Testing Times

The 27-year-old woman had just delivered a healthy baby girl at Cook County Hospital in Chicago when obstetrician Julie B. Schmidt entered the maternity suite bearing bad news. Just an hour earlier, in the throes of labor, the young woman had consented to an HIV test. Now, Schmidt had the result. As gently as possible, the doctor told the mother she had the virus. The new information demanded immediate action: With the right course of drugs and a dose of luck, the newborn could be saved from infection.

Ideally, drug treatment for a pregnant, HIV-infected woman and her fetus begins months ahead of delivery. That regimen reduces the chance of mother-to-child transmission to less than 1 percent. Without a diagnosis or antiviral drugs, that risk can soar to 25 percent.

The physicians at Cook County Hospital capitalized on their eleventh-hour discovery and gave the newborn drugs immediately. The baby escaped infection.

Medicine may have helped avert one tragedy that day, but such close calls underscore the need to discover more HIV infections earlier, and not just among pregnant women. People who are diagnosed with HIV infection soon after they acquire it get the most benefit from state-of-the-art therapies that can extend their lives by years.

The rest miss out on life-sustaining treatments and are at high risk of spreading the virus. An estimated 180,000 to 280,000 people in the United States–about one-quarter of those infected with HIV–don’t know it, and they cause about two-thirds of new infections, estimates Robert S. Janssen, who works on HIV and AIDS prevention at the Centers for Disease Control and Prevention (CDC) in Atlanta.

By and large, Janssen says, “when people learn that they’re infected, they take steps to reduce the risk of transmitting to partners.” That explains why the government’s current HIV-prevention initiative aims to shorten the time it takes to identify people infected with the virus.

A new rapid blood test, which helped diagnose Schmidt’s patient, is playing a central role in that effort. The test is being applied by a makeshift militia of researchers, health-care providers, and community-based health workers, trained to use it.

Some HIV fighters caution that hurried and imperfect implementation of rapid testing could mean that some people will get life-shattering information but won’t receive the life-saving care that should follow.

To help make decisions about whom to test, researchers are working to identify groups of people that are likely to have high proportions of undiagnosed HIV cases. Schmidt’s patient belonged to three of those groups: She was poor, African American, and presumably infected through heterosexual contact.

Rapid results

If Schmidt’s patient had received standard prenatal care, she might have been diagnosed months earlier. To help head off cases of mother-to-child HIV transmission, many obstetricians routinely offer HIV tests to women early in their pregnancies. Some states require such tests. But Schmidt’s patient, like many expectant mothers from inner cities, showed up for care only when she was in labor.

That’s one situation where rapid testing is invaluable, says Mardge H. Cohen of the Cook County Bureau of Health Services in Chicago. Last November, the Food and Drug Administration approved the OraQuick Rapid HIV-1 Antibody Test, which is manufactured by Abbott Laboratories of Abbott Park, Ill. Cohen runs the Chicago arm of a CDC-funded nationwide study that provided OraQuick kits to Schmidt’s staff.

“The case for rapid testing is pretty clear in labor and delivery,” says Cohen. Of hundreds of pregnant women to whom the investigators offered rapid HIV tests at four Chicago hospitals, including Cook County Hospital, nearly 80 percent accepted. About 1 percent have tested positive for HIV, and most got their results within an hour, Cohen and her colleagues report in the Sept. 12 Morbidity and Mortality Weekly Report. Had they used conventional HIV testing, none of those diagnoses could have been made in time for doctors to deliver antiviral treatment in the crucial hours around childbirth.

Diagnosing an HIV infection requires at least two steps. Conventionally, an initial blood screening, known as an HIV ELISA assay, can give either a negative or a preliminary positive indication that a person’s immune system has encountered HIV. The preliminary positive result typically leads to a repeat ELISA test and must be confirmed by another, more specific HIV test to rule out the possibility of false-positive results.

Because both tests must be done by specialized staff using sophisticated lab equipment, getting a result back to someone who has been tested can take 2 weeks.

The OraQuick rapid test, in contrast, can provide a person who may have HIV with nearly as reliable a result in under an hour and with little fuss. Its self-contained, handheld testing kit consists of a readout device, a loop for collecting a blood sample–or, in a version under FDA review, a saliva sample–and a vial containing a developing solution. Most health-care workers, once familiar with the test, can process a sample and interpret the results within 20 minutes. Data that FDA reviewed before approving the test in 2002 indicate that results from the rapid test are accurate more than 99 percent of the time. Still, any positive OraQuick result must be confirmed with a slower, more specific test to rule out even the small chance of error.

In January, FDA approved the use of rapid-testing kits in tens of thousands of facilities that weren’t previously capable of analyzing HIV test results in-house. Since then, state and federal health agencies have funded rapid testing in settings such as physicians’ offices, jails, sexually transmitted disease (STD) clinics, and mobile testing units that visit nightclubs, gay-pride events, and places where prostitutes and drug abusers congregate.

