P46.05
Background: Relationship dissolution among HIV serodiscordant couples enrolled in clinical trials bears a negative impact on adherence to study product and retention to clinic visits. We evaluated characteristics of couples that separated while enrolled in HIV prevention clinical trial in Kenya
Methods: We conducted univariate analysis to determine factors among HIV discordant couples associated with ever separating while enrolled at the Thika site in the Partners PrEP Study, a randomized controlled trial conducted in Kenya and Uganda. Participants reported relationship status at each visit. Data on relationship stability was extracted from the counselors' chart notes.
Results: Of the 496 HIV discordant couples enrolled, 98 (20%) reported ever separating. In relationships reporting separation, HIV uninfected persons were 34 [IQR 28, 39] years, 77% were male and duration of relationship was 3 [IQR1, 7] years. Compared to couples with children together, those without were likely to separate (30.1% vs 15.0%, OR=2.4, 95%CI 1.6-3.8, p<0.0001). Couples who did not consider themselves to be married (36.4% vs 18.6%, OR=2.5, 95%CI 1.2-5.2, p=0.016) and couples who had known their discordant status for less than a year at baseline (21.7% vs 13.1%, OR=1.8, 95%CI 1.0-3.4, p=0.05), were more likely to separate. Compared to monogamous relationships, people in polygamous unions were twice as likely to separate (OR=1.9, 95%CI 1.3-3.02, p=0.001). Couples who had stayed together longer were less likely to separate (P=0.001).
Conclusions: A significant proportion of HIV discordant couples enrolled in HIV prevention trials are likely to separate. The impact of this separation on retention to care should be explored.
