|
Cost-effectiveness implications of the timing of ART in HIV-infected adults |
Shackman B R, Freedberg K A, Weinstein M C, Sax P E, Losina E, Zhang H, Goldie S J |
|
|
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 comparison of early and deferred initiation of antiretroviral therapy (ART) in patients with human immunodeficiency virus (HIV) RNA levels of 10,000 to 30,000 copies/mL. Early treatment was initiated at a CD4 cell count of 350/microL, whereas deferred therapy was initiated at a CD4 cell count of 200/microL.
Study population The study used a hypothetical cohort of 1 million asymptomatic HIV-infected adults, aged 37 years old, presenting for medical care with HIV RNA levels of 10,000 to 30,000 copies/mL and CD4 cell counts of 350/microL. The hypothetical cohort was 88% male, based on the age and gender distribution of participants in the Dupont 006 trial (see Other Publications of Related Interest). The target population comprised patients with a starting CD4 cell count of 350/microL and HIV RNA levels of 10,000 to 30,000 copies/mL.
Setting A setting was not explicitly stated. The economic study was carried out in the USA.
Dates to which data relate The effectiveness, resource use and utility weight data were all derived from studies published between 1990 and 2002. The cost data were derived from sources published between 1999 and 2001. All of the costs were adjusted to 1999 levels.
Source of effectiveness data The effectiveness data were derived from a non-systematic review and synthesis of published studies, augmented by authors' assumptions.
Modelling The analysis used the Cost-effectiveness of Preventing AIDS Complications model, which is a computer-based state-transition simulation model of the progression of HIV disease (see Other Publications of Related Interest). From the hypothetical cohort of 1 million asymptomatic HIV-infected adults, one person entered the model at a time and each person was observed in the model until death. The health states were categorised as chronic and acute, with patients moving from chronic to acute health states when they developed opportunistic infections.
Outcomes assessed in the review The outcomes assessed were:
the monthly risk of developing opportunistic infections, acute mortality, and chronic acquired immune deficiency syndrome (AIDS) mortality;
the efficacy of antiretroviral regimens (increase in CD4 cell count);
the duration of efficacy for each ART;
acute drug toxicities from ART (adverse effects);
the effects of cholesterol on ART;
the probability of non-AIDS death because of cholesterol increases;
fat redistribution symptoms of ART;
the quality of life decrease associated with acute drug toxicities;
the health-related quality of life for patients who have not experienced an opportunistic infection;
the quality of life for acute opportunistic infection; and
the quality of life associated with fat redistribution symptoms.
Study designs and other criteria for inclusion in the review The authors did not report the inclusion or exclusion criteria used to identify studies for inclusion. However, they did state that they reviewed randomised controlled trials, cohort studies and other published literature.
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 In total, 31 studies provided the effectiveness evidence.
Methods of combining primary studies It was unclear whether the authors combined the results of the primary studies.
Investigation of differences between primary studies No differences between the primary studies were reported.
Results of the review The monthly risk of the development of opportunistic infections was reported for six groups of increasing CD4 cell counts. The six groups were 0 - 50/microL (group 1), 51 - 100/microL (group 2), 101 - 200/microL (group 3), 201 - 300/microL (group 4), 301 - 500/microL (group 5) and >500/microL (group 6).
The monthly probability of pneumocystis carinii pneumonia ranged from 3.700 for group 1 to 0.041 for group 6.
The monthly probability of Mycobacterium avium complex ranged from 1.220 for group 1 to 0.006 for group 6.
The monthly probability of toxoplasmosis ranged from 0.270 for group 1 to 0.003 for group 6.
The monthly probability of cytomegalovirus ranged from 1.857 for group 1 to 0.006 for group 6.
The monthly probability of fungal infections ranged from 1.123 for group 1 to 0.009 for group 6.
The monthly probability of other infections (e.g. bacterial infections, tuberculosis, Kaposi sarcoma) ranged from 1.123 for group 1 to 0.009 for group 6.
