CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

1020 COMPARING THE CHARACTERISTICS OF BRAZILIAN MSM USING APP FOR SEXUAL ENCOUNTERS Thiago S. Torres 1 , Raquel B. De Boni 1 , Mauricio Vasconcelos 2 , Paula M. Luz 1 , Brenda Hoagland 1 , Ronaldo I. Moreira 1 , Alex Garner 3 , Valdilea Veloso 1 , Beatriz Grinsztejn 1 1 Instituto Nacional de Infectologia Evandro Chagas, Rio de Janeiro, Brazil, 2 Escola Nacional de Ciências Estatísticas, Rio de Janeiro, Brazil, 3 Hornet Networks, Los Angeles, CA, USA Background: Geosocial networking (GSN) smartphone applications (apps) are becoming the main venue for sexual encounters among Brazilian men who have sex with men (MSM). As HIV infection risk has been increasing among MSM, the Brazilian government recently adopted pre-exposure prophylaxis (PrEP) which will be provided by the National Health System. Accordingly, an understanding of the profile of MSM who use apps for sexual encounters is needed to design combined prevention interventions tailored for this group. Methods: Two cross-sectional online surveys using the same questionnaire were conducted in July 2016 and July 2017. Inclusion criteria were ≥ 18 years of age, cisgender, HIV uninfected and residence in any of the ten selected Brazilian capitals Chi-square tests were used to compare characteristics of MSM recruited in the two studies. Results: A total of 2800 and 2497 MSM completed the survey in 2016 and 2017, respectively. Median age was 30 years (IQR25-36) in both years, but more MSM aged ≤24 years were included in 2017 (26.1% vs. 29.0; p=0.036). Compared to 2016, participants with the following characteristics were more represented in the 2017 survey: non-white (36.5 vs. 42.5; p<.001), bisexual (9.2% vs. 12.9%; p<.001), less educated (<12 years: 41.4% vs. 49.5%; p<.001) and lower monthly income (≤3 Brazilian MinimumWage - USD 250: 34.1 vs. 45.4; p<.001). Scores on “The HIV Incidence Risk for MSM” scale (≥10 points: 66.5% vs. 68.2%, p=0.19) and STI diagnosis (12.4% vs. 14.0%; p=0.08) were similar. Perceived likelihood of getting HIV was higher in 2017 (21.1% vs. 25.3%, p=0.001), although more MSM reported never testing for HIV (14.3% vs. 20.3%, p<.001) and receiving money for sex (3.3% vs. 5.5%; p<.001). Having a male steady partner (18.0% vs. 19.3%; p=0.22), binge drinking (71.7% vs. 73.6%; p=0.13) and stimulants use (22.5% vs. 20.6%; p=0.08) were similar. Awareness of PrEP and nPEP was lower in 2017, while willingness to use oral daily PrEP and non-occupational post-exposure prophylaxis (nPEP) was higher. Willingness to use PrEP injection and HIV self-testing (HIVST) was higher in 2016 (Table 1). Conclusion: Our results suggest that the use of GNS apps seems to have spread across younger and less educated MSM in 2017, which may explain the lower PrEP awareness in this year. As this population stratum is at the highest risk of HIV acquisition in Brazil, mobile health tools are a promising means to reach these individuals in order to increase awareness and demand of new prevention strategies.

