|
Cost-effectiveness analysis of strategies for hepatitis C screening in French blood recipients |
Jusot J F, Colin C |
|
|
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 health technologies studied represented screening strategies for hepatitis C virus among blood recipients in current use. The screening strategies included the following biological tests: alanine aminotransferase (ALT), a third generation enzyme immunoassay (EIA3), detection of hepatitis C virus RNA (HCV RNA), alanine amino-transferase plus a third generation enzyme immunoassay. These could be applied either before transfusion, or both before and after, giving eight possible strategies. A ninth strategy was 'no testing and no medical therapy'.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised three groups of blood recipients:
Group 1 - young adults less than 40 years old.
Group 2 - blood recipients who had received one to ten units of red blood cells (low-volume transfusion) or were hospitalised in a surgery department or were aged between 40 and 65.
Group 3 - blood recipients who received over ten units of red blood cells (high-volume transfusion).
Setting The setting was a hospital and the economic analysis was carried out in France.
Dates to which data relate Effectiveness data were collected from studies published between 1991 and 1998. Cost and resource use data were collected from a previously published study carried out in 12 French hospitals and the authors' institution. The price year was 1996.
Source of effectiveness data Effectiveness estimates were taken from a review or synthesis of previously completed studies and the authors' assumptions.
Modelling The authors used a 30-year decision analytic model to determine the cost-effectiveness of the screening strategies. One Markov model modelled the evolution of hepatitis C with interferon treatment; a second Markov model modelled the evolution of the stages of hepatitis C for which no interferon treatment was indicated; and a third Markov model was built taking into account false negative cases of the screening strategies and the hepatitis C cases of the 'do nothing' strategy.
Outcomes assessed in the review The review assessed progression rates calculated as annual constant probabilities.
Study designs and other criteria for inclusion in the review Effectiveness estimates were taken from a number of primary studies and one meta-analysis.
Sources searched to identify primary studies The estimates of diagnostic performance of tests were taken from a review of studies extracted from MEDLINE, Excerpta Medica, and Current Contents for the EIA3, and from MEDLINE, Embase, Pascal and Current Contents for HCV RNA.
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 Nine primary studies were included in the review.
Methods of combining primary studies Not applicable for the annual rate of chronicity of hepatitis C, as the rate was stated to be the same for all 3 studies. For the rate of complications of cirrhosis and hepatocellular carcinoma although only one value was presented the method used to reach that value, given several studies, was not presented.
Investigation of differences between primary studies Results of the review The probability of progressing from acute to chronic hepatitis was 0.7.
The probability of detecting a case was 0.7 without interferon treatment and 0.7 with interferon treatment.
The probability of recovery after chronic hepatitis was 0.02.
The probability of detecting a case was 0.02 without interferon treatment and 0.35 with interferon treatment.
The probability of compensated cirrhosis after chronic hepatitis was 0.012.
The probability of detecting a case was 0.012 without interferon treatment and 0.012 with interferon treatment.
The probability of cirrhosis complications after compensated cirrhosis was 0.056.
The probability of detecting a case was 0.056 without interferon treatment and 0.056 with interferon treatment.
The probability of liver transplantation after compensated cirrhosis was 0.
The probability of detecting a case was 0.05 without interferon treatment and 0.05 with interferon treatment.
The probability of hepatocellular carcinoma after compensated cirrhosis was 0.03.
The probability of detecting a case was 0.03 without interferon treatment and 0.03 with interferon treatment.
The sensitivity and specificity were 0.5 and 0.9 for ALT, 0.97 and 0.99 for EIA3, and 0.99 and 0.999 for HCV RNA.
Methods used to derive estimates of effectiveness Authors' assumptions about effectiveness were also used.
Estimates of effectiveness and key assumptions The authors assumed that, with a Knodell's score greater than or equal to 5 only subjects with chronic hepatitis were treated. A Knodell's score lower than 5 would lead to surveillance with three ALT tests within a year. They also assumed that liver biopsy had a specificity of 1 and there would be no false positive cases. Liver transplantation was indicated only for cirrhosis.
Measure of benefits used in the economic analysis The number of life years gained was used as the measure of benefits.
Direct costs Direct costs were not discounted, although the time horizon was 30 years. Resource quantities and unit costs were not reported separately. Direct costs were represented by the costs of biological tests, liver biopsy, treatment (interferon and liver transplantation) and follow-up. The quantity/cost boundary adopted was that of the hospital. The costs of complications from HCV were estimated by two gastroenterologists and three hepatologists. The costs of biological tests and of interferon were collected from a previously published study carried out in 12 French hospitals. The costs of liver biopsy, follow-up of hepatitis stages and transplantation were taken from the authors' institution. The price year was 1996.
Statistical analysis of costs The authors reported costs per patient and per year.
Indirect Costs Indirect costs were not included.
Currency French Francs (Ffr), with Ffr 1 = $0.13 = Euro 0.15.
