Medical Examiner

Does the Third World Deserve Second-Class AIDS Treatment?

Yes; some help is better than no help.

This September, the New England Journal of Medicine created a firestorm when it accused researchers of conducting unethical experiments to reduce HIV in newborns in Third World countries. The charges, comparing the work to the infamous Tuskegee experiment, have since halted at least one HIV study. But a closer look suggests that the critics suffer from an ethical blindness of their own.

The studies in question are seeking low-cost ways to prevent HIV-positive pregnant mothers from transmitting the virus to their newborns. A breakthrough 1994 study found a drug regimen that dramatically reduced HIV-transmission rates–from 26 percent of births to 8 percent. Doctors in industrialized nations rapidly adopted the strategy as the standard of care. At $800 a pop, however, poor countries can’t afford the full regimen. It’s also too complex. Mothers must take AZT, an antiviral, five times a day during pregnancy; receive intravenous AZT during labor; forgo breastfeeding to prevent transmission in milk; and give their babies liquid AZT four times daily for six weeks.

Sixteen studies to find simpler, cheaper methods are now testing everything from vitamin A to a shorter course of oral AZT. The U.S. Centers for Disease Control and Prevention and the National Institutes of Health fund nine of these, involving 12,211 women in Uganda, Thailand, Tanzania, and elsewhere. In every study, some mothers must serve as a comparison group, and that’s where the trouble begins. For two years researchers debated whether to compare new therapies to the full AZT regimen or to no treatment at all. In the end, 15 of the 16 trials gave the comparison group a placebo.

In their article reopening the debate, Peter Lurie and Sidney Wolfe, doctors with Ralph Nader’s Public Citizen Health Research Group, condemned these 15 trials. Researchers giving placebos, they said, were knowingly killing children. They quoted the Helsinki Declaration of the World Health Organization: “In any medical study, every patient–including those of a control group, if any–should be assured of the best proven diagnostic and therapeutic methods.” To them, the only valid question is how new treatments compare to the complex AZT treatment.

T he Journal’s editorial was even more pointed. It accused HIV researchers of “self-serving justification,” and likened their studies to the Tuskegee experiment. Conducted by the U.S. Public Health Service from 1932 to 1972, that study left 412 black syphilitic men untreated–even after penicillin became widely available–in an attempt to determine the natural history of syphilis. The experiment, halted only after exposure by journalists, remains a textbook example of unethical, racist research.

The Journal’s charges made front-page news. Only a pre-publication leak allowed NIH director Harold Varmus and CDC director David Satcher to publish their indignant defense the next week. In an unlikely spectacle, Republican Sen. Dan Coats stood up for poor Africans to oppose Satcher’s nomination for surgeon general. (Ultimately, the Senate Labor Committee approved Satcher 12 to 5.) Johns Hopkins researchers suspended an Ethiopian study with placebos. Other suspensions may follow.

Ethics, the critics insist, demand that U.S. researchers provide the American standard of care to Third World subjects to avoid exploitation. But they don’t. The critics’ arguments seem reasonable only if they ignore the facts:

This isn’t Tuskegee. In Tuskegee, doctors sought no public review, had no clearly beneficial aim, and deceived patients into getting no treatment. By comparison, the HIV trials were reviewed extensively and approved by local governments and the World Health Organization. The experimental therapies may offer great benefit. And some subjects receive placebos because, unlike penicillin for Alabama, the full AZT regimen is out of reach for poor countries.

The complete AZT regimen isn’t coming. By 2000, 6 million pregnant women in developing nations will carry HIV. Critics are right that a complex AZT regimen would be best for them. It is the standard of care. So are CAT scanners and heart-bypass surgery. But poor countries aren’t going to get any of these. Cost is a big reason–these countries typically spend under $20 per person on health care annually. But even if Glaxo Wellcome gave away the drugs, the complete regimen is still infeasible in much of Africa and south Asia. There, tubing and skills for intravenous AZT are scarce. The governments can’t afford to give moms formula to stop them from breast-feeding. Mothers also live too far from doctors to get therapy early in pregnancy and to keep up with necessary monitoring.

Something may be better than nothing. If trials compared low-cost therapy to the complete AZT regimen it’s likely that the new regimens will prove less effective. But such results are useless, since the full treatment is not an option. Without a placebo group, we still won’t know if any of the treatments are better than nothing and therefore worth giving.

We’ve been here before. These issues aren’t new. Cheap oral hydration, for example, has been studied and used to treat dehydration for diarrheal illness in developing countries. It works–but not as well as intravenous hydration. Far from condemning its use as unethical, however, doctors have embraced oral hydration as the model of locally appropriate therapy. Similarly, poor countries won’t adopt a new rotavirus vaccine, despite 600,000 deaths a year from diarrhea caused by the bug. Inoculating every child at $30 a dose would divert scarce resources from even better uses.

The core issue is whether it is ethical for Americans to conduct Third World studies seeking therapies cheaper and perhaps inferior to our own. It can be, but only if: 1) Subjects get the best care feasible locally (they do). 2) It’s done for local benefit (measuring a new therapy against placebo isn’t useful for the West; we’d switch only if it were as effective as the more complex therapy). 3) The new regimens are themselves feasible (at costs from pennies to $60 per patient, they seem to be). 4) The American standard is unavailable for legitimate reasons (I’ve already mentioned all the obstacles).

For the folks at Public Citizen, this last criterion really sticks in their craws. It offends them to see researchers accepting the high cost of American medical care as a given. Could drug companies lower their costs? Maybe, maybe not. But this is not the way to debate free-market pricing of technology. Don’t stop the research. Acknowledging reality isn’t unethical, but ignoring it can be.