HIV voluntary counseling and testing at an abortion clinic: missed opportunities for diagnosis.

Background: Most abortion clinics in the US do not provide opt-out HIV testing, and thus women electing abortion may not receive adequate screening for HIV. We sought to determine whether the standard practice of offering voluntary HIV counseling and testing (VCT) services to women electing abortion results in detection of undiagnosed HIV infection.

Study Design: Two thousand twenty-five women electing abortion were consecutively offered VCT at a public hospital-based, urban abortion clinic. Unlinked, anonymous HIV surveillance testing was conducted subsequent to voluntary testing without patients knowing of the surveillance study testing at the time. Multivariable analysis was used to determine factors associated with opting for VCT.

Results: Of the 13 women (0.6%) who tested positive for HIV, 8 had been previously diagnosed. Of the 5 women without previous diagnosis, only 1 opted for VCT. Overall, 530 (26%) women opted for VCT. Factors associated with opting for VCT in multivariable analysis (odds ratio, 95% confidence interval) were sex with an injection drug user (1.9, 1.1-3.2), presenting in the second trimester (1.3, 1.0-1.6), economic hardship (1.3, 1.0-1.5), partner difficulties (1.8, 1.3-2.4) and having public health insurance (1.6, 1.2-2.3).

Conclusions: HIV testing via VCT did not effectively diagnose previously undiagnosed HIV infections among women seeking abortion in this clinical setting.