A medical watchdog organization yesterday charged that the U.S. government was sponsoring nine medical studies in the developing world that are unethical because they fail to provide all pregnant women with a drug that could protect their infants from acquiring the AIDS virus during childbirth.

The studies seek to learn whether brief use of AZT, or other drugs, will decrease the mother-to-child transmission of human immunodeficiency virus (HIV). A complicated, three-part treatment with the antiviral drug AZT reduces that transmission by two-thirds, but is viewed by governments and AIDS researchers as far too expensive and cumbersome for the developing world.

The key criticism voiced yesterday was that the studies included some HIV-infected women who are being randomly assigned to receive no AZT or other "active" drug anytime during the experiment. In the countries where the studies are being carried out -- mostly nations of West and Central Africa -- AZT is largely unavailable.

"These are as unethical as any experiments we have ever seen since the end of the Second World War," said Sidney Wolfe, director of the Public Citizen Health Research Group, an organization that analyzes research, treatment and public health policy.

"What has happened here is Tuskeegee, part two . . . in which even more people will die," said Peter Lurie, an AIDS researcher who assisted Wolfe in analyzing the studies. Lurie's reference was to the infamous four-decade study in which black Alabama sharecroppers with syphilis were observed and not offered treatment until long after it was widely available.

In a letter distributed to reporters at a news conference here, Wolfe asked Donna E. Shalala, secretary of health and human services, to order researchers to provide all women in the studies with at least some AZT, since that drug has been shown to be more effective than nothing.

Seven of the studies are being funded by the National Institutes of Health (NIH), and two by the Centers for Disease Control and Prevention (CDC).

Various scientists and officials overseeing the experiments, however, said they reached consensus years ago that the experiments criticized by Wolfe were ethical and well-designed. They said it is essential to compare "short-course" AZT treatment to what infected pregnant women in the study countries now are getting -- which is nothing. All of the studies in question have been approved by research review boards in the United States and the host countries.

"All this debate came up in 1994, and it was felt that the best way to go, and the best interest of the developing world, was to have placebo-controlled trials, where you get accurate data quickly," said Joseph Saba, a research official of UNAIDS, the new AIDS program run by the United Nations, World Health Organization, World Bank, and several other international agencies.

"These studies are attempting to define regimens of treatment that are actually usable in most of the world. They have been put together with extraordinary support and consensus on an international level," said Jack Killen, director of the division of AIDS at the NIH's National Institute of Allergy and Infectious Diseases.

A landmark study, published in 1994, found that the mother-to-child rate of transmission of HIV could by cut from 23 percent to 8 percent if the pregnant woman was given AZT tablets for the last one-third of her pregnancy, intravenous AZT during labor, and if the newborn got the drug for the first six weeks of life. In a more recent study, the transmission rate with treatment dropped to 4.8 percent. The cost of that preventive treatment ranges from $400 to $900.

Because many countries with high prevalence of AIDS are unable to afford that regimen or, in many places, deliver the intravenous medicine even if it were available, many AIDS researchers wondered if a simpler, cheaper, but still effective alternative existed.

The countries include Ivory Coast, Uganda, Tanzania, South Africa, Ethiopia, Burkina Faso, Malawi, and Zimbabwe. In all, more than 12,000 women are involved in the U.S.-funded studies. All participation is voluntary, and the women are counseled on the infection and the details of the study.

In most of the studies, the length of time women take AZT before delivery, as well as the length of time their babies take it after birth, has been shortened. Pills have been substituted for the intravenous dose given during labor. In some of those three treatment intervals, a placebo is given instead of AZT. In most of them, there is one "arm" of the study in which women and babies get only placebo.

At the news conference, Wolfe and Lurie said that about 1,000 babies are likely to become infected with the AIDS virus because their mothers are randomly assigned to these all-placebo options. While they said they do not object to studies that give women placebo for some of the intervals, the all-placebo arms are unethical because they, in effect, knowingly give experimental subjects substandard care.

However, Timothy Dondero, of the CDC, said the sort of studies Wolfe and Lurie advocate would not answer the fundamental question: Is any drug treatment better than the treatment women are getting now?

Both Dondero and Saba said that in order to marshal international support and money for HIV testing, counseling and treatment for millions of pregnant women in poor countries, scientists must present unassailable proof of AZT's effectiveness -- which requires trials that include all-placebo arms.

Further, Saba said, doing studies without all-placebo arms would take far longer, delaying implementation of potentially effective treatment and resulting in far more HIV deaths.