Cost effectiveness of antiretrovirals — long term or short term?

Abstract

HIV/AIDS is one of the greatest tragedies of the new millennium. No other disease has caused so much human suffering and has thrown back so many development efforts in such a short period. In particular, in sub-Saharan Africa, all attempts to roll back this disease, or at least to improve the situation of those living with it, have to be warmly welcomed. Health economics plays an important role in this fight.[1] As resources are scarce, health policy makers have to invest effort into making the best use of the health workforce, of healthcare budgets and of managerial capacity in order to prevent as many infections as possible, to save as many lives as achievable and to reduce human suffering to the greatest extent with their limited resources. The article by Marseille et al.[2] published in this issue of Applied Health Economics and Health Policy gives clear evidence that home-based highly active antiretroviral therapy (HAART) is effective even in rural Africa. Mortality is reduced and quality of life is gained. However, the article also shows that the cost effectiveness has to be questioned, in particular, for resource-limited countries. At $US597 per disability-adjusted life-year (DALY) averted, it has to be questioned whether HAART should have that high a priority on the agenda of international aid. As Marseille et al.[2] discuss, peer HIV-prevention education for sex workers and adult male circumcision cost approximately $US4-7 and $US12 per DALY averted, respectively.[3,4] Thus, the basic package of healthcare in developing countries should be filled with a number of interventions before HAART is added. However, all these calculations should be handled with the greatest of care as they are based on short-term observations limited to the direct effects on the target group. Thus, they might strongly underestimate the long-term cost of the intervention on the entire system and, consequently, overestimate the cost effectiveness. Figure 1 illustrates a causal loop diagram of an antiretroviral intervention. In the short term, a cost-effectiveness analysis of HAART only has to consider the survival rate, as feedback loops do not exist between antiretroviral intervention and other elements of the healthcare system. Therefore, the total cost of illness can be calculated by simply adding the independent cost components of the HAART regimen and the treatment cost, and the cost effectiveness appears favourable. Fig. 1 Causal loop diagram of the cost of HIV/AIDS. Reproduced from Flessa,[5] with the kind permission of Springer Science + Business Media. ART = antiretroviral therapy; COI = cost of illness; HAART = highly active antiretroviral therapy; pop = population. [Figure omitted.] However, feedback loops with tremendous consequences on the cost of illness and the cost effectiveness of HAART exist in the long run.[6] For instance, antiretroviral drugs might cause resistance,[7] which might increase the number of AIDS cases and, thus, will have an impact on cost as more expensive drugs will have to be prescribed. Most importantly, there is a strong interdependency between AIDS and the economic potential of a nation. On the one hand, patients treated with antiretroviral drugs spend more time in a relatively healthy condition and, thus, can contribute to the GDP. On the other hand, the drug regime requires high resources. Assuming there is a fixed healthcare budget, these costs have to be deducted from the healthcare resources for other diseases. HAART might result in inappropriate budgets for those 'neglected diseases'. …