|
A prospective cost-consequence analysis of adding lamivudine to zidovudine-containing antiretroviral treatment regiments for HIV infection in the US |
Lacey L, Mauskopf J, Lindrooth R, Pham S, Saag M, Sawyer W |
|
|
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 addition of lamivudine to zidovudine-containing antiretroviral regimens for treating patients with human immunodeficiency virus (HIV) was investigated. No further details of the active drug were given. The comparator was placebo.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients with HIV infection. No further details of the population from which the sample was drawn, nor inclusion or exclusion criteria, were presented.
Setting The setting was secondary care. The CAESAR trial was conducted in Canada, Australia, Europe and South Africa. The costing was carried out for the USA.
Dates to which data relate The time period relating to the trial was not reported. However, the publication reporting further details of the trial was published in 1997. Resource use was estimated for the same time period as the trial. The price year was 1997.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out prospectively on the same sample of patients as that used in the effectiveness study.
Study sample In this report of the trial the authors did not suggest that power calculations had been carried out to rule out the influence of chance in their results. The method of sample selection was similarly not reported. These details are available in the original clinical study (see Other Publications of Related Interest). The initial sample was appropriate for the study question since it included patients with HIV infection. There were 471 patients in the placebo group and 907 patients in the lamivudine group.
Study design The analysis used a randomised, double-blind clinical trial, which was conducted in multiple centres worldwide. The method of randomisation was not reported in the present study. The time horizon was one year. The patients in the placebo group were followed for 331 days while those in the lamivudine group were followed for 342 days. No loss to follow-up was reported.
Analysis of effectiveness The basis of the analysis was intention to treat. The primary health outcomes were disease progression and the number of new or recurrent HIV-related illness events. Disease progression was defined as the occurrence of an acquired immunodeficiency syndrome (AIDS)-defining illness in a HIV-infected patient or a more serious AIDS-defining illness in a patient with AIDS, or death.
The authors reported that the patients in the two groups were similar in terms of the percentage that were male, the median age, the median CD4+ count, the number of the patients at non-AIDS disease stage, and the number of patients with prior antiretroviral experience.
Effectiveness results The rate of disease progression was 20% in the placebo group and 9% in the lamivudine group, (difference = -11, p<0.001).
The mean number of HIV-related illness events was 0.68 in the placebo group and 0.41 in the lamivudine group (difference = -0.27, p<0.001).
The rate of death was 6% in the placebo group and 3% in the lamivudine group (difference = -3, p<0.001).
Clinical conclusions The authors did not draw clinical conclusions independently of the cost conclusions. However, the results indicated that lamivudine was associated with statistically improved clinical benefits.
Measure of benefits used in the economic analysis The authors estimated the HIV illness progression avoided and HIV-related illness avoided. The authors explicitly stated that the life-years gained could not be estimated from the trial data without strong assumptions about the effects of lamivudine beyond the trial period.
Direct costs A perspective for the costing was not explicitly stated. Discounting was unnecessary since the time horizon was shorter than two years. The authors estimated the direct costs related to HIV-related illness and adverse events. These included the number and duration of hospital stays and the number of unscheduled outpatient visits. The unit costs and quantities were estimated from actual data using the patient's case report form, the health care costs and utilisation form, the Physician's desk reference, Medicare fee schedule, Medi-span and the resource-based relative value scale. The costs and the quantities were reported separately. The cost data were adjusted to 1997 US dollars.
Statistical analysis of costs The authors reported that they used "standard parametric techniques" to estimate 95% confidence intervals (CIs) and p-values for the differences in the mean values of the costs. They also used a bootstrap technique.
Indirect Costs The indirect costs were not estimated.
Sensitivity analysis A one-way sensitivity analysis on the cost-to-charge deflators was used to calculate the hospital costs per day and the average length of each hospital stay. These analyses addressed areas of uncertainty where the trial could not provide specific parameter estimates.
Estimated benefits used in the economic analysis The consequences of treatment for 100 treated patients were as follows.
Disease progression occurred in 20 of the placebo patients and 9 of the lamivudine patients (difference = -11 patients).
The mean number of HIV-related illness events was 68 in the placebo group and 41 in the lamivudine group (difference = -27 HIV-related illness events).
Cost results The total cost of treatment (without the lamivudine costs) was $6,186 in the placebo group and $3,541 in the lamivudine group.
The incremental cost of lamivudine (without the lamivudine costs) was -$2,645 (95% CI: -4,225 - -1,065; p=0.001).
The total cost of treatment (with the lamivudine costs) was $6,307 in the placebo group and $5,954 in the lamivudine group.
The incremental cost of lamivudine (with the lamivudine costs) was -$353 (95% CI: -1,975 - 1,269; p>0.2).
