For the cohort of 300 women, using a threshold CD4 count of 200 cells per μL, the projected life-years gained with suppressive therapy, over no intervention, were 142 and the additional costs were $224,000. The incremental cost-effectiveness ratio (ICER) was $1,558 per life-year gained. Using the cheapest international price for acyclovir, the ICER was $1,023 per life-year gained.
At a stringent cost-effectiveness threshold of $1,200 per life-year gained, suppressive therapy was cost-effective in 22% of simulations (67% with the lower price).
Using a threshold CD4 count of 350 cells per μL, the projected life-years with suppressive therapy, over no intervention, were 110 and the costs were $126,000. The ICER was $1,130 per life-year gained or $737 per life-year gained using the lower price. Acyclovir was cost-effective in 58% of simulations (86% at the lower price).
When including the additional cost and benefit of ART, the ICERs of suppressive therapy were higher with each CD4 threshold, and the likelihood of therapy being cost-effective was substantially lower.
The biggest impact on the model outcomes was from variation of the efficacy of suppressive therapy (without ART) and in the assumptions on the default rate (with ART). Variations in other inputs affected the model results, including the proportion of HIV-1-infected women who were HSV2 infected, the initial number of patients receiving suppressive therapy, the salary of the defaulter tracer, the HIV death rate, and the discount rate.