Modeling the cost-effectiveness of health programs: HIV testing and early treatment in the USA.

The recent results from the NIH HIV Prevention Trials Network (HPTN) 052 trial, confirmed the relationship between plasma viral load and degree of infectivity of people living with HIV (PLHIV); when PLHIV are treated with antiretroviral therapy (ART), their infectivity is significantly reduced. This reiterates the importance of 'treatment for prevention', as well as the therapeutic function of ART. This and other studies raise a number of important questions, including when to start ART. Given the substantial number of PLHIV that are unaware that they are infected, should policy-makers set-up specific programs to identify these PLHIV and get them into treatment early? Long et al. tried to answer this question by modeling the cost-effectiveness of an expanded screening and treatment program in the USA but how good was their model, how can modeling exercises assist policy-makers in answering these difficult questions and what are some of the broader implications?