Differences in HIV testing at delivery among women giving birth in public and private hospitals in Texas

Background: In Texas, opt-out HIV testing is mandated to be offered to pregnant women at the first prenatal visit and at delivery and reported on the birth certificate. . This study describes differences in prenatal and delivery HIV testing among women giving birth in Texas hospitals in 2001.

Methods: Data were obtained from birth certificate records and medical charts of a stratified random sample of 1499 women (and children) within 26 public and private hospitals in Texas. Analyses were conducted as a three-level hierarchical linear model to test differences in receipt of prenatal and delivery HIV testing. Hospital variables included funding type (Level 3) and service volume (Level 2). Level 1 covariates included race/ethnicity and prenatal HIV test. The outcome variable was having no test at delivery or missing test.

Results: The conditional probabilities of “no test” or “missing test” at delivery was higher for Caucasian women who had received a prenatal test (beta=0.29; p=0.00). High service volume was significantly associated with having received no or missing test at delivery .(beta=0.53; p=0.01). The probability of having received a prenatal HIV test decreased for public hospitals (beta = -0.42; p=0.00).

Conclusions: These data show that delivery HIV testing fluctuates with the service volume and prenatal testing fluctuates with funding source of the hospital, and that HIV testing at delivery is a function of race/ethnicity and prenatal HIV testing. Additionally, HIV testing requirements at delivery should be followed regardless of perceived risk associated with racial/ethnic membership or existence of a prior HIV test.