|
Screening for hepatitis C in injecting drug users: a cost utility analysis |
Stein K, Dalzel K, Walker A, Jenkins B, Round A, Royle P |
|
|
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 hepatitis C virus (HCV) using an enzyme-linked immunosorbant assay (ELISA), followed by polymerase chain reaction (PCR) to confirm the presence of HCV ribonucleic acid (RNA), was studied.
Study population The hypothetical target population comprised people with a history of injecting drug use whom were currently in contact with misuse services.
Setting The setting was primary care in the UK.
Dates to which data relate The effectiveness data related to studies published between 1997 and 2000. Resource use was measured by a survey of screening practice among drug services in England, a date for which was not reported. The range of dates to which the unit costs referred was also not provided, although the authors reported that the base year was 2001.
Source of effectiveness data The effectiveness data were derived from a review and synthesis of completed studies, augmented with authors' assumptions and expert opinion.
Modelling An epidemiological model of screening and diagnosis was combined with a Markov chain model of treatment. The model was used to combine the effectiveness and cost elements of screening, diagnosis and treatment. A cohort of 32-year-old patients, with equal proportions of males and females, was used. The model was run over a 50-year period with annual cycles.
Outcomes assessed in the review Since the treatment side of the model has been published already (see Other Publications of Related Interest), this abstract will report details of the screening side of the model, which reflects the emphasis in the current publication.
The following outcomes were assessed:
the underlying prevalence of HCV in people presenting to drug services;
the proportion of people presenting to drug services who are not current IDUs (and therefore not eligible);
the sensitivity and specificity of the ELISA;
the sensitivity and specificity of PCR;
the proportions with mild, moderate, and severe disease;
the proportion of those eligible who accept the ELISA test;
the proportion of those positive to both tests who accept biopsy;
the proportion with moderate disease who accept treatment;
complications of biopsy;
the proportions of complications resulting in hospital admission and death.
Study designs and other criteria for inclusion in the review The estimates were chosen from studies on the basis of methodological quality, how recently the study was published, relevance to the UK, generalisability of the study population, and sample size. Where possible, the authors used existing systematic reviews of good quality.
Sources searched to identify primary studies MEDLINE (from 1996 to 2001), EMBASE (from 1996 to 2001), DARE (August 2001, online), NHS EED (August 2001, online) and HTA (August 2001, online) were searched.
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 Ten studies informed the review of screening and diagnostics.
Methods of combining primary studies The primary studies were not combined to estimate a single parameter value.
Investigation of differences between primary studies Results of the review The underlying prevalence of HCV in people presenting to drug services was 32% (range: 67 - 90).
The proportion of people presenting to drug services who are not current IDUs (and therefore not eligible) was 61%.
The sensitivity of the ELISA was 97.2% (worst-case scenario 97.8%) and the specificity was 100% (worst-case scenario 91.6%).
The sensitivity of PCR was 100% (worst-case scenario 99.8%) and the specificity was 100% (worst-case scenario 99.3%).
The proportion with mild disease was 46% (case-mix 2 = 80%).
The proportion with moderate disease was 43% (case-mix 2 = 15%).
The proportion with severe disease was 11% (case-mix 2 = 5%).
The proportion of those eligible who accept the ELISA test was 49% (scenario = 79%).
The proportion of those positive to both tests who accept biopsy was 77% (range: 10 - 100).
The proportion with moderate disease who accept treatment was 50% (range: 20 - 90).
Complications of biopsy were 30%.
The proportion of complications resulting in hospital admission was 7%.
The proportion of complications resulting in death was 0.03% (range: 0 - 6).
Methods used to derive estimates of effectiveness Authors' assumptions and expert opinion were also used to estimate effectiveness.
Estimates of effectiveness and key assumptions From expert opinion, it was assumed that 100% of those who have had an ELISA will accept the PCR test.
The following arbitrary assumptions were used:
the proportion of people presenting to drug services who are not current IDUs (and therefore not eligible) was 40 to 80%;
the proportion who accept the ELISA was 0.1 to 0.86;
the proportion who accept biopsy was 0.1 to 1.0;
the proportion with mild disease case-mix 1 was 25%;
the proportion with moderate disease case-mix 1 was 60%;
the proportion with severe disease case-mix 1 was 15%; and
the acceptance rate for treatment was 0.2 to 0.9.
Measure of benefits used in the economic analysis The summary measure of benefit used was the quality-adjusted life-years (QALYs). The utilities were taken from a range of published studies, some clinicians' estimates and some arbitrary assumptions. The populations used for the studies were not reported. However, the methods used to estimate utilities were reported. These included both a visual analogue scale and time trade-off.
Direct costs The costing was carried out from the perspective of the NHS. It encompassed a range of costs relevant to this perspective. For instance, the authors considered the cost of estimating eligibility, counselling, diagnostic test costs, and the cost of liver biopsy. The costs were based on actual values and were estimated from sources such as published studies, NHS reference costs, the National Survey of Screening for Hepatitis C, and the Public Health Laboratory Service. Some hospital overheads were included in the analysis. Discounting was carried out at a rate of 6%. The price year was 2001.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not relevant to the study, as it was carried out from the perspective of the NHS.
