A major snag in tests for the virus that causes AIDS is that they often fail when people are newly infected. That may be about to change.

Researchers now say they want to use a genetic test that can unmask HIV cases that might otherwise be missed.

The secondary test should help doctors alert to more HIV-positive people in the early stages of their disease and prevent them from infecting others at a time when they’re highly contagious, said Dr. Frances Priddy, the study lead author and an assistant professor of medicine at Emory University in Atlanta.

“We’re really missing a chance to diagnose these people and counsel them about behavioral changes,” she added.

The findings were to be presented Friday at the 12th Conference on Retroviruses and Opportunistic Infections in Boston.

While no one knows how many HIV tests are given in the United States each year, officials estimate that 40,000 Americans become infected annually. Some 280,000 apparently don’t realize they’re HIV-positive; many are in the dark because it can take years for the immune systems of HIV-positive people to decline.

But people without symptoms can still transmit HIV through sex or shared drug needles. In fact, levels of the AIDS virus in the blood are often extremely high in the first weeks after infection, making infected people very contagious.

Unfortunately, the high “viral loads” don’t always translate into positive HIV test results. Traditional blood and saliva tests only pick up the presence of specific antibodies, the immune system’s soldiers that swing into action when germs are detected.

“People will usually test negative in the first month or more of infection because your body takes several weeks to develop an antibody,” Priddy said.

Enter a genetic test called NAAT, short for nucleic acid amplification testing. The test isn’t new: It’s been used to screen blood-bank donations by detecting bits of virus, according to Priddy. But researchers hadn’t given it much thought on the HIV front until now.

Priddy and colleagues used the test to screen 2,135 people who appeared to be HIV-negative after undergoing the traditional test. All the participants were visitors to Atlanta urban clinics specializing in sexually transmitted disease and drug treatment.

The secondary test turned up four cases of HIV, three of which were new; it wasn’t clear whether the fourth person had been infected recently.

By contrast, the traditional test caught 66 cases of HIV out of a total of 2,202 people screened. “You could probably say we’ll detect 6 to 10 percent more infections (with the genetic test) than we’d get with the regular test,” Priddy said. “That may not sound very exciting, but the point is you’re getting the people with acute (new) infection.”

Priddy said she didn’t know how much the test costs. There is another test that picks up early cases of HIV, but it’s expensive and used by doctors who suspect their patients are infected.

The news about the success of the genetic test is promising, said A. David Paltiel, chairman of the division of health policy and administration at Yale University.

However, he thinks the big picture remains the same: HIV tests aren’t given often enough.

In a study released earlier this month, Palatial and colleagues reported that it’s cost-effective to offer screening to all Americans.

“To the extent we make it routine, we’ll reduce the stigma associated with getting an HIV test,” he said.