Eighty-four percent of women in a multicenter study who were offered a rapid HIV test during labor agreed to be tested; the test proved to be highly reliable and thus gave women who had not known that they were infected an opportunity to receive treatment aimed at preventing transmission of the virus to their infant during delivery.1 The younger women were, the more likely they were to accept testing; black women and Hispanics had elevated odds of agreeing to take the test.

The study was conducted in 16 hospitals in six U.S. cities between November 2001 and November 2003. Women in labor with no documented HIV test results were offered a test that produces results in 20 minutes, along with standard testing to confirm the result; when it was feasible, those who tested positive were given antiretroviral prophylaxis. Infants born to HIV-infected women were tested for the virus at birth and were followed up for as long as six months. To be eligible for the study, women had to be in active labor or to be at least 34 weeks pregnant. Labor and delivery personnel provided HIV services 24 hours a day; researchers conducted study interviews and collected additional data from participants' medical records.

Of the 5,744 women who were offered rapid HIV testing, 84% were tested and enrolled in the study. Results of logistic regression analyses that adjusted for study site and other relevant factors indicated that women younger than 30 were significantly more likely than their older counterparts to agree to testing; the younger the women were, the wider the differential (odds ratios, 1.4 for women in their late 20s, 1.5 for those in their early 20s and 1.9 for teenagers). Black women had higher odds of agreeing to the test than did white women (1.8), and Hispanics had elevated odds when compared with non-Hispanics (2.4). Women who were less than 32 weeks pregnant were more likely to have the test than were those at more than 36 weeks' gestation (2.0), and women who had not had prenatal care had higher odds of testing than women who had made more than five prenatal visits (1.7). Compared with women admitted for delivery care between eight a.m. and four p.m., those admitted between late afternoon and midnight had reduced odds of opting to be tested (0.7).

For half of women tested, results were available within 66 minutes. (By contrast, the median interval for receipt of enzyme immunoassay results was 28 hours.) Nevertheless, some women who entered the hospital in active labor did not receive test results until after they had delivered. In multivariate analyses, the odds of such delay were markedly increased if the woman delivered within two hours after arriving at the hospital (odds ratio, 34.5); they were significantly, although less dramatically, elevated if she gave birth within 3-12 hours of admission (2.0-5.2). A woman's likelihood of receiving test results after delivery also was elevated if she had waited more than 90 minutes to learn her test result (2.2), if she was admitted between four p.m. and eight a.m. or on a weekend (1.6-2.3), if she was more than 36 weeks pregnant (1.7) or if she had made more than five visits for prenatal care (1.4).

Both the rapid test and enzyme immunoassay identified 34 women who were infected with HIV (for a prevalence of seven per 1,000); neither test yielded any false-negative results. The sensitivity of the rapid test (i.e., the proportion of infected women it correctly identified) was 100%, and its specificity (i.e., the proportion of uninfected women for whom it showed negative results) was 99.9%.

Eighteen of the HIV-infected women received prophylactic zidovudine therapy; the median interval from receipt of the rapid test result to the first dose of the drug was 33 minutes. All 34 infants born to women with HIV received antiretroviral prophylaxis, and 32 were followed up; two were HIV-infected at birth, and one was not infected at birth but tested positive by six weeks of age.

On the basis of these findings, the researchers note, the Centers for Disease Control and Prevention has recommended routine HIV testing for women in labor whose infection status is unknown. They point out, however, that the results are important for other countries as well, particularly in the developing world, where pregnant women whose HIV status is unknown may not see a clinician until they are in labor. "Rapid testing during labor can enable [such] women…to learn their HIV infection status so they can receive antiretroviral prophylaxis and be referred for comprehensive medical care and follow-up."

—D. Hollander


1. Bulterys M et al., Rapid HIV-1 testing during labor: a multicenter study, Journal of the American Medical Association, 2004, 292(2):219-223.