All patients admitted to Aspen Valley Hospital in Colorado are tested for the AIDS virus "as a matter of routine" unless they refuse to sign a consent form given out before the test.

Methodist Hospital in Houston, the city's largest, expects this month to start a similar program of routine, voluntary AIDS virus testing of patients who are scheduled for "invasive" procedures such as surgery or catheterization.

"I think a lot of other hospitals around the country are watching us to see how our program goes," said a spokeswoman for the 1,218-bed Houston hospital.

But St. Thomas Hospital in Nashville, Tenn., which last June became the first hospital in the U.S. to announce a policy of testing patients for the AIDS virus, recently backed off that policy in favor of the Department of Labor's "universal precautions." Those require health care workers to treat all patients as if they carried the AIDS virus.

At 49-bed Aspen Valley Hospital, which began routine HIV testing of patients Nov. 18, the outcome of the test has no bearing on whether a patient gets treatment, a spokeswoman said. It is too early, she said, to tell what percentage of patients are consenting to the HIV test.

To test or not to test for the AIDS virus has become a major question for the nation's 5,732 community hospitals. The vast majority do not routinely test patients for the human immunodeficiency virus (HIV) that causes AIDS. But many are considering such testing, and the resulting debate raises many troublesome issues, including accuracy of the tests, informed consent, confidentiality and cost.

A District of Columbia Hospital Association task force on AIDS last month recommended against universal testing of hospital patients "at this time." The report did urge hospitals to consider HIV testing of maternity patients -- voluntarily -- as part of routine prenatal care.

The report, "AIDS: Strategies for Hospitals," represents a consensus of a 17-member task force and suggests broad guidelines rather than specific policies.

No District hospital is conducting routine HIV testing of either patients or employes, said Joan H. Lewis, vice president of communications for DCHA and principal author of the AIDS report.

"They're all talking about it, or their boards are, but practically speaking, it's not happening at the moment," Lewis said.

"Right now, it's left up to the individual and his or her physician," said D.C. General Hospital medical director Dr. Lawrence Johnson, a task force member.

Advocates of routine testing for AIDS point out that many hospital workers want to know which patients are infected so that they can take special precautions. More extensive testing could also help health officials monitor the prevalence of infection and the rate at which the virus is spreading.

"We all need data," said Dr. William Schaffner, chief of infectious diseases at Vanderbilt University Medical Center in Nashville, which also is considering an HIV testing program. "Somebody needs to do it and do it right and study the results."

Schaffner said any Vanderbilt testing policy would be "rigorously evaluated" so that its results could "inform the national debate" on these issues.

"It's a major, major problem with a lot of emotion tied up in it," said Dr. John J. Lynch, director of the oncology program at Washington Hospital Center and head of that hospital's bioethics committee.

Though he understands that "folks are frightened and think testing is the answer," Lynch said it's "very frustrating" because testing alone is not going to help control the spread of AIDS and could even "drive away a certain percentage of the very people you want to test."

Lynch, who favors voluntary testing, said that the value of mandatory testing of hospital patients is offset by the "false sense of security" offered by a negative test result and the potential for inaccurate or false positives.

The impact of testing on individual patients is "tremendous," said Dr. John Johnson, a pediatric immunologist and assistant director of the AIDS Research Center at the University of Maryland School of Medicine in Baltimore.

"I have personally seen women who have been shot and beaten because they've gone home and told their sex partners that they've had a test for HIV."

Economics is another factor. It costs up to $100 per person to test a group of patients and provide counseling for those who test positive, Johnson said. For a hospital with 30,000 admissions a year, the annual bill for across-the-board HIV testing would reach $3 million.

In certain circumstances such as the case of a health care worker who was accidentally exposed to the blood of a possibly infected patient -- for example, during surgery or by a needle stick -- Johnson recommends voluntary testing of the person exposed.

"We very rarely have people refuse to be voluntarily tested" in such cases, he said.

But mandatory testing of hospital patients or other large groups would be "impossible to administer" , warned Dr. Reed Tuckson, District Public Health Commissioner. And to keep HIV status up to date, "you'd have to test everybody all the time."

Because of the "window" of two to 16 weeks between infection and development of antibodies that the AIDS test is designed to detect, routine testing would result inevitably in some false negatives. These are people who test negative despite the fact that they are infected and carry the virus.

What's more, because an HIV test usually takes about two days to process, it is impractical to test emergency room patients, who make up about 40 percent of the patients at some hospitals.

The American Hospital Association's committee on AIDS last month reiterated opposition to mandatory testing of patients or employes and reaffirmed support of "universal precautions" -- such as use of gloves and gowns -- for hospital workers as the best defense against contracting the virus.

The AHA recommended that hospital patients give consent before being voluntarily tested for AIDS, be told of the test results and receive follow-up counseling if they test positive for the virus. No patient should be denied treatment because of unwillingness to undergo an HIV test, the AHA stated. Mandatory testing is "not something that sound medical practice would condone," said Karen Arcidiacono, associate counsel of Greater Southeast Community Hospital and a member of the hospital association task force on AIDS.

She said she expects hospitals to treat "all patients the same" as far as the possible AIDS risk, and to protect employes by having them use masks, gowns, gloves and goggles, when appropriate, as recommended by the Centers for Disease Control and required by the Labor Department's Occupational Safety and Health Administration.On any given day, there are about 100 AIDS patients in District hospitals. George Washington University Hospital and the Washington Hospital Center account for about half of those, followed by Howard University and Capitol Hill hospitals. In New York City, which has more cases of AIDS than any other U.S. city, about 700 hospital beds are occupied by confirmed AIDS cases on a typical day.

Compounding the AIDS issue in hospitals is the concern over confidentiality.

"A positive HIV test result or the diagnosis of AIDS or ARC will have a major impact on an individual's life, and it is up to the institution and health professionals to protect that patient's confidentiality," the D.C. hospital association report concluded. Hospitals must be responsible for maintaining confidentiality and for disciplining any staff member who violates privacy rules, it said.

A sample one-page informed consent form in the report makes clear that a positive HIV test "probably" means the person has been infected with the virus and is still carrying it, though the test's reliability is "not always certain." The form also states that "a positive test does not necessarily mean that I will develop AIDS" and notes that some negative tests "might actually be positive" because of the delay between infection and development of antibodies that show up on the test.

Before consenting to take the HIV test, the form emphasizes, the patient may ask for counseling about the risk of infection, the test itself, its meaning and limitations, the ramifications of a positive result and who should be told if the test is positive.

The name of the physician who explains the test must be written in, and the form must be dated and signed by the patient and a witness. By signing the consent form, a person agrees that "I have been given all of the information I desire concerning the blood test and release of results, and I have no further questions at this time."