On May 4, a nurse at the Alexandria jail took Carson Bailey from his isolation cell to an empty classroom and told him that he had acquired immune deficiency syndrome.

Three days later, a guard told Bailey to drag his mattress to a dumpster behind the jail. His cell was disinfected, his clothing destroyed and he was driven to the District's Whitman-Walker Clinic, the city's best-known health and counseling center for gays.

But, within a week, doctors there told Bailey he did not have the fatal disease. They say that the medical staff at the Alexandria jail misused a new blood test to make their diagnosis.

"What has happened here is incredibly damaging," said Dr. Peter Hawley, the clinic's medical director. "This guy does not have AIDS. This is not a test for AIDS. It is a test for exposure to the virus that causes AIDS. It should only be used to screen blood. People who look at this as an AIDS test are making a dangerous mistake."

The test, designed to help protect the nation's blood supply, measures the presence of antibodies to the HTLV-III virus, which many medical experts think causes AIDS. As of last Friday it was available at the Whitman-Walker clinic. Maryland clinics began offering the test last week and Virginia is expected to make it available to anybody who wants it within the next 10 days.

As the test, approved by the Food and Drug Administration in March, has become more accessible to those who fear that they may have AIDS, public health officials, gay rights groups and medical researchers have banded together to urge people to avoid it. Many officials say the test may become a powerful tool for discrimination, that it will be used incorrectly to diagnose AIDS and that it will spread panic among members of communities most at risk.

"As a blood test, it is clearly useful," said Dr. Martin Wasserman, director of Arlington's Department of Human Services. "But its implications are pernicious and profound. There is the stigma of a fatal disease, questions of confidentiality and also worries about insurance and discrimination. The test cannot tell you if you have AIDS or whether you will get it. I tell people to stay away."

Advocates of the test say it is a highly accurate means of protecting blood banks from the HTLV-III virus. Blood products and transfusions used by hemophiliacs are linked to about 2 percent of the nation's AIDS cases, according to the Centers for Disease Control in Atlanta. Most of the nation's blood banks now use the test.

In Alexandria, officials defended both their use of the test for inmates with a high risk of having AIDS and their subsequent treatment of Bailey. They said they consulted first with their medical staff and had a judicial order to remove Bailey from the jail.

"As a screening device, it is really very good," said Dr. Martin Cader, director of Virginia's communicable disease control program. "I think most physicians are fully aware of the utility of the test."

Statistics from the CDC suggest that between 5 percent and 20 percent of those whose test results are positive may eventually develop AIDS. But that figure is based on limited data, and officials say a more exact figure may not be known for some time.

Many health professionals at clinics that have received money from CDC to give the test say that while they are concerned about the test's uses, they believe they are better able than blood banks to provide counseling and assure confidentiality. Many members of high-risk groups -- particularly homosexual men, who account for 73 percent of the more than 10,000 AIDS cases diagnosed so far -- would have to turn to blood banks without the alternative clinics.

"Until this test came out, we got about 2,000 calls each month," said Rodger McFarland, executive director of New York's Gay Mens Health Crisis, one of the largest and most respected AIDS assistance organizations in the country. "Now we are getting at least 4,000. Almost every one of those extra calls is somebody asking where they can get the test."

McFarland said that GMHC, along with most other gay organizations, has tried to limit the availability of the test as much as possible. He said GMHC tries to discourage callers from taking the test, telling them that positive results suggest only that a person has been exposed to the virus, not that he or she will ever develop AIDS.

The test's critics also say that because the incubation period for the disease can be several years, negative results do not ensure that a person will not contract the disease. There also are instances where the test registers a falsely positive or falsely negative result.

"This is so tragic," said Dr. Shirley Fannin of the Los Angeles County Department of Health Services. "People want information badly, but the test cannot tell you what you want to know."

Fannin said that last week a Los Angeles mortician refused to handle the body of a person who had died of AIDS, even though he was assured that there has never been a recorded instance of transmission of AIDS through anything other than sexual or intravenous contact.

"This test is going to get more widespread use," said Fannin. "And that means more abuse. But when we start equating a positive test result with a pariah, we are really in big trouble, because there are going to be a lot of pariahs around."

For the administrators of public institutions, such as jails, where the incidence of AIDS has become an increasing problem, the choices can be stark. As the number of cases rises -- the figure is now doubling each year -- corrections officials wonder what to do.

"You tell me what I should have done," Edward J. Prokop, Alexandria's chief deputy sheriff, said in response to questions about why he had Bailey removed from the jail. "I have a staff and 180 other inmates I have to worry about. I am not in a position where I can take chances. I cannot afford panic here."

Public health officials say they are afraid that the Bailey case is evidence that running the test itself can induce panic. Some suggest it illustrates clearly the social implications attached to use of a test that is not a final, accurate indication of AIDS.

"He didn't even need to have the test," said Whitman-Walker's director, Jim Graham. "The word going out there is that this is a test for AIDS. That's dangerous and untrue. The vast majority of those who register positive on the antibody test will never get AIDS."

Bailey, who was arrested on a drug charge, said in an interview that his family heard about his test when they read about it in the newspapers. "I went to kiss my mother," he said. "And she just turned away."

Medical experts say that members of high-risk groups should be cautious in their sexual activity and never donate blood. They also say that most people in the high-risk groups should base their behavior on the assumption that they would test positively.

"The point is that whether you are positive or negative does not matter," said Jeffrey Levi, political director for the National Gay Task Force. "If you are gay you should take precautions."

While doctors say Bailey's physical prognosis is good -- he has some immune dysfunction, but there is no reason to believe he will develop AIDS, they say -- he has been hurt psychologically by the incident.

"I felt like I was the only person in the world," Bailey said about being informed he had the disease. I didn't know which way to turn."

The legal and political implications have had health officials and activists worrying since the test was first given. The Legal Defense and Education Fund of Lambda, a New York homosexual rights organization, almost filed a lawsuit to stop the test.

"We know there are valid uses for this test," said Nancy Langer, the group's public information director. "But we have to add the huge social costs to the benefits. Every day we see the consequences of misinformed action on AIDS. There is an incredible mass hysteria about the disease, and I really believe the test can only add to that."