For patients receiving protease inhibitors, there was a 26% increase in low-density lipoprotein (LDL) cholesterol and a 15% decrease in high-density lipoprotein (HDL) cholesterol. For patients not receiving protease inhibitors there was a 13% decrease in HDL cholesterol.
Patients who experienced acute drug toxicity from ART incurred a reduction of 0.10 quality-adjusted life-years (QALY).
Other outcomes ascertained from the literature were reported in full.
Methods used to derive estimates of effectiveness Some estimates of effectiveness were supplemented by authors' assumptions
Estimates of effectiveness and key assumptions The authors assumed that when individual drugs in a regimen were substituted, owing to an acute or chronic drug toxicity, the efficacy of that regimen was unchanged and all subsequent regimens remained available to the patient.
The effects of changes in HDL and LDL cholesterol levels on coronary heart disease (CHD)-specific mortality in patients without lipid-lowering treatment was estimated using the risk ratio formulae derived by Schulman et al. (see Other Publications of Related Interest) from data in men with elevated cholesterol levels.
The increases in the probability of non-AIDS death (by age and gender) due to cholesterol increases were assumed to begin 2 years after the mean time of initiation of treatment. The authors conservatively assumed that they continued until death, although there was some evidence in the literature that lipid levels may improve after a change in antiretroviral regimen.
Measure of benefits used in the economic analysis The measure of benefit used was the QALYs. These values were adopted from published literature. The derivation of the quality of life weights assigned to the chronic health states were detailed elsewhere (see Other Publications of Related Interest). As far as the fat redistribution symptoms were concerned, the authors varied the quality of life effect of the symptoms to account for the mean effect of different levels of symptom severity. The future benefits were discounted at an annual rate of 3%.
Direct costs The costs to the hospital were included in the study. These comprised drug costs, test costs, and patient care costs. Resource use and the unit costs were derived from published sources. The costs of patient care were derived from charges reported in the AIDS Cost and Services Utilization Survey, and were converted to estimates of costs using a national cost-to-charge ratio. The costs of CD4 cell counts and HIV RNA tests were obtained from the Boston Medical Centre cost accounting system. The cost of lipid profile tests was obtained from a Medicare reimbursement schedule. Discounting was relevant, as the costs were incurred over the lifetime of a patient, and was conducted at an annual rate of 3%. The resource quantities and the unit costs were reported separately. The costs were expressed in 1999 US dollars but no adjustment for inflation was reported.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs No indirect costs were reported.
Sensitivity analysis A sensitivity analysis was conducted to explore uncertainty in the baseline analysis. The type of sensitivity analysis was not explicitly stated, but it would appear that a one-way sensitivity analysis was conducted. As well as the base-case scenario, the authors reported an alternative scenario that continued ART for patients with detectable HIV RNA after all four lines of ART had failed, while there was no additional benefit of "partial suppression" of HIV RNA. In this alternative scenario, the authors continued to assign the monthly cost of ART until each patient's death and conducted multiple sensitivity analyses to consider the cost-effect of the choice to continue ART for these patients. Additional sensitivity analyses were conducted to test the effect of antiretroviral drug choice, age and gender on the cost-effectiveness ratios, and the quality of life effect for fat distribution symptoms (one-way analysis). The ranges used were derived from published studies.
Estimated benefits used in the economic analysis The benefits were reported for each ART strategy.
Strategy 1: with increased CHD mortality risk, no therapy resulted in 4.30 QALYs, deferred therapy 9.31 QALYs, and early therapy 11.19 QALYs.
Strategy 2: without increased CHD mortality risk, no therapy resulted in 4.30 QALYs, deferred therapy 9.34 QALYs, and early therapy 11.25 QALYs.
Strategy 3: with lipid-lowering agent (pravastatin sodium), no therapy resulted in 4.30 QALYs, deferred therapy 9.31 QALYs, and early therapy 11.21 QALYs.
The duration of the benefits was over the lifetime of the patient. The benefits were discounted at an annual rate of 3%.
Cost results In the base case-scenario, the cost results of the three ART strategies were as follows.
