|
Modelling the cost effectiveness of lamivudine/zidovudine combination therapy in HIV infection |
Chancellor J V, Hill A M, Sabin C A, Simpson K N, Youle 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 Lamivudine (3TC) and zidovudine (ZDV) combination therapy in treatment of HIV infection, compared to ZDV monotherapy.
Economic study type Cost-effectiveness analysis.
Study population Patients treated at Chelsea and Westminster Hospital in London from 1987 to 1995, ages 17 to 72 years with a median age of 32 to 36 years.
Setting Hospital based HIV-clinic in London. The economic study was carried out in Middlesex, UK.
Dates to which data relate The transition probabilities for the comparator were based on patients treated at the clinic between 1987 and 1995. Costs data were based on the same patient population in 1994 and 1995, and efficacy of the combination therapy was obtained from trials reported between 1995 and 1997. Costs were reported in 1995 prices.
Source of effectiveness data A review of controlled trials published between 1995-1997 was undertaken to establishthe treatment effect of 3TC/ZDV combination therapy. A retrospective survival analysis was conducted to obtain transition probabilities for the alternative treatment (ZDV monotherapy).
Modelling A Markov model with four states was used:
A) HIV positive, non-AIDS, CD4 >200 and <500 cells/mm?3,
B) HIV positive, non-AIDS, CD4 <200 cells/mm?3,
C) HIV positive, AIDS diagnosed,
and D) death.
Outcomes assessed in the review The outcomes assessed were: the risk of disease progression, risk of AIDS, and risk of death.
Study designs and other criteria for inclusion in the review All available randomised clinical trials on 3TC/ZDV combination therapy.
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 6 clinical trials and 1 meta-analysis summarising information from 4 of the reviewed studies.
Methods of combining primary studies Investigation of differences between primary studies The differences between studies were discussed with respect to variation in dosing, combination of drugs, duration of follow-up, baseline case mix of patients, and outcomes.
Results of the review It was concluded that the relative risk (RR) estimates in reviewed trials were sufficiently similar to justify the assumptionsof uniform treatment effect of 3TC/ZDV on all disease state transitions. The RR statistic of 0.509 (95%CI: 0.365-0.710) from the meta-analysis was selected as a conservative estimate to modify baseline transition probabilities in the Markov chain.
Measure of benefits used in the economic analysis The measure of benefits was life-years gained. The Markov model was used to extrapolate baseline (with only ZDV treatment available) transition probabilities from short-term to long-term, i.e. over 20 years. 1-year transition probabilities were estimated using the Kaplan-Meier method. Quality of life assessment was not undertaken.
Direct costs Costs were discounted by 6%. Resource quantities and costs were not reported separately. Costs, including administrative overheads, of treating patients in the HIV centre and respective prices were obtained from the Chelsea and Westminster Hospital in 1994-1995. Estimated costs of community care were obtained from a published study. The Markov model was used to extrapolate annual costs over 20 years. Costs were calculated using 1995 prices.
Sensitivity analysis A one-way sensitivity analysis was conducted by changing treatment effect (RR) estimates, duration of effect of 3TC/ZDV therapy, case mix of starting cohort, annual costs, and discount rate. Analysis covered uncertainty related to aspects of variability in the data and generalisability of the results.
Estimated benefits used in the economic analysis Incremental life-years gained were 0.9 years for the base case, discounted by 0%.
Cost results The expected 20 year costs per patient were 59,551 in the 3TC/ZDV combination therapy group, discounted by 6%. Respective costs in ZDV monotherapy group were 44,612, discounted by 6%. Hence, the incremental cost of 3TC/ZDV treatment compared to ZDV was 5,939, discounted by 6%.
Synthesis of costs and benefits The incremental cost/life-year gained was estimated to be 6,276 (in 1995 prices) in the base case where costs were discounted by 6% and benefits by 0%. The incremental cost/life-year gained was 9,330 when both costs and benefits were discounted by 0%, and 10,201 when both were discounted by 6%. Results were also sensitive to the relative risk estimate used (RR=0.509). Using the 95% confidence intervals 0.365 and 0.710 yielded respective incremental cost-effectiveness ratios of 5,337 and 9,075, costs discounted by 6% and benefits by 0%.
Authors' conclusions Under reasonable assumptions, the predicted cost-effectiveness of 3TC/ZDV combination therapy compared favourably with previously reported economic analyses of various HIV treatments.
CRD COMMENTARY - Selection of comparators A justification was given for the comparator used. The comparator chosen, ZDV monotherapy, was the most widely prescribed antiretroviral regimen before the advent of combination therapy.
Validity of estimate of measure of benefit The estimate of treatment effect was based on an overview of all available randomized controlled trials and was therefore likely to have high validity.
Validity of estimate of costs Adequate details were given of the sources of estimated resource use and prices, but resource use and costs were not reported separately. No important cost items were reported as being omitted.
Other issues The authors' conclusions seem to be justified assuming that the ZDV monotherapy is a relevant comparator. Incremental analysis comparing 3TC/ZDV combination therapy to other available combination therapies would have been useful. The authors, however, discussed the results in comparison to other relevant studies in the field. The generalisability to other settings was addressed by testing sensitivity on an alternative cost scenario, and different case mix of patients. HIV clinic and community care costs may not be generalisable to countries other than the UK. Use of different discount rates for costs and benefits in the base case seemed to favour the more expensive combination therapy. Sensitivity analysis on some key parameters, while using the same discount rate for both costs and benefits would have been useful.
Source of funding Funded by Glaxo Wellcome UK Ltd.
Bibliographic details Chancellor J V, Hill A M, Sabin C A, Simpson K N, Youle M. Modelling the cost effectiveness of lamivudine/zidovudine combination therapy in HIV infection. PharmacoEconomics 1997; 12(1): 54-66 Indexing Status Subject indexing assigned by NLM MeSH Adult; Cost-Benefit Analysis /economics; Drug Combinations; Female; HIV Infections /drug therapy; HIV-1; Humans; Lamivudine /economics /therapeutic use; Male; Models, Economic; Zidovudine /economics /therapeutic use AccessionNumber 21997008267 Date bibliographic record published 30/09/1998 Date abstract record published 30/09/1998 |
|
|
|