Resource Use in AIDS Treatment in Poor Nations may be Improved by Biostatistics Research
Now, biostatistician Andrea Foulkes at the University of Massachusetts Amherst, with colleagues at Philadelphia's Wistar Institute and elsewhere, propose a tool for prioritizing laboratory-based CD-4 cell count testing by linking cell counts to other patient data. They report details of their new "prediction-based classification" (PBC) system in the current issue of PLoS Medicine. Researchers in the United States, Argentina, South Africa, Canada, the U.K. and Malaysia took part in the study at seven sites around the world.
Foulkes, who with others at UMass Amherst has been actively driving this research, says, "By using these new statistical tools, we can decide how to allocate resources to the patients who need them the most. In other words, we identify which patients are most likely to benefit from secondary testing." PBC could reduce by nearly 57 percent the number of CD-4 tests needed during the first year of ART.
The study is a retrospective analysis that modeled CD-4 counts from 1,000 HIV-infected individuals. The researchers used estimates derived from the model to predict, from CD-4 counts taken at the start of treatment plus white blood cell counts and lymphocyte percentage measurements taken later, whether CD-4 counts would be above the threshold recommended for starting ART and how a patient would do over time.
Luis Montaner at the Wistar Institute says, "Our algorithm could be used as a triage tool to direct available laboratory CD-4 testing capacity to high-priority individuals, that is, those likely to experience a dangerously low CD-4 count." He and colleagues believe that with additional testing and refinement, their PBC system could increase the ability of medical and laboratory facilities in poorer countries to maintain AIDS treatment.
"Our data raises the possibility that we could save money in order to save more lives," Montaner points out. Foulkes, Montaner and colleagues say that more studies are needed to demonstrate the long-term feasibility, clinical effectiveness and cost-effectiveness of the PBC approach and whether the accuracy of its predictions can be improved.