• When asked whose idea it was to get tested, 42% of reproductive-age women report a doctor or other health care provider, 39% report self, and 7% report an insurer (Table J.7).
• Pregnant women are more than twice as likely as non-pregnant women to report that they were recently tested for HIV upon the recommendation of a health care provider (70% versus 30%) (Table J.8).

• According to multivariate analyses, different factors are predictive of HIV test use for pregnant and non-pregnant women. Among women who are not pregnant, being at HIV risk, African American, poor, living in a metropolitan area, being age 20–39, highly educated, and having been formerly married increase HIV test use. Decreased HIV test use occurs among teenagers and residents of the Northeast and Midwest. Being at HIV risk triples the odds of HIV test use among non-pregnant women (Table J.9).
• Among pregnant women, many of the sociodemographic predictors of HIV testing observed among non-pregnant women lose significance (i.e., age, race, poverty, and metropolitan area residence). This suggests that pregnancy is serving as a triggering event for testing, irrespective of the woman's characteristics. The role of education is reversed for pregnant women. Here, lower educational attainment is predictive of HIV testing. Different geographic patterns emerge for pregnant women, with residents of the South more likely to be tested than residents of other areas. Being at HIV risk quadruples the odds of HIV testing among pregnant women (Table J.9).

National Center for Health Statistics. Report of final mortality statics, 1995. Monthly Vital Statistics Report 1997; 45(11:Suppl 2).

Wilson JB. Human immunodeficiency virus antibody testing in women 15–44 years of age: United States, 1990. Advance Data from Vital and Health Statistics. Number 238, Hyattsville, Md.: National Center for Health Statistics, 1993.