AIDS researchers in Uganda and the United States have found a way to reduce mother-to-child transmission of the AIDS virus far more cheaply and easily than previously possible.

The strategy cuts transmission in half by using about $3 worth of the long-acting antiviral drug nevirapine. Although less effective than the method used in the United States, which requires about $800 in drugs and months of treatment, this one is probably within reach of even the poorest countries.

Preliminary results of an experiment comparing nevirapine to an extremely short course of the antiviral drug AZT in 600 pregnant women were announced yesterday in Kampala, Uganda and at the National Institutes of Health, the federal agency that financed the study.

"I think it's a major breakthrough," said Joseph Saba, an official with UNAIDS, the joint United Nations and World Health Organization AIDS program. "Now we have a number of choices, to be used in different contexts and different countries."

Brooks Jackson, a physician at Johns Hopkins University who helped run the experiment in Uganda, agreed that results were significant.

"In terms of its ability to save lives, this single-dose regimen will potentially save more lives than any other HIV intervention to date in developing countries," Jackson said.

Nevirapine, a member of the "non-nucleoside reverse transcriptase inhibitor" family of drugs, is made by the German company Boehringer Ingelheim. Douglas Wilson, a physician and executive at the American division, said the company is committed in principle to making the drug as widely available as possible.

"We've only received the information in the last day or two," he said.

Between 600,000 and 800,000 infants are born with HIV infection each year, virtually all in the developing world. (There are believed to be fewer than 600 HIV-infected infants born in the United States each year, although the government no longer collects such statistics.)

In the absence of any preventive treatment, HIV-positive women pass the virus to their offspring about one-quarter of the time. With breast feeding, the infection rate climbs to one-third or higher, depending on the mother's health and the age at which the child is weaned.

A study completed in the United States in 1994 showed that transmission could be cut from 24 percent to 8 percent if the woman took the drug AZT for the last two months of pregnancy, and the baby for the first six weeks after birth. Since then, researchers have sought to learn how much that regimen could be pared down before it became useless.

One-half of the 600 Ugandan women got a single nevirapine pill during labor, with their newborns also getting a single dose, in liquid form, after birth. The other half of the women got AZT every three hours during labor, with their babies getting the drug twice a day for one week.

Four months after birth, 13 percent of the babies getting nevirapine were infected with the AIDS virus, compared with 25 percent of the babies getting AZT.

It's uncertain whether AZT was entirely ineffective. The study originally included a group of women who got a placebo. However, that part of the experiment was stopped when a study in Thailand showed that a truncated course of AZT (but a more elaborate one than in the Uganda experiment) showed benefit.

All women after that got either AZT or nevirapine. However, 19 women had already gotten a placebo and delivered six HIV-positive infants. That transmission rate -- about 33 percent -- is higher than the 25 percent in the AZT part of the study, although because the number of cases is small, this may be a statistical fluke.

Nevirapine appears to be so useful because it lasts an astonishingly long time. In the mother, the single dose works for at least a week, inhibiting HIV in the cell-rich (and probably virus-rich) milk called colostrum that's secreted in the first few days of nursing. On the seventh day of life, babies who'd gotten a single dose of nevirapine still had bloodstream concentrations 38 times the levels necessary to inhibit the virus, Jackson said.

He and his colleagues are devising an experiment to see whether periodic doses of nevirapine given to infants during the first six months of life, followed by weaning, can cut transmission even further.

At a retail cost of about $3, and presumably far less if the manufacturer gives a deep discount, the drug cost in a nevirapine-based strategy will be negligible. However, there are other impediments to treatment. Counseling and HIV testing -- both essential parts of a prevention program -- cost between $5 and $29 per patient, depending on the country.

"We shouldn't create overly high expectations," said Saba, the UNAIDS official. "We should be excited, but you will not find it everywhere tomorrow."