|
Cost-effectiveness of earlier initiation of antiretroviral therapy for uninsured HIV-infected adults |
Schackman B R, Goldie S J, Weinstein M C, Losina E, Zhang H, Freedberg K 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 The use of antiretroviral therapy for uninsured adults infected with the human immunodeficiency virus (HIV). Antiretroviral treatment consisted of the administration of indinavir, zidovudine, and lamivudine, followed by second-line therapy with two nucleoside reverse transcriptase inhibitors and one protease inhibitor. The two alternative strategies considered were early antiretroviral therapy (EART) and deferred antiretroviral therapy (DART). EART was administered to patients with CD4 cell counts of 500 cells/microL. DART was administered after the patient presented for the first time CD4 cell counts of below 200 cells/microL.
Economic study type Cost-effectiveness analysis and cost-utility analysis.
Study population The study population comprised a hypothetical cohort of 1 million HIV-infected individuals (80% male) with a mean age of 36 years, who presented for medical care with 500 CD4 cells/microL. These patients were aware of their HIV infection, had never received antiretroviral treatment, lacked private insurance, and met financial eligibility requirements for AIDS Drug Assistance Programmes (ADAPs) and Medicaid.
The ADAPs were assumed to offer coverage to uninsured asymptomatic HIV-infected individuals (although only covering drug costs). Medicaid was assumed to cover health services and medication costs, but only to uninsured, HIV-infected, low-income individuals with symptoms (i.e. CD4 cell count below 200 cells/microL).
Setting The setting was not stated, but it might have been a hospital. The economic study was carried out in the USA.
Dates to which data relate The effectiveness data appear to have been collected from studies published between 1987 and 2001. The cost data appear to have been collected from studies published between 1993 and 2001. The price year was 1998.
Source of effectiveness data The effectiveness data were derived from a non-systematic review of the literature and some authors' assumptions.
Modelling A computer-based, state-transition simulation model of the progression of HIV disease (Freedberg et al., see Other Publications of Related Interest) was used to estimate the health benefits and costs. The health states were categorised as chronic, acute (both according to the CD4 cell count and HIV RNA) or death, although the number of health states was not reported. The patients moved from chronic to acute states when they developed opportunistic infections such as Pneumocystis carinii pneumonia, Mycobacterium avium complex, toxoplasmosis and cytomegalovirus. The time horizon of the model was lifetime. The cycle length appears to have been 1 month.
Outcomes assessed in the review The outcomes assessed in the review were:
the HIV RNA setpoint distribution;
the average monthly rate of CD4 decline without antiretroviral therapy;
the monthly risks of opportunistic infections, acute mortality, and chronic acquired immunodeficiency virus syndrome mortality;
the efficacy of antiretroviral therapy; and
the efficacy of second-line therapy in patients experiencing failure with the initial regimen.
In the paper, the authors reported the percentage of patients with no detectable HIV RNA below 500 copies/microL at 12 and at 24 weeks.
All these estimates appear to have been included as parameters in the model.
Study designs and other criteria for inclusion in the review At least one cohort study, one clinical trial, one randomised clinical trial, and several statistical data studies (all published) were included in the review. The authors did not report any criteria used to include studies in the review.
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 At least 14 published studies were included in the review.
Methods of combining primary studies The authors did not report the method used to combine the results of the individual studies, although a narrative method appears to have been used since the authors identified the references from which the estimates of effectiveness were directly obtained.
Investigation of differences between primary studies The differences between the primary studies do not appear to have been investigated.
Results of the review The percentages of patients with no detectable HIV RNA below 500 copies/microL were 22% at 12 weeks and 60% at 24 weeks.
The results for the remaining estimates of effectiveness were not reported.
Methods used to derive estimates of effectiveness The authors formulated several assumptions to derive estimates of effectiveness.
Estimates of effectiveness and key assumptions The authors assumed that:
all patients were initially enrolled in an ADAP and became eligible for Medicaid on the basis of disability when their CD4 cell count fell below 200 cells/microL;
all patients would remain enrolled in the programme once they became eligible; and
first- and second-line regimens were efficacious for 2 years for patients who had not experienced treatment failure in the interim, and all antiretroviral therapy was discontinued at the time of second-line treatment failure.
