Model Warns of Worsening TB Epidemic

ANAHEIM, CALIFORNIA--The death toll from tuberculosis, a scourge making a comeback across the globe, may skyrocket if doctors fail to start screening patients for multidrug-resistant strains, scientists reported here on 23 January at the annual meeting of the American Association for the Advancement of Science (which publishes ScienceNOW). The study's authors argue that the World Health Organization should change its policy to promote such screening.

A team led by University of California, San Francisco, epidemiologist Sally Blower wanted to know how many people in developing countries might die depending on how aggressively doctors treated drug-resistant strains of Mycobacterium tuberculosis. Using a computer model that Blower developed a few years ago to trace the path of the epidemic, her group found that the death rate would triple in Africa--from 55 to 170 deaths per 100,000 people a year--if the current laissez-faire approach to drug-resistant TB were to continue. Plugging into the model new numbers in which physicians seek out and treat patients with drug-resistant strains, the death rate dropped to 10 per 100,000 a year.

The World Health Organization (WHO) now advocates a treatment strategy in which new tuberculosis patients get a course of four drugs and are monitored to assure that they really take them. This approach can sometimes backfire, says Blower, because once a patient carries a bug that is resistant to one or two drugs, blindly continuing the therapy increases the chances of mutations that make a strain able to withstand a third or fourth drug. Blower's model suggests that the WHO standard could accelerate the spread of such multiresistant strains. To prevent this from happening, she says, patients should be screened and treated with alternative, more expensive therapies when necessary.

WHO epidemiologist Christopher Dye says that many developing countries already screen for drug-resistant cases, but usually only after a patient has failed to respond to treatment twice. "There is now discussion, led by WHO, about whether patients should be screened earlier," he says. But even adhering to the current WHO policy is too expensive in many countries--a failure that WHO can do little to address on its $3 million annual budget for TB control, Dye says. He says that models such as Blower's, although useful, "begin with the unrealistic assumption that we live in a world of unlimited resources."