|
Screening for human T cell leukaemia/lymphoma virus among blood donors in Sweden: cost effectiveness analysis |
Tynell E, Andersson S, Lithander E, Arneborn M, Blomberg J, Hansson H B, Krook A, Nomberg M, Ramstedt K, Shanwell A, Bjorkman A |
|
|
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 Screening for human T cell leukaemia/lymphoma virus in blood donors.
Economic study type Cost-effectiveness analysis.
Setting Primary care. The study was carried out in Sweden.
Dates to which data relate Effectiveness estimates were based on a 1 year screening programme carried out in 1994. A retrospective study carried out in 1994 covered data from 1988-1994. The effectiveness estimates were partly based on information from studies published in 1990, 1994 and 1995. The resource use data were based on assumptions relating to the protocol, without any reference to datesbeing made. Also, the price year was not reported.
Source of effectiveness data Effectiveness data were derived from opinion based on a 1 year national screening programme, a pilot study and results from other, unpublished, studies.
Modelling A probabilistic model was used in order to calculate the cost-effectiveness of the three strategies investigated. The model included the benefits of excluding positive cases from future blood donation, in terms of morbidity and mortality outcomes.
Methods used to derive estimates of effectiveness The effectiveness of screening was assumed based on previous evidence from the literature, a 1 year national screening programme in 1994 and the authors' own pilot study. This was reported in terms of sensitivity and specificity of the screening tests.
Estimates of effectiveness and key assumptions Key assumptions used in the model were that the prevalence of human T cell leukemia/lymphoma virus is 1.3/100,000, the annual incidence is 5 in 10 million, the risk of transmission from infected blood is 15%, the risk of developing adult T cell leukemia/lymphoma after transfusion with infected blood is 2.5% after 30 years, mortality is 100%, the risk of developing tropical spastic paraparesis after transfusion with infected blood is 1% after 3 years but with 0% mortality.50% of blood transfusion recipients are assumed to survive for 3 years and 14% for 30 years. It was assumed that no donor infected with the virus was missed by screening due to the 'high sensitivity of the enzyme linked immunosorbent assays', and the specificity was reported to be 99.8% with the initial test, which was then enhanced by the results of the protocol's additional tests.
Measure of benefits used in the economic analysis The measures of benefit were the number of transmissions and deaths avoided. These measures were estimated by means of a probabilistic model incorporating epidemiological and demographic data from Sweden.
Direct costs In spite of the time span covered by the interventions, costs were not reported as discounted. The quantities of resource use were mainly presented implicitly in the protocols. The costs measured were those associated with laboratory use and tests. The cost associated with avoided infections was also included in the calculations, although the costs associated withsociopsychological effects of screening and counselling for those testing positive or indeterminate were omitted. The source of costs and the price year were not clearly reported. A probabilistic model was used in the calculation of costs. It is stated that "actual laboratory costs" were used in the calculation of screening and confirmation costs. The viewpoint was that of the Swedish health service.
Currency US dollars ($), with some individual figures converted to UK pounds Sterling ().
Sensitivity analysis No sensitivity analysis was performed.
Estimated benefits used in the economic analysis The numbers of transmissions prevented each year were as follows: Strategy A, 1.59; Strategy B, 1.49; and Strategy C, 1.35. The figures for the number of deaths avoided annually were, Strategy A, 0.0056 (1 in 180 years); Strategy B, 0.0051 (1 in 195 years);and Strategy C, 0.0047 (1 in 210 years).
Cost results The total annual costs were as follows:Strategy A, $3.02m (1.88m), Strategy B, $0.32m (sterling conversion not given) and Strategy C, $0.17m (sterling conversion not given).
Synthesis of costs and benefits The cost per transmission prevented, the cost per case of disease prevented, and the cost per death prevented were used to combine costs and benefits (neither discounting nor the price year used were reported in calculating the figures that follow). The cost per transmission avoided was $5.59 million for strategy A, $0.70 million for strategy B and $0.44 million (275,000) for strategy C. The corresponding figures for cost per death avoided were $540 million, $63 million, and $36 million. The figures per case of disease avoided were $222 million for Strategy A, $25 million for Strategy B and $15 million for Strategy C. While it was 18 times higher than the cost of testing new donors only, the strategy of testing every donation would yield only one additional positive donor in 7 years relative to the former.
Authors' conclusions Based on these results, the Swedish National Board of Health and Welfare decided that only new blood donors would be screened for infection with the virus.
CRD COMMENTARY - Selection of comparators The reason for the choice of comparators is clear. The comparators investigated were (a) screening every donation, (b) screening new donors followed by retesting at 5 years, and (c) screening new donors only. The no-screening option was also used as comparator. You, as a database user, should consider whether these are commonly used strategies in your own setting.
Validity of estimate of measure of benefit The validity of the measure of benefit cannot be assessed, given the information provided. Epidemiological data was obtained from 'limited published information'.
Validity of estimate of costs The methodology of the cost calculations was not adequately reported, with the price year and the source of costs not being clearly reported. The costs associated with psychological effects of screening in patients were not included in the analysis.
Other issues The conclusions reached by the authors may not be justified given the uncertainties in the data (sensitivity analyses were not carried out). Comparisons with findings from UK, the USA and France were provided, with differences being explained in terms of the differing study methods employed. The generalisability of the study results is limited to nonendemic countries, which are more likely to be western European and Scandinavian countries. The results were not presented selectively.
Bibliographic details Tynell E, Andersson S, Lithander E, Arneborn M, Blomberg J, Hansson H B, Krook A, Nomberg M, Ramstedt K, Shanwell A, Bjorkman A. Screening for human T cell leukaemia/lymphoma virus among blood donors in Sweden: cost effectiveness analysis. BMJ 1998; 316: 1417-1422 Indexing Status Subject indexing assigned by NLM MeSH Adult; Blood Donors /statistics & Blood Transfusion /adverse effects /economics; Blotting, Western /economics; Cost-Benefit Analysis; Enzyme-Linked Immunosorbent Assay /economics; Humans; Incidence; Leukemia-Lymphoma, Adult T-Cell /economics /epidemiology /prevention & Mass Screening /economics /methods; Models, Economic; Prevalence; Program Evaluation; Retrospective Studies; Risk Factors; Sweden /epidemiology; control; numerical data AccessionNumber 21998008113 Date bibliographic record published 28/02/1999 Date abstract record published 28/02/1999 |
|
|
|