With strategy 1, the lifetime costs per patient were $61,600 for no therapy, $139,700 for deferred therapy, and $152,900 for early therapy.
With strategy 2, the lifetime costs per patient were $61,600 for no therapy, $140,200 for deferred therapy, and $153,500 for early therapy.
With strategy 3, the lifetime costs per patient were $61,600 for no therapy, $141,100 for deferred therapy, and $154,600 for early therapy.
The lifetime costs of each strategy under the continued ART scenario were reported in full in the paper.
Synthesis of costs and benefits An incremental cost-effectiveness analysis was carried out. In the base-case scenario, and for all three ART strategies, deferred therapy was dominated by early therapy. The incremental cost-effectiveness ratio (ICER) for early therapy versus no therapy was $13,000 per QALY gained.
In the continued ART scenario, deferred therapy was no longer dominated by early therapy. The ICER of early therapy versus deferred therapy was $26,000 per QALY.
In the sensitivity analysis to test the effect of antiretroviral drug choice, age and gender on the results, the deferred therapy was again dominated by early therapy. The ICER of early therapy versus no therapy remained almost unchanged at $14,000 per QALY.
The cost-effectiveness ratios appear to be improved in the older cohort (men aged 55 years). This probably reflects the greater life expectancy benefit of successfully treating older, high-risk patients with pravastatin.
When assuming 20% and 40% permanent reductions in health state quality of life weights due to fat redistribution symptoms in the base-case scenario, deferred therapy was again dominated by early therapy. The ICERs of early therapy versus no therapy were $17,000 per QALY (20%) and $24,000 per QALY (40%), respectively.
By contrast, in the scenario in which antiretroviral costs continued until death, early therapy was not dominant versus deferred therapy.
The authors stated that other chronic toxicities associated with ART (e.g. insulin resistance, decreases in bone mineral density), although not considered, are unlikely to affect the cost-effectiveness ratio for early versus deferred therapy, unless the effect of these chronic toxicities on non-AIDS mortality is substantially higher than the CHD effect.
Authors' conclusions The early initiation of antiretroviral therapy (ART), in accordance with the ART guidelines of the US Department of Health and Human Services, remains cost-effective when chronic drug toxicities are considered. The authors also recommended that any decision about initiating ART for individual patients should consider the magnitude of treatment-related adverse effects, particularly fat redistribution symptoms, on quality of life, as it is the most important factor.
CRD COMMENTARY - Selection of comparators The choice of the comparators was explicitly stated. The authors justified their choice by reference to published guidelines for ART, and the considerable controversy that exists as to the appropriate time to initiate ART for HIV-infected patients. You should decide if this represents a valid comparator in you own setting.
Validity of estimate of measure of effectiveness The authors did not explicitly state that a systematic review of the literature was carried out. The sources searched, inclusion criteria and data extraction methodology were not reported. It is therefore possible that the data from the available studies were used selectively. The authors did not note any differences between the studies, or report on the methods used to combine the estimates, if indeed they were combined. Given the level of reporting, it is difficult to judge the quality of the effectiveness parameters used as model inputs.
Validity of estimate of measure of benefit The measure of benefit was the QALYs. These were derived using a model. The authors referred the reader to another study for the methods and data used to derive the quality of life weights assigned to chronic health states. However, the authors acknowledged that because the baseline quality of life weights were derived from patients in the HIV Cost and Services Utilization Study survey, most of whom were not receiving highly active ART, they might have underestimated the offsetting quality of life benefits of ART in their study.
Validity of estimate of costs The authors did not explicitly report the cost perspective adopted. However, it would appear from the costs included that it was likely to have been that of the service provider. It is difficult to assess whether all the relevant costs were included. The authors used published sources to derive the resource use quantities and prices, but sensitivity analyses of the quantities or prices were not conducted. This may limit the interpretation of the study findings. The authors stated that using different data on patient care costs did not have a large effect on the cost-effectiveness ratios for the treatment strategies examined. Where only charges were available, the authors converted them to estimates of costs using a national cost-to-charge ratio. All the costs were expressed in 1999 US dollars but details of how they were adjusted were not reported. All the costs were discounted at an annual rate of 3%.
