that in 1995, they offered testing to about 73% of their pregnant patients; they counseled 67%; and they tested about 39%. It is interesting to note that despite the fact that it is not legal to test in Massachusetts without counseling, clearly this is happening in some practices. Also, it is clear that far fewer women are tested than are counseled or offered testing. (2) Having an HIV clinical practice policy in place is the single best predictor of whether a provider counsels, offers, or performs a test. Client characteristics are also predictors of whether women are offered or receive testing in Massachusetts. Specifically, African Americans are more likely to be offered a test; Hispanics are more likely to be tested; and privately insured patients are less likely to be tested. Ms. Allen noted that these findings indicate that providers continue to use a risk assessment model. She observed that providers do not seem to be getting a clear message about what the PHS guidelines say: that is, they think they are following the guidelines when they counsel based on risk. (3) Survey findings indicate that provider attitudes do not seem to make a difference in whether the provider counsels, offers a test, or tests; however, they do make some difference in the likelihood of the practice having a policy in place.
In a separate but parallel study conducted among HIV-infected women who had experienced pregnancy, women were asked whether they thought testing should be mandatory. Nearly all—24 of 26 interviewees—said yes, "because of the baby." Ms. Allen noted that this finding should be taken as evidence of the strong feelings HIV-infected women have about their babies, not necessarily as the best public policy to pursue. The patient interview also indicated that HIV-infected women want to have a good relationship with their providers and that providers can greatly influence patients' decisions. However, it appears that often providers do not recognize the importance of this relationship. Finally, Ms. Allen noted that having a case manager can influence women's acceptance of testing, particularly women who are not from the dominant culture.
Rachel Royce, an epidemiologist from the School of Public Health, University of North Carolina, presented an overview of efforts in her state to prevent perinatal HIV transmission through prenatal HIV counseling and testing. She presented results of a survey of prenatal care providers and a study of women offered testing during prenatal care.
Immediately after the ACTG 076 results were reported, North Carolina's health officer sent a letter to all prenatal care providers in the state informing