Loading...
Monitoring viral loads (VLs) during antiretroviral therapy (ART) allows patients to avoid unnecessary switches to expensive, second-line ART and identifies true treatment failures earlier. However, questions have been raised about whether limited funds should be used to develop laboratory infrastructure for VL monitoring when they could be used to expand access to treatment. In addition, the relative cost-effectiveness of monitoring CD4-cell counts and VLs simultaneously, rather than only CD4-cell counts, has been debated (JW AIDS Clin Care Dec 12 2011).
In the present study, investigators used Markov modeling to compare three strategies of treatment monitoring: CD4-cell count, VL, and symptom-based. Patients were assumed to be treatment-naive at initiation of first-line ART and were followed for 6 years. Cohort data from sub-Saharan Africa were used to estimate costs (in 2011 U.S. dollars), life expectancy, treatment-failure rates, and other factors.
In the base-case analysis, symptom-based monitoring yielded an average per-person cost of $4028 over the 6-year period and a life expectancy of 64 months. With testing every 6 months, average raw costs were $630 lower with CD4-cell count monitoring and $621 lower with VL monitoring; corresponding gains in life expectancy were 0.9 and 2.3 months, respectively. With testing every 12 months, the life-expectancy advantages of CD4-cell count and VL monitoring remained roughly the same, but the cost savings increased such that both types of monitoring were cost-effective relative to symptom-based monitoring. A wide range of sensitivity analyses confirmed these findings.
Hamers RL et al. Cost-effectiveness of laboratory monitoring for management of HIV treatment in sub-Saharan Africa: A model-based analysis. AIDS 2012 Aug 24; 26:1663.
Comment
The WHO is gradually shifting toward encouraging use of viral load monitoring whenever feasible. VL monitoring has been found to be cost-effective in several studies, but the present analysis is the first to show that it may actually yield raw cost savings. If the study had accounted for a reduction in drug resistance from early identification of treatment failure, the evidence for VL monitoring would have been even stronger. Furthermore, VL monitoring may have public health benefits not captured in the model, including more rapid identification of patients for intensive adherence counseling and reductions in incident opportunistic infections, mortality, and transmission of both HIV infection and drug resistance. Given the potential for cost savings and improved outcomes, efforts to expand the necessary infrastructure for VL monitoring are more essential than ever.