CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

indicating the need for developing HIV prevention strategies tailored to this population.

Northwestern University, Chicago, IL, USA Background: Pre-exposure prophylaxis (PrEP) is highly efficacious at preventing HIV but is dependent upon optimal adherence, including sustained use during high risk periods. PrEP uptake is escalating among young men who have sex with men (YMSM) and transgender women (TW), but evidence suggests that up to one-third of YMSM/TW PrEP users discontinued use in a 6-month period, which eliminates its protective benefit. The current analyses examined longitudinal predictors of PrEP discontinuation. Methods: Data came from RADAR (N=1100+), an ongoing longitudinal cohort of YMSM/TW (aged 16-29) in Chicago. Using data from 7 visits at 6-month intervals (collected 2015-2018), mixed effects longitudinal regression models examined change in sexual behaviors and psychosocial factors as predictors of PrEP discontinuation (i.e., use at prior visit, no use at current visit). Predictors included change from the prior to current visit in condomless anal sex (CAS), number of sex partners, relationship status, substance use, and depression, as well as current insurance status. Models adjusted for demographic characteristics. Results: PrEP use among HIV-negative YMSM/TW increased from 8.4% (visit 1) to 28% (visit 7). PrEP discontinuation similarly increased from 12.6% (visit 2) to 20% (visit 7). In a multivariate model, YMSM/TWwho had increases in CAS across visits were less likely to discontinue PrEP (Odds Ratio [OR]=0.93, 95% Confidence Interval [CI]: 0.89-0.98), while those who entered a serious relationship were more likely to discontinue (OR=1.85, 95% CI: 1.08-3.19). Number of sex partners, substance use and depression were not associated with discontinuation. We observed no race or gender identity differences in PrEP discontinuation. In a separate model, we examined the association between current insurance status and discontinuation, adjusting for demographics. YMSM/TWwho had insurance were significantly less likely to discontinue PrEP (OR=0.54, 95% CI: 0.32-0.92). Conclusion: That YMSM/TWwho increase CAS are less likely to discontinue PrEP is encouraging. Among those entering relationships, it remains unclear how and when YMSM/TW discontinue PrEP and whether or not transmission risk remains after discontinuation. Insurance status is a key structural determinant of the ability to sustain PrEP use and reduce transmission risk. These findings point to encouraging trends and opportunities for structural and behavioral intervention. Uwe Koppe 1 , Ulrich Marcus 1 , Stefan Albrecht 1 , Klaus Jansen 1 , Heiko Jessen 2 , Barbara Gunsenheimer-Bartmeyer 1 , Viviane Bremer 1 1 Robert Koch Institute, Berlin, Germany, 2 Praxis Jessen² + Kollegen, Berlin, Germany Background: Use of HIV pre-exposure prophylaxis (PrEP) is increasing, but some users discontinue PrEP. We investigated former PrEP users in Germany and compared them to current PrEP users in order to elucidate reasons for stopping and implications for HIV prevention. Methods: From 24th July to 3rd September 2018 we recruited current and former PrEP users on geolocation dating apps for MSM, community-based HIV testing sites, and a community website in Germany for an anonymous online survey. Risk factors were assessed with logistic regression models adjusting for age, country of origin, and annual gross income. Results: We recruited 212 former PrEP users and 2,005 participants currently taking PrEP. 78.7% completed the survey. Most participants identified as male (99.1%, trans*: 0.4%, intersexual: 0.3%, non-binary: 0.2%) and indicated Germany as their country of origin (74.8%) with no significant differences between current and former PrEP users. The median age of former PrEP users (33 years, IQR: 27-41) was lower than of current PrEP users (38 years, IQR: 31-45). The reasons for discontinuing PrEP are shown in Table 1 (multiple responses allowed). Former PrEP users were much more likely to have used PrEP intermittently or on demand (OR = 2.8, 95% CI 2.0, 4.0). In addition, former PrEP users were more likely to be unhappy with their current sex life (OR = 4.1, 95% CI 2.6, 6.6). Most former PrEP users indicated that they always (35.8%) or often (27.9%) use condoms since stopping PrEP, whereas 35.8% indicated using condoms during half or less of their sexual acts. Compared to current PrEP users, former users were more likely to always or often use condoms (OR = 7.9, 95% CI 5.4, 11.6). Conclusion: The analysis identifies important reasons for discontinuing PrEP, some of which could be overcome if PrEP were covered by health insurances. More than a third of former PrEP users reports inconsistent condom use

991 FACTORS ASSOCIATED WITH REFUSING OR STOPPING PrEP AMONG AT-RISK MSM IN KENYA Elizabeth Wahome 1 , Grace Mwashigadi 2 , Bernadette Kombo 1 , Makobu Kimani 1 , Elisabeth M. van der Elst 1 , Khamisi Mohamed 1 , Mark Brockman 2 , Susan M. Graham 3 , Eduard Sanders 1 1 KEMRI Wellcome Trust Rsr Prog, Kilifi, Kenya, 2 Simon Fraser University, Burnaby, BC, Canada, 3 University of Washington, Seattle, WA, USA Background: There are limited data on reasons for refusing or stopping programmatic pre-exposure prophylaxis (PrEP) among men who have sex with men (MSM) in Kenya, a country rolling out PrEP since May 2017. We assessed factors associated with refusing or stopping PrEP in this population, using a mixed methods approach. Methods: Since June 2017, at-risk MSM followed at monthly visits in an HIV-1 vaccine feasibility cohort study in the coastal Kenya were offered PrEP with adherence and risk reduction counselling, monthly rapid HIV-1 antibody testing and X-pert RNA Qual testing if acute HIV-1 risk criteria were met. We assessed factors associated with refusing or stopping PrEP at the last available visit for those who refused PrEP and the date of PrEP discontinuation for those who stopped using generalized linear modeling with log-link Poisson regression and robust error variance. Variables associated with refusing or stopping PrEP at P<0.2 in the bivariable analysis were included in the multivariable model. We also conducted 2 focus groups discussion (FGDs) and 12 in-depth interviews among purposively sampled MSM who were eligible but did not start (N=6) or discontinued PrEP (N=6). Interviews and FGDs were recorded, transcribed and analyzed using a grounded theory framework. Results: Of 178 MSM offered PrEP, 36 (20.2%) did not start and 142 (79.8%) started, of whom 31 (17.4%) stopped after a median of 4.3 (interquartile range: 1.7–8.9) months. In multivariable analysis, paying for sex (adjusted prevalence ratio [aPR] 1.6, 95% CI 1.0–2.5) was an independent predictor of refusing or stopping PrEP, after adjustment for religion and self-reported unprotected sex, anal sex position, and receipt of payment for sex. In qualitative analysis, participants who had refused or had stopped PrEP showed limited knowledge and misconceptions about PrEP, and often had low perception of HIV risk. Pill burden, side effects, stigma, and storage challenges were cited as reasons for stopping. There was a strong preference for long-acting oral or injectable PrEP as alternatives to daily oral PrEP. Conclusion: Over one third of at-risk MSM followed up in the cohort study refused or stopped taking PrEP. MSM reporting paying for sex may have low perceived HIV-1 risk. Ongoing community engagement and education are needed to correct misconceptions, raise awareness, and decrease stigma in order to improve uptake and continuous use of PrEP among Kenyan MSM.

Poster Abstracts

990 STOPPING HIV PREEXPOSURE PROPHYLAXIS: REASONS AND IMPLICATIONS

CROI 2019 388