Sensitivity analysis One-way sensitivity analyses were conducted on ten-year survival rate, prevalence rate of hepatitis C, recovery rate after interferon treatment, sensitivity of ALT, EIA3, HCV RNA detection, cost of testing with ALT, EIA3 and HCV RNA, costs of subsequent disease stages, costs of interferon, and progression rates between different stages.
Estimated benefits used in the economic analysis Life expectancy increased by 0.0085 years and 0.0004 years for subjects under 40 and for those who received low-volume transfusions. For those who received high-volume transfusions, 0.0047 life years were gained during the first year, 0.0033 during the second year, and 0.0030 during the third year, with EIA3 prescribed before and after transfusion and confirmed by HCV RNA. Results for the other strategies were not given.
Cost results Total costs were Ffr 6,954.55 with EIA3 after transfusion for subjects aged under 40. Total costs amounted to Ffr 1,681.95 with EIA3 after transfusion for subjects who received low-volume transfusions. Total costs amounted to Ffr 3,760.90 with EIA3 before and after transfusion twice a year for subjects who received high-volume transfusions during the first year, Ffr 487.62 during the second year, and Ffr 205.70 during the third year. Results for the other strategies were not given.
Synthesis of costs and benefits In the base case the lowest incremental cost-effectiveness ratio was Ffr 776,474 with EIA3 after transfusion for subjects aged under 40.
The lowest incremental cost-effectiveness ratio was Ffr 2,636,500 with EIA3 after transfusion for subjects who received low-volume transfusions.
It was stated that the lowest incremental cost-effectiveness ratio was Ffr 800,191 with EIA3 before and after transfusion twice a year for subjects who received high-volume transfusions during the first year, Ffr 147,764 during the second year, and Ffr 68,567 during the third year. No other results were given
Sensitivity analysis revealed that the lowest incremental cost-effectiveness ratio was Ffr 690,221 per life year gained for subjects under 40 years of age, given a 95% survival at 10 years.
Authors' conclusions The authors argued that the research showed high costs and few life years gained for the dominant strategy even with the most favourable risk group.
CRD COMMENTARY - Selection of comparators The justification for the comparators was that they represented the most frequently used screening strategies in France. You, as a user of the database, should decide if these health technologies are relevant to your own setting.
Validity of estimate of measure of effectiveness The authors did not state that a systematic review of the literature had been undertaken. More details about the search strategies employed and the method of identifying, selecting and combining effectiveness estimates could have been provided. Effectiveness estimates were taken from a number of primary studies and one meta-analysis. Additional effectiveness estimates (quality of life) were based on the opinion of experts, which was appropriate. However, no justification for the assumptions was given. The model was adequately described. The main problem was the lack of presentation of results by strategy, thus hindering transparency.
Validity of estimate of measure of benefit The estimation of benefits was obtained directly from the effectiveness analysis. The use of quality-adjusted life years as the main outcome measure would have made it possible to compare the results with those from studies reporting on similar health technologies.
Validity of estimate of costs Good features of the cost analysis were that all relevant cost categories seem to have been included, the price year was reported, and sensitivity analyses were performed. This permits the replication of the results in other settings and improves the generalisability of the results. However, resource quantities and unit costs were not given, and cost estimates were derived from local sources. Hence, they may not apply to other settings. Also, direct costs were not discounted, although the time horizon was 30 years. However, the main problem was the lack of presentation of results by strategy, thus hindering transparency.
Other issues The authors made appropriate comparisons of their findings with those from other studies, and, through the sensitivity analysis, partially addressed the issue of generalisability to other settings. The authors did, however, appear to present their results selectively in that they reported cost-effectiveness ratios only for the most cost-effective strategies. The study considered three types of blood recipients and this was reflected in the authors' conclusions. The authors acknowledged that the model did not take account of early treatment of the disease and compensation of blood recipients, and other stages of chronic hepatitis and cirrhosis.
Implications of the study The authors argued that this study showed high costs and few life years gained for the dominant strategy even for the more favourable risk group. Moreover, prospective studies are needed that more accurately estimate the cost-effectiveness of screening and management of hepatitis C. It was unfortunate that the study was hampered by a lack of transparency in the presentation of the results (as described above).
Source of funding Financial support from Fonds d'Orientation de la Recherche en Transfusion Sanguine (FORTS) of l'Agence Francaise du Sang
Bibliographic details Jusot J F, Colin C. Cost-effectiveness analysis of strategies for hepatitis C screening in French blood recipients. European Journal of Public Health 2001; 11(4): 373-379 Indexing Status Subject indexing assigned by NLM MeSH Adult; Blood Transfusion /adverse effects /economics; Cost-Benefit Analysis; Enzyme-Linked Immunosorbent Assay; France /epidemiology; Hepatitis C /blood /diagnosis /epidemiology /etiology; Humans; Markov Chains; Mass Screening /economics /methods; Middle Aged; Prevalence; Survival Rate AccessionNumber 22002000129 Date bibliographic record published 30/09/2002 Date abstract record published 30/09/2002 |
|
|
|