Synthesis of costs and benefits The incremental cost per one-person reduction in disease progression was -$3,333 (95% CI: 17,624 - -19,375).
The incremental cost per HIV-related event avoided was -$1,319 (95% CI: 7,075 - -8,970).
When the length of stay was altered, the incremental cost per one-person reduction in disease progression was -$3,503 (95% CI: 21,964 - -9,146).
When the length of stay was altered, the incremental cost per HIV-related event avoided was -$1,387 (95% CI: 9,649 - -$3,857).
In terms of the sensitivity analyses, the authors commented that the results were robust to changes in the cost-to charge deflator, but not to the length of hospital stay.
Authors' conclusions Lamivudine treatment was associated with a saving in the costs of treating human immunodeficiency virus (HIV)-related illness and adverse events. The cost-saving completely offset the incremental cost of lamivudine during the first year of treatment.
CRD COMMENTARY - Selection of comparators Lamivudine and standard treatment (zidovudine) were compared to placebo and standard treatment (zidovudine). The authors used zidovudine and placebo as a comparator because this enabled them to show the active value of lamivudine. However, they pointed out that guidelines in the USA recommend the use of triple combination therapies that include a protease inhibitor, which were not assessed in the CAESAR trial. You should consider the relevance of the technologies examined to your own setting.
Validity of estimate of measure of effectiveness The analysis used a randomised, controlled double-blind study, which was appropriate for the study objective. The study sample was representative of the population as it included patients with HIV infection. The patients were shown to be comparable at analysis.
Validity of estimate of measure of benefit The estimation of benefits was obtained directly from the effectiveness analysis. The authors justified their choice of benefit estimate and also discussed their reasons for not using a more generic measure of benefit such as life-years gained.
Validity of estimate of costs A perspective for the cost analysis was not explicitly stated. Therefore, it is not possible to comment on whether all the costs relevant to the analysis were included. Since the overall cost difference between the treatments was small, omissions in cost may substantially affect the results and conclusions from the study. The costs and the quantities were reported separately. This allows the reader to form a better understanding of the decision problem and to replicate the results in their own setting. Resource use was taken from a single study, and bootstrap and sensitivity analyses were carried out to address uncertainty in the estimates. The unit costs were taken from a variety of sources relevant to the authors' setting and were derived from actual data. The sensitivity analysis explored the uncertainty in the cost-to-charge ratio. The authors pointed out that their cost data might have underestimated the savings with lamivudine, as some health services usage data were not captured in the analysis (lower for the intervention group).
Other issues The authors made appropriate comparisons of their findings with those of other studies, pointing out a number of similarities. The issue of generalisability was not discussed in relation to the results of the costing analysis. However, the effectiveness results were taken from the CASAER trial, which was not conducted in the USA. The authors discussed the issue of using these results within a US setting, and also mentioned this as a limitation of the study. The authors did not present their results selectively and their conclusions accurately reflected the scope of the analysis. However, in terms of the costs, the authors concluded that lamivudine was a cheaper treatment alternative. A 95% CI surrounding the estimated cost-saving included zero, suggesting that the cost difference was not statistically significantly different from zero at a 5% level of significance. The CIs for the incremental cost-effectiveness ratios also included zero, further suggesting there was no statistically significant difference in cost-effectiveness. The authors discussed some limitations of their study. For example, the fact that they did not measure all heath care resource use.
Implications of the study The authors expressed some concern that treatment with lamivudine and zidovudine had now been replaced with triple combination therapies, although lamivudine and zidovudine remained the basis of these therapies. They did not make any recommendations for changes in policy or practice concerning the use of lamivudine and zidovudine. No suggestions for further work were made.
Bibliographic details Lacey L, Mauskopf J, Lindrooth R, Pham S, Saag M, Sawyer W. A prospective cost-consequence analysis of adding lamivudine to zidovudine-containing antiretroviral treatment regiments for HIV infection in the US. PharmacoEconomics 1999; 15(Supplement 1): 23-37 Other publications of related interest Randomised trial of addition of lamivudine or lamivudine plus loviride to zidovudine-containing regimens for patients with HIV-1 infection: the CAESAR Trial. Lancet 1997;349:1413-21.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Ambulatory Care /economics; Anti-HIV Agents /economics /therapeutic use; Cost-Benefit Analysis; Double-Blind Method; Drug Costs; Drug Therapy, Combination; Female; HIV Infections /drug therapy /economics; Health Resources /economics /utilization; Hospitalization /economics; Humans; Lamivudine /economics /therapeutic use; Length of Stay; Male; Middle Aged; Prospective Studies; United States; Zidovudine /economics /therapeutic use AccessionNumber 21999008277 Date bibliographic record published 31/12/2003 Date abstract record published 31/12/2003 |
|
|
|