Monitoring of antiretroviral therapy in low-resource settings

Abstract

Recent international debate has focused on the need for CD4 count monitoring, viral load testing, or both, for HIV treatment in resource limited-settings. Andrew Phillips and colleagues address this question using a computer simulation model,1Phillips AN Pillay D Miners AH Bennett DE Gilks CF Lundgren JD Outcomes from monitoring of patients on antiretroviral therapy in resource-limited settings with viral load, CD4 cell count, or clinical observation alone: a computer simulation model.Lancet. 2008; 371: 1443-1451Summary Full Text Full Text PDF PubMed Scopus (146) Google Scholar and conclude that the “benefits of viral load or CD4 cell count monitoring over clinical monitoring alone are modest.” We have concerns with the main results in table 4 and the methods that underlie them. First, Phillips and colleagues do not eliminate “weakly dominated” strategies—ie, strategies that are more expensive and incrementally less cost-effective than more expensive strategies—which is a crucial step in cost-effectiveness analysis.2National Institute for Clinical ExcellenceGuide to the methods of technology appraisal. National Institute for Clinical Excellence, London2004Google Scholar, 3Weinstein MC Decision rules for incremental cost-effectiveness analysis.in: Jones A Elgar companion to health economics. Edward Elgar Publishing, Cheltenham2006Google Scholar They calculate the incremental cost-effectiveness ratio of the viral load monitoring strategy by erroneously comparing it with a weakly dominated strategy—CD4 decline from peak. Had they eliminated the weakly dominated strategies, the final results would be as shown in our table.TablePhillips and colleagues' table 4, with weakly dominated strategies eliminatedLife-yearsCosts (US$)Cost per life-year gained (US$)Viral load >500 copies per mL10·5640553494New WHO 3/4 event10·2228672662Multiple WHO 3 events/new WHO stage 4 event10·012308927New WHO stage 4 event9·752067.. Open table in a new tab Second, the incremental cost-effectiveness ratio for viral load monitoring, when calculated correctly (US$3494 per life-year gained), should be compared with an external standard (eg, 3× the per-capita gross domestic product4Commission on Macroeconomics and HealthMacroeconomics and health: investing in health for economic development. World Health Organization, Geneva2001Google Scholar). Reporting it as “not the most cost-effective” does not allow a policy maker to compare its value with any “willingness to pay” threshold. When calculated correctly, the results also suggest that all CD4 monitoring strategies are weakly dominated by clinical monitoring strategies. These results are at odds with our published work5Goldie SJ Yazdanpanah Y Losina E et al.Cost-effectiveness of HIV treatment in resource-poor settings—the case of Côte d'Ivoire.N Engl J Med. 2006; 355: 1141-1153Crossref PubMed Scopus (221) Google Scholar which finds that CD4 monitoring is cost-effective in Côte d'Ivoire compared with clinical monitoring alone. This finding calls into question whether the value of CD4 monitoring has been underestimated and, therefore, whether the incremental value of viral load monitoring has been overestimated by comparison. We declare that we have no conflict of interest. Monitoring of antiretroviral therapy in low-resource settings – Authors' replyStephen Lawn and colleagues point out that viral load testing has other uses besides being used as a regular monitoring tool for deciding when to switch to second-line treatment. As well as possibly being a suitable basis for promoting adherence in some settings (although the effectiveness compared with other approaches is unproven), its measurement in pregnant women being assessed for mother-to-child transmission, and as a check on patients in whom treatment has failed according to immunological and clinical criteria, are likely to be of some benefit. Full-Text PDF