Math Experts Suggest AIDS Will Be Halted; Critics Fault Assumptions in Model
August 2, 2002
A mathematical model based on San Francisco predicts it might be possible to stop the AIDS epidemic using available drug therapies. If the model by University of California-Los Angeles researchers is correct, the pandemic could be slowed, even eradicated, after 50 years of widespread use of HIV drugs.
Mathematician Sally Blower and her UCLA colleagues believe that "the United States has a moral imperative to start pouring money into eradication throughout the world ... These drugs, even though they cannot cure individuals who are treated, could head us towards eradication," Blower said in an interview. The study, "Could Widespread Use of Combination Antiretroviral Therapy Eradicate HIV Epidemics?" appears in the Lancet Infectious Diseases (2002; Vol. 2; No. 8).
The research uses a mathematical method called "uncertainty analysis" to calculate the impact of a range of variables, offering possible outcomes from decades of use of HIV drugs in San Francisco. And even making what Blower considers worst-case assumptions -- such as the evolution of drug-resistant viruses that remain infectious and an increase in unsafe sex among treated individuals -- the model predicts the drugs will stop the San Francisco epidemic before the end of this century. The development of improved drugs would further improve the outlook, she said.
The primary problem with all mathematical models, experts say, is the validity of assumptions built into them. "Despite being a statistician, I would argue against making public policy based on math modeling," said Johns Hopkins School of Medicine mathematician Stephen Gange.
The UCLA team assumes that HIV treatments lower the amount of virus in the bloodstream by at least half and up to 100-fold, and that drug-resistant strains will arise but will be less infectious. Gange works with Dr. Ronald Gray in monitoring HIV treatment outcomes in female patients in the United States and Uganda. Their data, which was crunched through Gange's mathematical models and presented at the 14th International AIDS Conference in Barcelona, found that HIV medicines decreased by an essentially insignificant amount the risk that an individual might transmit HIV sexually. That modest effect "would be obliterated," Gray said, by the worst of behavioral changes now seen in the United States, such as unprotected sex. The main reason treatment won't be enough to slow the epidemic, Gange said, is that most patients manage to sustain a 50 percent or less reduction in viral load, which is below the bottom end of the UCLA model.
Blower counters that her model is based on San Francisco, where about 70 percent of the HIV population is in treatment and most have experienced viral load decreases of more than 50 percent. She said any community that attains the same levels of treatment should experience similar declines in HIV over time. The San Francisco model, she said, "could be a global model." However, Blower concedes that if only a small percentage of the HIV population is treated, "you'll have very little impact on transmission" -- which is why she says 90 percent coverage must be the goal.
Newsday (New York City)
07.31.01; Laurie Garrett
Report Says U.S. Minorities Get Lower-Quality Health Care; Moral Implications of Widespread Pattern Noted
This article was provided by U.S. Centers for Disease Control and Prevention. It is a part of the publication CDC HIV/Hepatitis/STD/TB Prevention News Update. Visit the CDC's website to find out more about their activities, publications and services.