Abstract Body

Partner HIV testing during pregnancy has been difficult to achieve in sub-Saharan Africa as men seldom attend antenatal appointments. Home-based testing may be an effective alternative method to test partners and identify discordant couples in pregnancy. A randomized clinical trial was conducted in Kenya to determine whether home-based partner education and HIV testing (HOPE) during pregnancy results in higher uptake of testing by partners and increased detection of HIV-discordant couples compared to those who receive a clinic invitation letter.

Pregnant women attending a first antenatal visit at Kisumu East District Hospital from October 2012 to May 2013 were randomized to receive the HOPE intervention or a clinic invitation letter for the woman’s partner (INVITE). The HOPE intervention was a scheduled home visit by a male/female pair of community health workers within 2 weeks of enrollment with couple HIV testing and counseling. In both arms, women had follow-up visits at 6 and 14 weeks postpartum at the clinic and couples had 6 month postpartum home visits. Relative risks were calculated between the two arms.

Among 1,101 women screened, 620 (56%) were eligible, and 601 (97%) were enrolled. At enrollment, mean age of women was 24.9 years and 19.1% were HIV positive. Retention was high at 6 months postpartum (88% of women, 86% of men). During the study period, 233 (87%) of 247 men in HOPE reported being tested for HIV during the study period compared to 108 (39%) of 240 men in INVITE (Relative Risk [RR] 2.10; 95% CI: 1.82-2.42). Furthermore, 217 (88%) of 248 women in HOPE knew their partner’s status compared to 98 (39%) of 254 women in INVITE (RR 2.27; 95% CI: 1.93-2.67), 192 (77%) of 248 women in HOPE had been tested as a couple compared to 62 (24%) of 254 women in INVITE (RR 3.17; 95% CI 2.53-3.98) and 33 (13%) of 248 women in HOPE were identified as being in a discordant partnership compared to 10 (4%) of 254 women in INVITE (RR 3.38; 95% CI: 1.70-6.71).

Home partner testing resulted in significantly more partner and couple HIV testing, disclosure, and identification of discordant couples compared to a clinic invitation letter to the male partner. This intervention has the potential for prevention of incident HIV infection in pregnancy and MTCT as well as testing harder to reach populations. There is a need to develop strategies to adapt and scale-up similar interventions in settings with high HIV prevalence and low male partner participation in PMTCT.