|
||||||||
|
HEALTH POLICY Better Cost–Benefit Ratio Found for AIDS Treatments Than for Some Heart Attack, Breast Cancer TherapiesA pair of studies led by HSPH and HMS researchers give the best evidence yet for the cost-effectiveness of both antiretroviral treatment against HIV and drug-resistance testing to guide drug selection.
Milton Weinstein co-authored two new studies that demonstrate the cost-effectiveness of HIV treatment and drug-resistance testing. Photo by Pam Murray Kenneth Freedberg, director of epidemiology and outcomes research at the Partners AIDS Research Center, led a study which found that patients who got a potent three-drug antiviral cocktail nearly doubled their quality-adjusted life expectancy. Freedberg and coworkers developed a comprehensive computer model of HIV disease, allowing the investigators to assess the effects of antiretroviral treatment on viral RNA levels, CD4 cell counts, opportunistic infections, quality of life, and death. Freedberg and his colleagues used data from an early cohort study to create a baseline for the natural history of HIV infection without treatment. They modeled the effects of an antiretroviral cocktail of zidovudine, lamivudine, and indinavir for one million hypothetical patients based on results of a 1997 trial of patients with advanced AIDS; they used results of other trials to model other regimens and patients. The study appears in the March 15 New England Journal of Medicine. Co-authors include Sue Goldie, assistant professor of decision science, and Milton Weinstein, the Henry J. Kaiser professor of health policy and management and biostatistics, both at HSPH, and others at MGH, HSPH, Boston University, Yale, and the Community Research Initiative of New England. The three-drug combination resulted in an increase in unadjusted life expectancy from 1.97 years to 3.51 years. With adjustment for quality of life, the corresponding figures were 1.53 quality-adjusted life years (QALYs) for untreated patients and 2.91 QALYs for those receiving the three-drug cocktail. Cost of LivingThe drugs increased average total medical costs from $45,000 to $77,000 (including health care costs due to longer life), which works out to a cost of $23,000 per QALY saved—slightly less (or more cost-effective) than thrombolytic therapy in suspected myocardial infarction. When the authors modeled a newer combination regimen with improved efficacy reported in recent clinical trials, and assumed that it would be started when disease is less advanced, they found a cost-effectiveness as low as $11,000 per QALY. Analysis of two-drug therapy yielded results less effective and cost-effective than three drugs."Although costly, antiretroviral therapy is more cost-effective than many therapies for non-HIV diseases," the authors write. These include radiation therapy for early-stage breast cancer ($30,000 per QALY gained), treatment of hypercholesterolemia ($47,000), and dialysis in patients expected to live for less than six months ($150,000). Bug ResistanceIn the March 20 Annals of Internal Medicine, Weinstein, Freedberg, Goldie, and colleagues at HSPH, MGH, Community Research Initiative, and the Robert Wood Johnson Medical School in New Jersey report on their analysis of the clinical impact and cost-effectiveness of genotypic antiviral resistance testing. These tests, which identify HIV variants that have mutated to become drug-resistant, can be used in guiding treatment decisions to avoid giving patients drugs that will be ineffective. They are now typically used in patients for whom one treatment regimen has failed, but concerns about growing drug resistance have led to consideration of testing before initial therapy. With an extension of the mathematical model described above, the researchers used data from two previously completed trials to estimate the effects of resistance testing on the probability of opportunistic infections, quality-adjusted life expectancy, and costs.For testing after treatment failure to guide the choice of subsequent therapy, the investigators found that unadjusted life expectancy rose from 78.3 months without testing to 81.3 months with testing; when adjusted for quality of life, the figures were 60.9 months and 63.1 months. Cost per QALY gained was estimated at $17,900 based on U.S. trial data and $16,300 when data from a French trial were used. Estimates of the cost-effectiveness of resistance testing in patients who have never been on drugs varied, depending on the expected reduction in treatment failure, which depends on the prevalence of resistant HIV in the population. If 20 percent of patients are resistant and testing reduces failure by 25 percent, the cost per QALY would be $22,300. But if only 4 percent of patients are resistant, the cost per QALY would be $69,000. One recent study in Boston found that 16 percent of patients with newly diagnosed HIV infection show resistance. "These analyses show that there is a strong economic, as well as a clinical, justification for the practice of using genotypic resistance testing to guide second-line therapy for HIV infection," Weinstein said. "While many insurance carriers pay for these tests, many Medicaid programs do not. The evidence supporting increases in primary resistance prior to therapy is growing rapidly, and our study suggests a compelling case that resistance testing for these patients would be cost-effective in communities where the prevalence of resistant virus among newly infected patients is high." Paying Up FrontIn a third study, also published in the March 15 New England Journal, researchers with the HIV Cost and Services Utilization Study Consortium examined changes in total medical expenses for HIV-infected patients since the introduction of highly active antiretroviral therapy. Although the drugs added significantly to costs, they found this was more than offset by decreases in hospital costs because patients treated with the new regimens spent less time hospitalized. Average annual costs declined from $20,300 in 1996 to $18,300 in 1998. However, the authors note that among certain groups of HIV patients—women, blacks, the less educated, and those with public health insurance—hospital costs still exceeded drug costs in 1998, suggesting that these patients received suboptimal care. They also note that total lifetime HIV costs are likely to increase as patients live longer.—Tom Reynolds |