Measure of benefits used in the economic analysis The summary measures of benefit used were the number of life-years gained (LYG) and the number of quality-adjusted life-years (QALYs) associated with each of the treatment alternatives studied. The evaluation of health states was obtained from data from four clinical trials (Freedberg et al., see Other Publications of Related Interest). A lifetime period was considered for the estimation of the health benefits, and a discount rate of 3% was applied. In addition, the authors reported the mean CD4 cell count, the incremental number of deaths per 1,000 patients, the incremental opportunistic infections per 1,000 patients, and the mean time of antiretroviral therapy during the first 5 years of treatment, when EART was compared with DART.
Direct costs The direct costs considered in the baseline analysis appear to have been those of the health service. Both drug costs and health care costs associated with routine and acute care were included. The monthly routine costs for different CD4 cell count levels and 3-month costs for acute episodes were reported. The costs used were obtained from published studies, Medicaid and Medicare data, ADAPs data, the 1998 Red Book, the Boston Medical Center cost accounting system, and several authors' assumptions. Therefore, the basis of the cost estimation was current data, authors' assumptions and modelling techniques. The resource quantities were not reported separately from the unit costs. The costs were converted to 1998 prices using the medical component of the Consumer Price Index. Discounting was appropriately performed, at a rate of 3%, given that a lifetime period was considered for the estimation of the costs. The costs reported were the total costs per patient. The estimated costs were adjusted by applying national cost-to-charge ratios.
Statistical analysis of costs No statistical analyses of the costs were reported.
Indirect Costs Although the authors reported that a societal perspective was adopted, the costs related to lost productivity arising from morbidity and mortality were not specifically reported to have been considered in the economic evaluation. In addition, the cost categories included in the patient costs were not described. Therefore, it cannot be stated whether the indirect costs were actually included in the economic analysis.
Sensitivity analysis Sensitivity analyses were performed to assess the robustness of the study results to variations in the parameters included in the model. The variables investigated were treatment efficacy, antiretroviral drug costs and quality of life. These analyses might have been one-way. Analyses of different scenarios, which considered the administration of third- and fourth-line therapies, were also performed. Further analyses (limited to a 5-year period) assessed the cost-effectiveness of the strategies when a third-party perspective was adopted, and the financial budget impact of three US states when different scenarios (according to the level of coverage) were accounted for. The areas of uncertainty investigated were variability in the data and analytical methods.
Estimated benefits used in the economic analysis The discounted average number of LYG per patient was 7.02 years for NART patients, 8.51 years for DART patients, and 9.10 years for EART patients.
The discounted number of QALYs obtained per patient was 6.23 for NART patients, 7.64 for DART patients, and 8.21 for EART patients.
During the first 5 years of treatment EART presented better results than DART in terms of a lower mean CD4 cell count (383 versus 238 cells/microL), 51 fewer deaths per 1,000 patients, and 72 fewer opportunistic infections per 1,000 patients.
The mean duration of antiretroviral therapy was 2.94 years for patients who received EART and 2.20 years for patients who received DART.
Cost results The discounted lifetime costs per patient associated with each of the alternative strategies were $69,900 for NART, $98,000 for DART, and $104,100 for EART.
Synthesis of costs and benefits The estimated health benefits and costs were combined by calculating incremental cost-effectiveness ratios (ICERs) as the additional cost that had to be spent to obtain a further LYG, or an extra QALY, when the different treatment strategies were compared.
The incremental cost per additional LYG with EART compared with NART was $16,500.
The incremental costs per additional QALY were $17,300 for EART versus NART, $10,800 for EART versus DART, and $20,000 for DART versus NART.
DART was stated to be weakly dominated by EART since it presented a higher ICER when compared with NART.
The results from the sensitivity analyses showed that DART remained weakly dominated by EART, except for the situation in which long-term treatment side effects were assumed to reduce quality of life by 20% for EART. In this case, DART was no longer dominated and the ICERs of EART versus DART varied between $67,200 and $22,900, according to different scenarios.
When the financial budget impact of three US states was analysed according to different coverage scenarios, the results showed that differences in federal-state cost-sharing formulae led to substantially different impacts of EART and DART on state and county budgets in both scenarios considered. Overall, when all government payers (3 states) were considered, the undiscounted total cost per patient for these payers was $29,100 with DART and $40,600 with EART. The additional costs of medications for EART were partially offset by savings from averted HIV-related morbidity.
Authors' conclusions Antiretroviral therapy initiated at 500 CD4 cells/microL is cost-effective from a societal perspective when compared with therapy initiated later (when the CD4 cell counts are lower than 200 cells/microL).
