Freedberg KA, Kumarasamy N, Losina E, Cecelia AJ, Scott CA, Divi N, et al. Clinical impact and cost-effectiveness of antiretroviral therapy in India: starting criteria and second-line therapy. AIDS 2007 Jul;21 Suppl 4:S117-28.
To project the life expectancy, cost, and cost-effectiveness associated with different strategies for using antiretroviral therapy (ART) in India, to inform treatment programs.
This paper presents a modeling study using the previously published Cost-Effectiveness of Preventing AIDS Complications (CEPAC) International Model, which incorporates data on natural history, treatment efficacy, and costs of care.(1,2) This state-transition model simulates disease progression in a hypothetical cohort of HIV-infected patients and tracks outcomes monthly. Patients move among chronic disease, acute events, and death with the intention of capturing the frequency of clinical events and resource use. In the absence of effective ART, the HIV RNA level determines the rate of decrease in CD4 cell count. Disease progression is characterized by decreases in CD4 cell count, and CD4 cell count determines the incidence of opportunistic infections (OI) and mortality resulting from HIV disease. ART that effectively suppresses HIV RNA generally results in an increase in CD4 cell count at rates reported in clinical trials and a decrease in the incidence of OIs and mortality independent of changes in CD4 cell count. Cotrimoxazole is initiated at CD4 cell counts <200 cells/µl, or occurrence of OIs, and is included with all ART strategies in the absence of major toxicity; its use decreases the incidence of bacterial infections and pneumocystis jiroveci pneumonia, but increases rates of mild fungal infections. One million patients are simulated for each treatment strategy. Costs are reported in US dollars; costs and life-years were discounted at 3% per year.
The simulated cohort consisted of treatment-naive, HIV-infected adults with age, sex, initial CD4 cell count, and HIV RNA levels similar to those of the cohort of 6,918 patients who presented for care at the Y.R. Gaitonde Centre for AIDS Research and Education (YRG CARE) in Chennai, India from 1996 to 2005. The mean age was 32.6 years, 66% were male, and the initial mean CD4 count was 318 cells/µl at the time of presentation.
There was no intervention in this study.
The primary outcomes were life expectancy in months and lifetime costs per person. The study projects the cost-effectiveness (CE) associated with first-line ART and the incremental CE of switching to a second-line regimen after diagnosing first-line treatment failure.
Model inputs: In the model, ART was initiated on the basis of CD4 cell count (200, 250, or 350 cells/µl, measured every six months) or with the occurrence of severe OI. ART was stopped if it failed, defined as a 90% decline from peak CD4 count during treatment or with the appearance of severe OIs that occurred six or more months after ART initiation. In strategies with two regimens available, the patient was switched to the second regimen after a 50% decline from peak CD4 count or upon severe OI six or more months after start of the first regimen. Baseline OI incidence in the absence of ART was derived from the YRG CARE cohort; estimates of OI mortality and of the effect of ART on mortality were taken from the placebo arm of the ANRS in Côte d'Ivoire.(2) Estimates of cotrimoxazole efficacy were also based on that trial and US data (annual cost of $9). Virologic efficacy of ART was based on published clinical trial data. The first-line regimen consisted of nevirapine/stavudine/lamivudine (annual cost of $222, 55% viral suppression at 48 weeks). The second-line regimen consisted of ritonavir-boosted indinavir/didanosine/lamivudine (annual cost of $1,435, 65% viral suppression at 48 weeks). Adherence rates were derived from the same sources of the efficacy data, and the maximal duration of HIV RNA suppression was ten years. Resource utilization was estimated from the YRG CARE cohort.
Base case analysis: Patients not receiving ART or co-trimoxazole starting with mean CD4 counts of 318 cells/µl had a projected discounted life expectancy of 34.5 months and a discounted lifetime cost of $530. Cotrimoxazole prophylaxis alone increased discounted life expectancy by 0.4 months for an additional discounted cost of $50. With one line of ART available, cotrimoxazole prophylaxis in combination with stavudine/lamivudine/nevirapine, starting at a CD4 count <200 cells/µl, increased discounted life expectancy to 62.4 months, at a total discounted lifetime cost of $1,540. Starting ART earlier at a CD4 count of <250 cells/µl added another 1.3 months, weakly dominated starting ART at a CD4 count <200 cells/µl, and had an incremental CE ratio of $430 per year of life saved (YLS) compared with no treatment. Starting ART at a CD4 count of <350 cells/µl increased life expectancy to 64.7 months with an incremental CE ratio of $550 per YLS, compared with starting at a CD4 count <250 cells/µl. With the addition of a second protease-inhibitor based regimen, the life expectancy and costs of all ART strategies increased. Starting ART at a CD4 count of <200 cells/µl provided a discounted life expectancy of 84.8 months, at a discounted lifetime cost of $4,980. This strategy also weakly dominated the strategy of starting ART at a CD4 count of <250 cells/µl. The incremental CE ratio for the two-line ART strategies increased progressively from $1,060 per YLS for starting ART at <250 cells/µl compared to no treatment, to $1,530 per YLS for starting at a CD4 count <350 cells/µl compared to starting at <250 cells/µl. Depending on CD4 starting threshold, adding second-line therapy increased discounted life expectancy 22.4-24.2 months with discounted lifetime cost increases of $3,440-3,790. For all CD4 starting thresholds examined, the CE ratio for adding a second-line compared to a single line ranged from $1,850-$1,880 per YLS.