Other issues The authors did not compare their findings with other published results. However, this might have been due to a lack of published literature in this specific area. The generalisability of the results to other settings or countries was not specifically discussed. The cost estimates and efficacy data appear to have been based on US estimates and evidence, which may limit the generalisability of the study beyond the USA. The results appear to have been fully reported and the authors' conclusions accurately reflected the scope of the study.
The authors highlighted some of the limitations of their study. First, the population used to model ART efficacy came from clinical trials. Such a population is generally considered to be more adherent to drug regimens and clinic visits than a non-trial population. Second, the effects of cholesterol levels on CHD mortality were derived from results in men with elevated cholesterol levels who were not HIV-infected, and published non-fasting lipid-level data on the effect of ART on cholesterol level were used. This might have biased the analysis against early ART. The authors did not compare treatment with a lipid-lowering agent with alternative treatment modalities (e.g. dietary intervention). The authors explained that they omitted this parameter because they assumed that most patients with elevated LDL cholesterol levels had additional CHD risk factors. However, the authors did not consider potential interactions between HIV disease or ART and these other factors.
Implications of the study The authors suggested that their analysis "does not intend to substitute for decision by individual patients and clinicians about when it is best to initiate ART". The authors did not make any recommendation for policy change, nor did they identify further research, although the discussion highlighted some areas where assumptions were made, implying the need for greater information.
Source of funding Supported in part by grants R01 AI42006 and P30 AI42851 from the National Institute of Allergy and Infectious Diseases, Bethesda (MD); by cooperative agreements U64/CCU114927 and U64/CCU119525 from the Centers for Disease Control and Prevention, Atlanta (GA); and by grant LM07092-07 from the National Library of Medicine, Bethesda (MD), USA.
Bibliographic details Shackman B R, Freedberg K A, Weinstein M C, Sax P E, Losina E, Zhang H, Goldie S J. Cost-effectiveness implications of the timing of ART in HIV-infected adults. Archives of Internal Medicine 2002; 162: 2478-2486 Other publications of related interest Staszewski S, Morales-Ramirez J, Tashima KT, et al. Efavirenz plus zidovudine and lamivudine, efavirenz plus indinavir, and indinavir plus zidovudine and lamivudine in the treatment of HIV-1 infection in adults. New England Journal of Medicine 1999;341:1865-73.
Schulman KA, Kinosian B, Jacobson TA, et al. Reducing high blood cholesterol level with drugs: cost-effectiveness of pharmacologic management. JAMA 1990;264:3025-33.
Torrance GW, Feeny DH, Furlong WJ, Barr RD, Zhang Y, Wang Q. Multiattribute utility function for a comprehensive health status classification system: Health Utilities Index Mark 2. Medical Care 1996;34:702-22.
Shackman BR, Goldie SJ, Freedberg KA, Losina E, Brazier J, Weinstein MC. Comparison of health state utilities using community and patient preference weights derived from a survey of patients with HIV/AIDS. Medical Decision Making 2002;22:27-8.
Indexing Status Subject indexing assigned by NLM MeSH AIDS-Related Opportunistic Infections /prevention & Adipose Tissue; Adult; Aged; Anti-HIV Agents /administration & Body Composition; Body Constitution; CD4 Lymphocyte Count; Cholesterol /blood; Cohort Studies; Cost-Benefit Analysis; Drug Administration Schedule; Female; HIV /genetics /isolation & HIV Infections /blood /drug therapy /economics /immunology; Health Care Costs /statistics & Humans; Male; Middle Aged; Quality of Life; RNA, Viral /blood; Randomized Controlled Trials as Topic; Time Factors; United States; Viral Load; control; dosage /economics; numerical data; purification AccessionNumber 22002008284 Date bibliographic record published 28/02/2005 Date abstract record published 28/02/2005 |
|
|
|