CRD COMMENTARY - Selection of comparators The justification for the comparator chosen was that NART would allow comparisons of the alternative treatment strategies in terms of the incremental cost-effectiveness. You should consider whether any of the treatment strategies considered at analysis are widely used treatment strategies for HIV-infected patients in your own setting, and the degree of coverage these patients receive.
Validity of estimate of measure of effectiveness The authors did not state that a systematic review of the literature had been undertaken. The methods used to find and select the primary studies were unclear. Therefore, it is impossible to be certain whether the study was subject to bias due to the exclusion of relevant studies. Although most of the effectiveness estimators were derived from clinical trials, the quality and validity of the data collected from these studies were not reported. Moreover, most of the effectiveness estimators used to model the health benefit results were not reported. Several authors' assumptions were used to derive some of the effectiveness estimators, and were justified with reference to the medical literature. The authors reported that the effectiveness results of this study might be biased because for several reasons. First, the study sample in some of the studies included did not coincide with the study population analysed in the present study. Second, the patient eligibility criteria for coverage of Medicaid or ADAPs vary considerably across states and through time. Finally, the potential benefit derived from the fact that patients receiving antiretroviral treatment may be less likely to infect others was not considered in the analysis.
Validity of estimate of measure of benefit The estimation of health benefits was modelled using instruments that appear to have been appropriate. The estimated health benefits were appropriately discounted, at a rate of 3%, given that the study was performed in the USA. The use of QALYs allows comparisons with other health care interventions.
Validity of estimate of costs Although the authors reported that a societal perspective was adopted in the baseline analysis, they did not state whether they had included the productivity loss derived from morbidity and mortality associated with HIV-infection. Therefore, the perspective would appear to be more closely related to that of the health service. There was a detailed description of the sources of the costs used. However, the categories of costs mentioned were too broad, and it could not be clearly inferred whether all the relevant categories related to the health service perspective were included. Although the price year was identified and appropriate adjustments appear to have been performed, the fact that the resource quantities were not reported separately from the costs hinders reflation exercises in other settings. Further, the applicability of the cost results to other settings may be limited. Sensitivity analyses, in which the effectiveness estimators and costs were varied in order to analyse uncertainty surrounding the study results, were performed.
Other issues The authors compared their results with those from other studies. They reported that similar cost-effectiveness results have been found for Pneumocystitis carinii pneumonia prophylaxis and Mycobacterium avium complex prophylaxis, which are two of the currently used therapies for HIV treatment. Although some of the effectiveness estimators were derived from a cohort study that included mainly white homosexual men, the authors reported that the natural history of HIV infection predicted by the model was consistent with observations among women, non-whites, and injection drug users. Therefore, the results of this analysis may be generalisable to uninsured HIV-infected patients, independent of gender and race. However, the issue of generalisability to other settings was not addressed.
Implications of the study The authors recommended that states should consider the implementation of programmes to expand access to EART in accordance with current treatment guidelines through Medicaid waivers or other Medicaid projects.
Source of funding Funded by the National Library of Medicine, the National Institute of Allergy and Infectious Diseases, and the Centers for Disease Control and Prevention.
Bibliographic details Schackman B R, Goldie S J, Weinstein M C, Losina E, Zhang H, Freedberg K A. Cost-effectiveness of earlier initiation of antiretroviral therapy for uninsured HIV-infected adults. American Journal of Public Health 2001; 91(9): 1456-1463 Other publications of related interest Freedberg KA, Scharfstein JA, Seage GR, et al. The cost-effectiveness of preventing AIDS-related opportunistic infections. JAMA 1998;279:130-6.
Gulick RM, Mellors JW, Havlir D, et al. Simultaneous versus sequential initiation of therapy with indinavir, zidovudine, and lamivudine for HIV-1 infection: 100-week follow-up. JAMA 1998;280:35-41.
Indexing Status Subject indexing assigned by NLM MeSH Anti-HIV Agents /blood /economics /immunology /therapeutic use; Budgets /statistics & CD4 Lymphocyte Count; Cost of Illness; Cost-Benefit Analysis; Drug Administration Schedule; Drug Costs /statistics & Female; Florida; HIV Infections /drug therapy /economics; Health Services Accessibility /economics; Humans; Life Expectancy; Male; Massachusetts; Medicaid /economics; Medically Uninsured; Models, Econometric; New York; Outcome Assessment (Health Care); Patient Selection; Quality-Adjusted Life Years; Sensitivity and Specificity; State Government; Time Factors; numerical data; numerical data AccessionNumber 22001008219 Date bibliographic record published 28/02/2005 Date abstract record published 28/02/2005 |
|
|
|