based on 3 risk perception questions. The survey also generated an HIV risk score (CalcR), which estimated an individual’s 1 year risk of acquiring HIV compared to the average risk for MSM based on reported condomless anal sex acts, STIs and needle sharing events. SPR and CalcR scores categorized risk as low, medium, high and very high. Participants randomized 1:1 to the intervention armwere provided the CalcR category; control subjects received standard risk reduction counseling. Participants were contacted at 8 weeks to determine if they had started PrEP (primary endpoint). Fisher’s exact test was used to compare the difference in PrEP uptake between arms. Cohen’s kappa coefficient evaluated the agreement between the 2 risk scores. Results: Of 171 participants (n=85 control and n=86 CalcR intervention), median age was 32, 29% Latino, 60%White, 8% Black. Most had heard of PrEP (81%), and 57% thought they were good PrEP candidates. Participants had a median of 5 partners in the past 6 months (IQR: 3-10). SPR had poor agreement with the CalcR score (kappa=0.176) with more than one-third underestimating their HIV risk. At week 8, n=135 participants were reached for follow up, notably n=59 (70%) in control and n=76 (87%) in the intervention arm (p<0.001). Only 14 subjects reached for follow up initiated PrEP with no difference between randomized groups (CalcR 11%, control 10%, p=1.0). The most common reasons for not starting PrEP were low risk perception (36%), and not wanting to take pills (18%); 13% reported waiting for a PrEP visit. Conclusion: In this cohort of at-risk MSM, providing an objective HIV risk score alone did not increase PrEP uptake, perhaps due to discordance between self- perceived and actual HIV risk. Further, many participants did not think their risk was high enough to use PrEP. Thus, HIV testing may be a crucial time to correct misperceptions about HIV risk and initiate same day PrEP to facilitate greater PrEP uptake. 1022LB DISTRIBUTION OF ACTIVE PrEP PRESCRIPTIONS AND THE PrEP-TO-NEED RATIO, US, Q2 2017 Aaron J. Siegler 1 , Farah Mouhanna 1 , Robertino Mera Giler 2 , Scott McCallister 2 , Howa Yeung 1 , Jeb Jones 1 , Jodie L. Guest 1 , Michael Kramer 1 , Cory Woodyatt 1 , Elizabeth Pembleton 1 , Patrick S. Sullivan 1 1 Emory University, Atlanta, GA, USA, 2 Gilead Sciences, Inc, Foster City, CA, USA Background: Cumulative unique persons starting oral TDF/FTC for PrEP in the United States since 2012, including those actively on PrEP and those who have discontinued PrEP, is estimated to be 140,000. This study is the first to describe the magnitude and distribution of active PrEP prescriptions. Methods: Data on active PrEP prescriptions, defined as ≥1 day of PrEP in Q2 2017 for unique persons, were generated from a national prescription database. An algorithm that includes a minimum 30 day prescription period was used to identify each TDF/FTC for PrEP prescription. Active PrEP prescriptions were calculated per population (PrEP prevalence) by region, gender, and age. HIV diagnoses from 2016, based on CDC surveillance data, were used as an epidemiological proxy for PrEP need. The ratio of PrEP prescriptions per new HIV diagnosis (PrEP-to-need ratio) was used to describe the distribution of prescriptions relative to need. Results: A total of 61,298 unique individuals had active PrEP prescriptions: 58,603 male and 2,695 female; 6,422 aged ≤24; 24,144 aged 25-34; 15,197 aged 35-44; 10,786 aged 45-54; and 4,866 aged ≥ 55. Nationally, PrEP prevalence was 23.2/100,000 and the PrEP-to-need ratio was 1.5. Males had higher prevalence (45.5/100,000) than females (2.0/100,000), and more than four times the PrEP-to-need ratio (1.8 and 0.4). Persons aged ≤24 had low prevalence (12.3/100,000) and low PrEP-to-need ratio (0.8). The Northeast region had the highest prevalence (38.5/100,000), and the Midwest (18.7/100,000) and South (18.8/100,000) the lowest. The PrEP-to-need ratio was three times higher in the Northeast (2.9) than in the South (0.9). States with Medicaid expansion had higher prevalence (27.0/100,000) than states without expansion (17.1/100,000), and more than double the PrEP-to-need ratio (2.1 and 0.9).

Poster Abstracts

1021 HIV RISK PERCEPTION AMONG MEN WHO HAVE SEX WITH MEN: A RANDOMIZED CONTROLLED TRIAL Jill Blumenthal 1 , Donald E. Mulvihill 1 , Sonia Jain 2 , Sara Graber 1 , Marvin Hanashiro 1 , Eric Ellorin 1 , Xiaoying Sun 2 , Richard Haubrich 3 , Sheldon Morris 1 1 University of California San Diego, San Diego, CA, USA, 2 University of California San Diego, La Jolla, CA, USA, 3 Gilead Sciences, Inc, Foster City, CA, USA Background: Despite greater access to PrEP, one barrier to HIV prevention is inaccurate risk perception by MSM. Providing information about objective HIV risk could improve PrEP uptake. Methods: PrEP Accessability Research and Evaluation 2 (PrEPARE2) was an RCT examining if providing a calculated HIV risk score affects PrEP uptake in at-risk HIV- MSM recruited from HIV testing sites. Participants received a baseline survey assessing demographics, risk behaviors and HIV risk perception. Self-perceived risk (SPR) score was the perceived likelihood of acquiring HIV

CROI 2018 391