Sensitivity analysis: Relative efficacy and costs of second-line ART and whether therapy was stopped after failure had the greatest impact on results. In addition, the initial mean CD4 count of the cohort had a substantial effect on life expectancy. Continuation of ART after clinical or virologic failure also affected life expectancy, as well as lifetime costs.
The authors conclude that in India, ART will lead to major survival benefits and is cost-effective by World Health Organization (WHO) criteria. The availability of second-line regimens will further increase survival, but their cost-effectiveness depends on their relative cost compared with the first line regimens.
While there is no standardized quality rating for modeling studies, the authors note several limitations. First, they had to use data derived from outside of India for some estimates, including HIV-related mortality. They note, however, that based on the sensitivity analysis, these data had little impact on study results. The authors also did not consider economies of scale that may occur during a rollout. Further, the analysis was patient-based, not population-based, and does not capture issues of HIV transmission. Additionally, this reviewer notes that this study bases ART efficacy at least partially on clinical trial data, which may not be achieved in real life, developing world settings.
Approximately 785,000 people in India are in need of ART.(3) As of early 2007, nearly 57,000 patients had initiated ART through the government program and the program now aims to provide ART to 300,000 more adults and 40,000 children over the next five years.(4) The WHO suggests that health interventions with an incremental CE ratio less than three times the per person gross domestic product (GDP; $,1,758 for India) would be considered "cost-effective" and those less than the GDP itself would be "very cost-effective."(5) Other modeling studies of HIV in India and other developing countries have also been performed.(6,7)
The authors note that India now has more persons living with HIV than any other country in the world. With increasing treatment needs, the importance of determining the most cost-effective treatment strategy for this population is paramount. Additional CE studies are needed to model alternative treatment strategies, such as tenofovir-based regimens, which have fewer toxicities but higher costs, and different degrees of treatment efficacy, which may vary considerably by setting. The cost-effectiveness of the current first-line regimen and survival benefits found in this study are notable, as is the fact that the incremental CE ratio for the second-line regimen is near the three times GDP threshold for cost-effectiveness cited by the WHO. Further decreases in the cost for ART, especially with second-line regimens, could greatly impact survival and treatment accessibility in the developing world.
- Freedberg KA, Losina E, Weinstein MC, Paltiel AD, Cohen CJ, Seage GR, et al. The cost effectiveness of combination antiretroviral therapy for HIV disease. N Engl J Med 2001 Mar 15;344(11):824-31.
- Goldie SJ, Yazdanpanah Y, Losina E, Weinstein MC, Anglaret X, Walensky RP, et al. Cost-effectiveness of HIV treatment in resource-poor settings--the case of Cote d'Ivoire. N Engl J Med 2006 Sep 14;355(11):1141-53.
- WHO. Summary country profile for HIV/AIDS treatment scaleup: India. Accessed: 30 March 2007.
- Kumarasamy N, Solomon S, Chaguturu SK, Cecelia AJ, Vallabhaneni S, Flanigan TP, et al. The changing natural history of HIV disease: before and after the introduction of generic antiretroviral therapy in southern India. Clin Infect Dis 2005 Nov 15;41(10):1525-8.
- WHO. The World Health Report 2002: Reducing Risks, Promoting Healthy Life.
- Over M, Marseille E, Sudhakar K, Gold J, Gupta I, Indrayan A, et al. Antiretroviral therapy and HIV prevention in India: modeling costs and consequences of policy options. Sex Transm Dis 2006 Oct;33(10 Suppl):S145-52.
- Over M, Marseille E, Sudhakar K, Gold J, Gupta I, Heywood P, Hira S. HIV/AIDS treatment and prevention in India: modeling the costs and consequences. 2007.
- Ravenga A, Over M, Masaki E, et al. The economics of effective AIDS treatment. 2007.