Missed opportunities

One-hour HIV testing will be instrumental not just in low-income maternity wards, but also throughout inner cities where people seek health care only sporadically, says Sam Niedbala of OraSure Technologies, the Bethlehem, Pa., company that created the OraQuick test. Niedbala says that the quickness of the test can prevent a frequent and frustrating outcome of conventional testing: About 20 percent of HIV-infected people fail to return to a clinic or doctor’s office for the test results.

“With the rapid test,” Niedbala says, “they can have a test done . . . while they’re standing there.”

“Rapid testing certainly has a role in emergency rooms and [STD] clinics, where you get one crack at the patient,” says Daniel Klein, chief of infectious diseases at Kaiser Hospital in Hayward, Calif. However, he says, it’s less critical at his hospital and others that serve relatively affluent populations because patients who request HIV tests tend to return for the results. What’s needed in such settings, says Klein, is more frequent use of conventional tests among people who don’t request them but whom doctors suspect to be at substantial risk of infection.

To understand what clues doctors could use to identify people who might benefit from HIV testing, Klein and his colleagues combed through medical charts of 440 people whom doctors at the hospital had diagnosed with HIV in 1998. Men, ethnic minorities, and people who got tested only after a physician suggested it generally had more severely ravaged immune systems than did other HIV-positive people. The degree of immune deficiency in a person correlates with how long HIV has gone undetected. The researchers reported their findings in the Feb. 1 Journal of Acquired Immune Deficiency Syndromes.

Another pattern also stood out. A year or more before testing positive for HIV, “a significant minority of patients had STDs but weren’t tested for HIV,” Klein says. If all patients treated for STDs had been tested for HIV at the same time, many HIV cases might have been turned up earlier, he contends.

The implication, Klein says, is that “you get your greatest bang for your buck if you go after people with STDs” or a history of STDs and no corresponding HIV tests. “Someone who had gonorrhea 10 years ago and is now monogamous and happily married could have picked up HIV 10 years ago,” he says.

The CDC data also point to factors associated with late diagnosis, which researchers typically define as that occurring less than a year before AIDS symptoms set in. Hispanics and African Americans are about twice as likely as non-Hispanic whites to get a late HIV diagnosis, researchers reported in the June 27 Morbidity and Mortality Weekly Report.

Heterosexuals who don’t use intravenous drugs have a relatively low risk of getting HIV, but if they do become infected, they are more than twice as likely as men who have sex with other men, and 4 times as likely as intravenous-drug users, to have their infection diagnosed late. Many belatedly diagnosed heterosexuals are immigrants who may have been infected overseas and who have limited access to health care in this country.

Many people are diagnosed late because they wait to get tested until they show symptoms of AIDS, which can take a decade to develop following infection with HIV, says CDC’s Janssen.

Not so fast

Like Klein, Rochelle P. Walensky of Massachusetts General Hospital in Boston considers widespread testing the key to earlier diagnosis, whether or not it involves rapid testing. Rather than having test providers try to single out people who may be infected, she advocates systematically offering tests to all patients in settings where HIV is prevalent.

Before such all-patient screenings became routine at Boston Medical Center, Walensky says, 9.3 people sought testing each month on average, and 1.3 people tested positive. After staff began preemptively offering tests to all patients in 1999, an average of 33.8 got tested each month and 2.3 HIV infections were diagnosed monthly, Walensky and her colleagues reported in the April 22, 2002 Archives of Internal Medicine. Not only does that get more infected people into treatment, but it also cuts off potentially extensive branches of HIV transmission.

At the February 2003 Conference on Retroviruses and Opportunistic Infections in Boston, Walensky and her colleagues reported that 1.6 percent of preemptive HIV tests at Boston Medical Center and three other Massachusetts hospitals resulted in an HIV diagnosis, whereas statewide only 0.9 percent of tests that people had sought out themselves led to an HIV diagnosis.

While advocating widespread testing, Walensky cautions against switching too hastily to rapid tests. Despite the relatively high precision of the OraQuick assay, she says, its results aren’t definitive without confirmation.

Furthermore, says Jean F. McGuire, a former assistant commissioner of public health in Massachusetts and one of Walensky’s research collaborators, the sheer number of rapid tests being conducted means that a small but significant number of false positives are turning up. False positives “create incredible turmoil for clinicians” and undermine public confidence in health services even though confirmatory testing soon reveals the error, she says.

Also, rapid testing’s implementation might outpace the delivery of services that connect HIV-infected people with medical care, McGuire says. She also worries about ad hoc testing at, for example, bars and nightclubs. “I’m troubled about people who might be in altered states making decisions” to get tested, she says.

If full-service counseling and referral services for further testing and care aren’t on hand where people get tested, they may not know what to make of positive results, McGuire says. Even with rigorous current standards for posttest counseling, one in three people in Massachusetts waits at least a year after being diagnosed with HIV before seeking medical care, McGuire says. That fraction could grow if hurried, haphazard counseling accompanied rapid testing.

She encourages rapid testing that’s conducted by organizations that have fine-tuned their procedures for testing and counseling and that have strong roots in a community. In the end, she points out, it’s what people do with the results of a medical test that determines whether testing makes a difference.

****************

If you have a comment on this article that you would like considered for publication in Science News, send it to editors@sciencenews.org. Please include your name and location.