cm, and emergency surgery for ruptured AAAs. The model was run over a five-year period. AAAs under 6cm were considered to be a low risk and were screened annually.
Outcomes assessed in the review The outcomes assessed in the review were the prevalence of AAAs of different sizes, the growth rates of aneurysms, and the risk of rupture of AAAs of differing sizes.
Study designs and other criteria for inclusion in the review Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included Six studies appear to have been included in the review but the study types were not reported.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The results of the review were not reported separately from the estimates based on authors' judgement. Consequently, all results are reported below.
Methods used to derive estimates of effectiveness The authors used their own judgement as to the risk of rupture of aneurysms of different sizes, the prevalence of such aneurysms and the annual growth rate, using published information as a starting point for their estimation. The methods used to calculate the final estimates were not reported in the paper but may be obtained from the authors on request.
Estimates of effectiveness and key assumptions The overall prevalence of AAAs in the study population was estimated to be 0.078.
The prevalence of AAAs between 3.0 and 3.9 cm was 0.053; between 4.0 and 4.9cm was 0.014; between 5.0 and 5.9 cm was 0.005, and at least 6cm was 0.006.
The annual risk of rupture was: 0.000 (3.0 - 3.9cm), 0.006 (4.0 - 4.9cm), 0.035 (5.0 - 5.9cm), and 0.100 (at least 6cm).
Annual growth rates for AAA were 0.26cm (3.0 - 3.9cm); 0.46cm (4.0 -4.9cm) and 0.66cm (5.0 - 5.9cm).
The mortality rate from elective surgery was assumed to be 5%.
It was assumed that all patients requiring emergency treatment for ruptured AAAs died.
It was assumed that 25% of men offered screening would refuse.
Normal life expectancy for a 70-year-old male in England and Wales was assumed to be 12 years.
Life expectancy for men with an untreated AAA greater than 6cm was assumed to be 2 years.
Measure of benefits used in the economic analysis The measures of benefits used in the economic analysis were the additional lives saved as a result of screening and the additional years of life saved.
Direct costs The costs of screening (fixed costs and variable costs) and elective/emergency treatment for AAAs were estimated from the perspective of the English National Health Service using two different sources. Firstly a 1992/93 study undertaken at a Manchester hospital, which included a pilot screening programme for 600 men. Secondly costs were taken from a 1988 study of a screening programme for 2,500 men. Estimates of costs using the Manchester data used 1992/93 prices, whilst cost for the larger screening study used 1986 prices. Costs do not appear to have been discounted, although benefits were discounted at a rate of 5% per annum. Costs were not reported separately from resources used.
Statistical analysis of costs No statistical analysis of costs was performed.
Indirect Costs Indirect costs were not included.
Sensitivity analysis A series of one way sensitivity analyses was undertaken to examine the impact of varying the baseline assumptions. The following parameters were varied: the probability of refusing screening, the elective procedure mortality rate, the prevalence of AAAs of at least 6cm in the population, the risk of annual rupture of AAAs at least 6cm in size, and the annual rupture risk of AAAs between 5.0 and 5.9 cm.
Estimated benefits used in the economic analysis Over the five year period, for a study population of 2,500 patients, an incremental 10.2 lives would be saved as a result of screening. An incremental 101.7 years of life (undiscounted) or 77 years (discounted) would also be saved.
Cost results The total five-year costs of the screening programme using the 1986 costs were 122,706 and the costs associated with no screening programme over the same period were 14,510.
Synthesis of costs and benefits Using the 1986 costs the incremental costs per life saved were 10,636.3 and the incremental cost per life year saved (with benefits discounted at 5% per annum ) was 1,381. Using undiscounted benefits the cost per life year saved was 1,064. Using cost data from the 1992/93 study the figures were 1,698 and 1,285 respectively. Results were robust in sensitivity analysis, although the annual risk of rupture for AAAs was found to be a sensitive parameter. Doubling the risk from 10 to 20% reduced average cost per undiscounted life year saved from 1,206 to 615, whilst reducing the risk to 2%, increased the average costs per life year saved to 65,097.
Authors' conclusions The cost-effectiveness of screening for aneurysms of at least 6cm is favourable compared to that for other widely accepted treatments. Screening for AAAs of at least 5cm is not as favourable, with costs per life year saved ranging between 21,000 and 30,000. AAA screening should be developed and further evaluated within centres of excellence.
CRD COMMENTARY - Selection of comparators The authors compared the intervention screening for AAAs of at least 6m with a do nothing alternative (no screening programme), which represents current practice in England and Wales. You, as a user of the database, should consider whether this represents current practice in your own setting.
Validity of estimate of measure of benefit Effectiveness data were based on a mixture of data obtained from the literature and from the authors' assumptions. Although a systematic review does not appear to have been undertaken, the authors do clearly state that available data were limited and, in many instances, were not appropriate for their model. It is not clear exactly when the authors relied on published estimates and when they used their own judgement. Detailed methods used to estimate benefits were not presented in the paper for the sake of brevity, but are available from the authors upon request. The lack of data from a controlled trial increases the possibility of bias in the analysis. However, uncertainty about the epidemiological data and treatment outcomes was investigated in the sensitivity analysis. In the economic analysis the authors used lives saved and full quality life years saved as outcome measures. Both these measures are appropriate for cost-effectiveness analysis. However, full quality life years saved is incorrectly presented as a quality-adjusted life year (QALY), when it appears to be the unadjusted total life expectancy for a 70 year old male. No valuation method was used to adjust standard life expectancy for quality.
Validity of estimate of costs Costs were presented from the perspective of the English NHS and all relevant costs from this perspective appear to have been included. Cost estimates were taken from two different sources, but no information was provided separately from costs on resources used. The authors acknowledged this limitation and again stated that further information was available on request. Prices used in the analysis were taken from these two different sources and were varied in sensitivity analysis. Although the model used in the study was of five years duration, the authors do not appear to have discounted their estimates of costs, although they do correctly present both undiscounted and discounted benefits in their analysis. Future analyses may also wish to consider whether there are any substantial costs to others in society such as the costs of informal care or any losses in household productivity. It would also have been helpful to have inflated costs from 1986 to 1992/93 to facilitate comparison between the two different sources of cost data.
Other issues The authors correctly considered the incremental cost-effectiveness of the screening programme, and compared this cost-effectiveness with that of other widely accepted health care interventions, and they also made comparison with other similar studies. The issue of the generalisability of their findings to settings outside the English National Health Service was not considered and more information would be required from the authors on the estimates of resource use to undertake such an analysis. The authors acknowledged the limited data available for the economic analysis, and stated that further evaluation of screening programmes for AAA in centres of excellence should be undertaken.
Bibliographic details St Leger A S, Spencely M, McCollum C N, Mossa M. Screening for abdominal aortic aneurysm: a computer assisted cost-utility analysis. European Journal of Vascular and Endovascular Surgery 1996; 11(2): 183-190 Indexing Status Subject indexing assigned by NLM MeSH Aged; Aortic Aneurysm, Abdominal /economics /mortality /prevention & Computer Simulation; Cost-Benefit Analysis; Data Interpretation, Statistical; Great Britain /epidemiology; Hospital Costs /statistics & Humans; Male; Mass Screening /economics /statistics & Models, Economic; Quality-Adjusted Life Years; Risk Factors; Sensitivity and Specificity; control; numerical data; numerical data AccessionNumber 21996000307 Date bibliographic record published 30/06/2001 Date abstract record published 30/06/2001 |
|
|
|