Study Terms AIDS Testing Accurate

BOSTON--The AIDS test is so accurate that even people at very low risk of the disease can be checked with virtually no risk of being falsely identified as infected, military research shows.

The findings discount one frequently made argument against broad AIDS testing--the fear that healthly people will mistakenly be found to carry HIV, the AIDS virus.

"The public should know that if they choose to, they can do early diagnosis of HIV infection cheaply, they can do it in a logistically feasible way and they can do it with a great deal of accuracy," said Dr. Robert R. Redfield of the Walter Reed Army Institute of Research in Washington.

The latest findings, published in today's New England Journal of Medicine, were based on the first 20 months of military screening of all recruits. Just one of the 135,187 people tested from rural areas was misclassified as infected.

The AIDS test is also widely used to screen donated blood for the virus and to check homosexual men and drug addicts, who are at high risk of AIDS. However, experts disagree over whether such screening should remain largely voluntary or be expanded to include mandatory testing of people at low risk of AIDS, such as all hospital patients.

Opponents argue that, among other things, such broad testing could falsely label many healthy people as infected. For instance, Dr. Klemens B. Meyer of the New England Medical Center has suggested that of every 10 low-risk people with positive AIDS tests, nine of them might actually be free of the virus.

Redfield, who is one of the nation's most persistent advocates of widespread testing, said the latest findings show this is untrue. He believes doctors should routinely check for AIDS just as they do for cancer, heart disease' and other illnesses.

"An early diagnosis of a treatable, preventable disease is something we in medicine should strive for, not run away from," he said.

Redfield suggests that AIDS screening be part of routine testing of hospital patients, pregnant women, marriage license applicants and clients of venereal disease and drug abuse programs. The aim is to spot the infection early, when its symptoms can often be controlled, if not cured, and when carriers can be persuaded not to spread the disease to others.

Dr. Harvey Fineberg, dean of the Harvard School of Public Health, said that while the latest data help build the case for widespread screening, they are "not sufficient to shift the balance altogether."

Issues of confidentiality of test results and possible discrimination against infected people still must be settled, he said. Promoting condom and spermicide use, fighting drug abuse and providing clean needles are more economical ways to stop the spread of AIDS.

Dr. Robin Weiss of the Institute of Medicine agreed that despite the test's accuracy, it should remain voluntary, at least for now.

"There is no overwhelmingly compelling reason at this moment to endorse mandatory widespread population screening," she said, "because the answer to the question, `What is the purpose of the test, and what would you do with the results?' is unclear. There is no treatment. There is no way to ensure that people will change their behavior and not spread the disease."

Others question whether commercial labs will perform the AIDS test as accurately as the military.

Dr. Lawrence Miike of the Congressional Office of Technology Assessment acknowledged the military's "phenomenally high accuracy rate" but added, "I'm not worried about them. I'm worried about labs that advertise, `Come get your HIV test.' I get calls from people who say they are the victims of false positives."

Meyer agreed. "The question is not how well the best laboratories can do," he said. "The issue is how well most laboratories are performing, whether they will maintain this performance and whether it is good enough."

To this, Redfield responded, "I think it's an unusual time when the Department of Defense can do things cheaper, quicker and more accurate than the private sector."