Abstract
The Centers for Disease Control and Prevention and the American Congress of Obstetricians and Gynecologists recommend universal prenatal HIV testing to prevent perinatal HIV transmission in the U.S.; since the 1990s perinatal HIV transmission has declined. In 2006, 74% of women with a recent live birth reported testing for HIV prenatally or at delivery. We used Pregnancy Risk Assessment Monitoring System data from 36 states and New York City from 2004 to 2013 (N = 387,424) to assess characteristics associated with lack of self-reported testing and state-to-state variability in these associations. Overall, 75.2% (95% confidence interval [CI] 75.0-75.5) of women with a recent live birth reported an HIV test. There were significant differences in testing prevalence by state, ranging from 91.8% (95% CI 91.0-92.6) in New York to 42.3% (95% CI 41.7-43.5) in Utah. In adjusted analysis, characteristics associated with no reported testing included being married, white, non-Hispanic, multiparous, not smoking during pregnancy, and having neither Medicaid nor Special Supplemental Nutritional Program for Women, Infants, and Children. White married women were 57% (adjusted prevalence ratio [aPR] 1.57, 95% CI 1.52-1.63) more likely to report no test compared to white unmarried women. Multiparous married women were 57% (aPR 1.57, 95% CI 1.51-1.64) more likely to report no test compared to multiparous unmarried women. Women who were married, white, non-Hispanic, and multiparous women were 23% less likely to be tested than other women combined. Marital status was significantly associated with lower prevalence of testing in 35 of the 37 reporting areas, and race was significant in 30 of 35 states with race information. The prevalence of reported HIV testing during pregnancy or at delivery remains below 80%. Opportunities exist to increase HIV testing among pregnant women, particularly among certain subpopulations.