Sensitivity analysis Extensive one-way sensitivity analyses were carried out to investigate the uncertainty "associated with plausible variation in the values of inputs". The ranges were defined from a review of the literature review, clinician opinions and arbitrary assumptions.
Estimated benefits used in the economic analysis The number presenting was 101,081.
The number screened (accepting ELISA) was 30,213.
The number eligible was 1,555.
The number responding to treatment was 544.
The number of QALYs gained (over no screening) was 303.
The number-need-to-screen to obtain one treatment responder was 186.
Cost results The cost of screening was 3,568,314.
The cost of follow-up was 2,106,619.
The cost of treatment over the no screening alternative was 3,437,539.
The cost of screening sequelae over the no screening alternative was -585,459.
Synthesis of costs and benefits The cost per QALY was 28,120 (range: 11,062 - 278,372).
The authors reported that the base-case estimates changed considerably with variation in the following parameters:
the proportion of HCV-positive patients accepting liver biopsy;
the proportion of people who accept treatment;
treatment response;
the proportion of people eligible for treatment;
the mortality rate associated with biopsy complications;
assigning those who are current IDUs a follow-up outpatient appointment;
the utility assigned to the health state of chronic hepatitis; and
the utility of successful drug treatment.
Authors' conclusions "Screening for HCV (hepatitis C virus) in IDUs (injecting drug users) in contact with services is moderately cost-effective and reasonably stable when explored in extensive one-way sensitivity analyses."
CRD COMMENTARY - Selection of comparators The authors aimed to estimate the cost-utility of screening and treating IDUs. They compared the costs and benefits associated with screening and treating individuals with the costs and benefits of not screening and treating individuals. The comparator was therefore the natural choice in these circumstances.
Validity of estimate of measure of effectiveness The authors did not state that a systematic review of the literature was undertaken, but they appear to have followed systematic reviewing methods. The sources searched were clearly reported and quality criteria were used to determine which primary sources were to be included in the review. The steps taken to assess the validity of data when being extracted, such as whether a single or multiple reviewers contributed, were not reported. The primary studies were not combined to estimate a single parameter estimate and, where there were differences between primary studies, the authors seem to have used these differences to determine the ranges for the sensitivity analysis. The authors did not, however, explicitly discuss potential reasons for differences between the published studies.
Validity of estimate of measure of benefit The utilities were estimated from published studies and clinician estimates. These were incorporated into the decision model to estimate the QALYs. This measure of benefits was appropriate for the studies. In addition, it increases the comparability of the results to broader technologies unrelated to hepatitis C.
Validity of estimate of costs The costing was carried out from the perspective of the NHS. All the categories of cost relevant to this perspective were included, such as overheads and the costs of each diagnostic test. The large difference in cost between the screening and no screening alternatives suggests that omissions in costs, which may have occurred, are unlikely to affect the conclusions drawn. The unit costs were reported separately. The quantities were determined using the decision model.
Other issues The authors did not compare their results with other findings in the discussion, because the other studies had been based outside of the UK setting and were not as comprehensive. The issue of generalisability was not explicitly addressed, although the authors thoroughly discussed many of the variable estimates and their sources, and the implications of this for the results. For instance, they discursively considered the effect of the high rate of missed appointments for IDUs and the impact of this on the results. The results were not presented selectively and several graphs were presented to illustrate further the impact of uncertainty on the results.
The conclusions accurately reflected the results presented and related directly to IDUs who were the target population of the study. Extensive limitations were presented. These centred on uncertainty in the parameter estimates, and ways to overcome these were suggested. The objectives of the study may have been better addressed had the authors been able to relate their conclusions to the current policy on screening and issues that were clarified as a results of the analyses.
Implications of the study The authors did not make any recommendations for policy or practice as a result of their study. They did, however, suggest that further work in examining the impact of uncertainty in more detail and overcoming structural limitations of the model may be useful.
Bibliographic details Stein K, Dalzel K, Walker A, Jenkins B, Round A, Royle P. Screening for hepatitis C in injecting drug users: a cost utility analysis. Journal of Public Health 2004; 26(1): 61-71 Other publications of related interest Shepherd J, Waugh N, Hewitson P. Combination therapy (interferon alpha and ribavirin) in the treatment of chronic hepatitis C: a rapid review and systematic review. Health Technology Assessment 2000;4:1-67.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Antiviral Agents /economics /therapeutic use; Cost-Benefit Analysis; England; Enzyme-Linked Immunosorbent Assay /economics; Female; Hepatitis C /diagnosis /drug therapy /economics; Humans; Interferons /economics /therapeutic use; Male; Markov Chains; Mass Screening /economics; Polymerase Chain Reaction /economics; Quality-Adjusted Life Years; Substance Abuse, Intravenous /complications /economics /virology AccessionNumber 22004008151 Date bibliographic record published 30/11/2004 Date abstract record published 30/11/2004 |
|
|
|