The idea is to get the public to use a sophisticated electronic questionnaire that would get an instant medical risk assessment and to help persuade those who seem not to be at risk to stay away and give medical professionals time to focus on patients who are.

After the computer leads someone through the questions, either online or over the telephone, the patient would be advised according to what the symptoms seem to be. The options the system might offer include: stay home in voluntary quarantine, go to a designated site for follow-up care, move to an isolation facility or just relax.

The system could be updated with localized information about the hazard and the nearest appropriate medical help, said one of its developers, Georgetown University biodefense coordinator Michael D. McDonald. Programmed correctly, it could be used to respond to severe acute respiratory syndrome, smallpox, nerve agents, radiological hazards and unforeseen diseases or emergencies.

The project is being coordinated by McDonald and Harvard psychiatrist Stephen E. Locke. They have secured no formal funding commitments, but Georgetown officials are conducting meetings that include discussions of the concept with public health officials.

Proponents say the electronic assessments would ease pressure on hospitals while protecting the so-called "worried well" from gathering where they would be in close quarters with people who actually had been exposed to pathogens or chemical agents.

Some experts, though, question whether the screening tool might give bad advice, and they said they doubt many people would seek or accept electronic advice when they are fearful. Others said they worry the system might inadvertently discourage some sick people from seeking help that they need.

Most local public health officials and bioterrorism experts are not familiar with the program, but several said anything that might limit a post-event surge of people to hospitals would be welcome. "If it doesn't cost much and keeps anybody away from the ER, it's worthwhile," said Arlington County Health Director Susan Allan.

"Many patients we saw in our emergency room during the anthrax attacks would not have needed to come in for care if they could have had simple questions answered," said the director of a large emergency room who spoke on the condition of anonymity. "You could lop 20 or 30 percent off the top."

"It's a very exciting and potentially valuable tool," said Michael S.A. Richardson, the D.C. Health Department's chief medical officer. "It gives the public a way to access important information in a crisis so they can act appropriately . . . The physical facilities are not going to be able always to handle large amounts of people efficiently. Large groups of people should not be out on the streets."

Richardson wants the District to anchor the Washington region testing of the software, and triage system leaders at Georgetown and Harvard say they plan to make the national capital area a "test bed" for the project, along with Pittsburgh and Massachusetts. Another group is working on a similar concept in Los Angeles, they said.

But McDonald and Locke said that some experts have doubts. "Somebody is scared to death, and you suggest going to the computer and starting to answer questions," said Ken Alibek, a former Soviet bioweapons expert who defected and now is a biodefense specialist at George Mason University. "Some people would do it, but the great majority wouldn't be able to do it. The major engine is fear."

Concern about the "surge capacity" of health care facilities has sharpened since the anthrax attacks of 2001, when about 20,000 people in the region were given preventive antibiotics, mostly by public health agencies.

Thousands of others went to doctors and emergency rooms complaining of symptoms. Few turned out to be at risk of contracting anthrax. The rest, McDonald said, were somatizing -- experiencing physical symptoms caused solely by psychological distress. Experts say physical ailments rooted in the mind account for up to half of office visits to primary care doctors.

"Even in normal times, somatization is the leading reason why people seek care from a doctor," said Locke, who also is president of the American Psychosomatics Society. "People go to doctors with physical complaints in which careful evaluation fails to reveal an organic cause from one-third to one-half of the time."

In a major outbreak, stress-related symptoms could run wild, causing huge lines and flooding facilities with people who do not need to be there, McDonald and Locke say.

"Conservatively, the somatizers will likely outnumber those actually exposed by 5 to 10 times," according to a report on the screening tool by Harvard researchers supervised by Locke. "However, it is conceivable that this number may in fact be much larger."

Georges C. Benjamin, executive director of the American Public Health Association and former Maryland health secretary, said he knows telephone triage protocols can be effective because he used to write them for private health plans.

"You absolutely can sort people into high-risk and low-risk groups," he said. "The research needs to be done to figure out what you do with people who are giving you exactly the symptoms they read about or saw on television . . . And you have to make sure that you identify people who are both somatizing and [also] sick."

SARS, smallpox, chemical agents, radiological bombs or an entirely new "engineered agent" -- a combination of deadly pathogens -- could trigger a far more intense public response than anthrax did, according to experts working on the triage project.

"When you put a lot of people under stress, you suddenly have a huge surge in demand of people needing help determining whether their symptoms are due to stress or exposure," Locke said.

Models of the questionnaire ask for personal and contact information, take the patient through descriptions of symptoms, inquire whether the patient often somatizes and ask about the patient's attitudes about health care and life.

Dan Hanfling, director of emergency management and disaster medicine for Inova Health System, said of the anthrax scare that brought large numbers of the uninfected people to emergency rooms, "That opens a window of experience on what we are likely to face in the next event." He added, "We need to preserve the ability to deliver acute health care. This will have hiccups, but I think it's a great step in the right direction."

Some say they think the system needs extensive testing. "I think the concept would be fine, but I see all kinds of possible problems," said Philip S. Brachman, an Emory University epidemiology professor who spent 32 years at the Centers for Disease Control and Prevention in Atlanta. "It has to be field-tested very carefully in a random way with good controls by an independent group."