In 1994, the US Public Health Service (PHS) recommended the use of zidovudine during pregnancy to prevent the perinatal transmission of HIV. In 1995, the agency recommended that all pregnant women be counseled and encouraged to undergo voluntary HIV testing. To evaluate the implementation of those guidelines, the researchers analyzed data from the Behavioral Risk Factor Surveillance System (BRFSS) for the years 1994 through 1999. In each state, data are collected monthly through telephone interviews with a random sample of adults age 18 or older. The researchers restricted the sample to women ages 18 to 44. Data were weighted and aggregated across states for each year to obtain national estimates.

During each year, the sample comprised more than 30,000 women, and most of the characteristics did not change over time. A majority were white (70-76 percent), married (56-58 percent), high school graduates (89-90 percent) with health insurance (81-84 percent) who had undergone a Papanicolaou test (92-93 percent). The percentage of women who were pregnant did not change from 1994 to 1999 (range: 4.4-5.1 percent).

Following the introduction of the PHS guidelines, HIV testing among pregnant women increased from 41 percent in 1995 to 56 percent in 1999. During the period, the rate of testing among non-pregnant women remained fairly stable, suggesting that the guidelines had an effect. Even so, the fact that just over half of pregnant women were tested suggests that more education efforts are needed. The percentage of pregnant women who reported HIV testing was lower than in several studies of women recruited in prenatal clinics, each of which reported percentages of more than 70 percent. Similarly, a population-based sample of postpartum women in several states found that 58-81 percent of women interviewed in 1996 and 1997 were tested. The BRFSS does not ask whether a test was offered or refused; the lower rate of testing in this study may result from the inclusion of women who were not offered a test, or who refused.

Younger age, not being married or employed, living in the South, and having health insurance were associated with having been tested for HIV. Regional differences in testing prevalence rates may be influenced by differences in state legislation regarding counseling and testing pregnant women. HIV/AIDS seroprevalence may also affect testing. Of the four regions, the South accounts for the largest number and proportion of AIDS cases.

"Monitoring of HIV testing among pregnant women will continue to be important to recognizing changes in the prevalence of testing. These data are useful, along with those from HIV/AIDS surveillance, in assessing the effect of new policies or guidelines such as the Institute of Medicine's recommendation for universal HIV testing as a routine component of prenatal care," the authors concluded.

Back to other CDC news for August 20, 2001

Previous Updates | Search the CDC archive