CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

839 THE MACARTI STUDY: CLOSING THE GAPS IN HIV CARE AMONG YOUTH IN ATLANTA, GA Andres Camacho-Gonzalez 1 , Scott Gillespie 1 , Lateshia Thomas-Seaton 2 , Krystal Frieson 3 , Sophia A. Hussen 1 , Ashley Murray 4 , Zaneta Gaul 6 , Madeline Sutton 4 , Chanda Graves 1 , Rana Chakraborty 1 1 Emory Univ, Atlanta, GA, USA, 2 Children’s Hlthcare of Atlanta, Atlanta, GA, USA, 3 Atlanta VA Med Cntr, Decatur, GA, USA, 4 Emory Univ, Atlanta, GA, USA, 5 CDC, Atlanta, GA, USA, 6 ICF Intl, Atlanta, GA USA Background: Georgia ranks 5th among US states for new HIV diagnoses; black youth and young adults are disproportionately affected. Ensuring early diagnosis and HIV care are vital steps to decrease the HIV burden in Georgia, consistent with the National HIV/AIDS Strategy. Methods: Newly identified HIV+ young people aged 18-24 years were enrolled into the Metropolitan Atlanta Community Adolescent Rapid Testing Initiative (MACARTI) intervention and compared to standard of care (SOC) participants. MACARTI combined a formative phase that informed a later strategy of non-traditional venue HIV testing, pre-test and post-test motivational interview sessions, and case management support. Demographic, behavioral, and clinical variables along with linkage (within 3 months of diagnosis) and retention rates (missed visit rate/100 visits) were collected. Means and standard deviations (SD) were calculated for continuous variables; frequencies and percentages with 95% confidence intervals (CI) are shown at baseline and at 12-month follow-up time. Clinical values were compared between arms using parametric and non- parametric statistical tests. Results: Ninety-eight participants were enrolled; 49 each in the MACARTI and SOC arms; 85%were male; 91%were Black; mean age=21 years (SD:1.8). The MACARTI study screened 435 participants for an HIV-positivity rate of 11.3%. Overall 64% of participants were linked to care; linkage was higher for the MACARTI arm compared with SOC (88% vs. 41%, p≤0.001). Mean linkage time for MACARTI participants compared to SOC was 0.46 (IQR: 0.23-0.85) vs. 5.31 (IQR:1.35-17.03) months (p<0.001). Missed visit rates in the MACARTI armwere significantly lower than the SOC arm [14.4 (95% CI:10.3 – 19.8) vs. 26.1 (CI:20.8- 32.2) attended visits per 100 visits scheduled, respectively]. Mean CD4+ T-cell counts increased within both arms, however, values were significantly higher in the MACARTI arm at 12-month follow-up (474 vs. 278 cells/mm3; p=0.006). Also, mean HIV-1 RNA levels decreased in both arms at 12-month follow-up, with borderline significance across the MACARTI arms and SOC arm [79 vs. 283 copies/ml p=0.068 respectively]. Conclusion: The MACARTI intervention successfully identified and linked HIV-positive black youth to care in Atlanta. The intervention also decreased missed visit rates and improved CD4 counts and viral suppression rates. MACARTI may serve as an HIV linkage and care model for other areas with HIV-affected black youth. 840 RETENTION, ART USE, AND VIRAL SUPPRESSION AMONG YOUNG ADULTS NEWLY LINKED TO HIV CARE Fidel A. Desir 1 , Peter F. Rebeiro 2 , Richard D. Moore 1 , Heidi M. Crane 3 , Karyn Gabler 4 , Joseph B. Margolick 1 , Ronald Bosch 5 , Michael A. Horberg 6 , Michael J. Silverberg 7 , Keri N. Althoff 1 1 The Johns Hopkins Univ, Baltimore, MD, USA, 2 Vanderbilt Univ, Nashville, TN, USA, 3 Univ of Washington, Seattle, WA, USA, 4 BC Cntr for Excellence in HIV/AIDS, Vancouver, British Columbia, Canada, 5 Harvard Univ, Boston, MA, USA, 6 Kaiser Permanente Mid-Atlantic States, Rockville, MD, USA, 7 Kaiser Permanente Northern California, Oakland, CA, USA Background: Retention in care (RIC), antiretroviral therapy use (ART), and HIV viral suppression (VS) are critical milestones for persons living with HIV (PLWH) to achieve soon after an HIV diagnosis. The United States (US) National HIV/AIDS Strategy specifies subgroups of particular importance, including young black men who have sex with men (MSM), black women, and Hispanics. Our objective was to examine trends in these indicators among important subgroups of young adults newly linked to care in the US. Methods: Young adults (18-<30 years) with ≥1 CD4 or HIV RNA measure within 5 years of linking to care in 15 US clinical cohorts in the North American AIDS Cohort Collaboration on Research and Design were included. We evaluated 3 Department of Health and Human Services indicators: 1) RIC was the percentage of patients with ≥1 HIV care visit in January to June who also had visits in each following semester of a 24-month period, ≥60 days apart; 2) ART was the percentage with ≥1 HIV care visit who were prescribed ART for ≥1 month; and 3) VS was the percentage with ≥1 HIV care visit and ≥1 HIV RNA measure who had an HIV RNA ≤200 copies/mL at their last measurement in the calendar year. Cross-sectional annual estimates from 2004-14 were produced for each subgroup. Log binomial models with generalized estimating questions for repeated measures and an ordinal variable for calendar time were used to estimate p-values for trend. Changes over time in disparities between subgroups were evaluated using an interaction term and a nested models approach. Results: Among 9,432 young adults newly linked to care, ART and VS increased in all subgroups over time (all p-trends<.01; Figure 1). However, there was no change in RIC (all p-trends>.05). Black MSM had lower RIC and VS compared to white and Hispanic men, with a widening of the disparity in VS over time relative to white men (p-interaction<.001). Hispanic men had higher percentages of ART compared to black MSM and white men, with a narrowing disparity over time relative to white men (p-interaction<.001). Compared to Hispanic women, black and white women had lower percentages of RIC and ART. The disparity in ART relative to white women decreased for Hispanic and black women over time (p-interaction=.035). Conclusion: The low proportions of RIC, ART, and VS in young adults newly linked to care, and changes over time in disparities between key subgroups, demonstrate the need for continued efforts to improve the percentages of young PLWH that achieve these milestones. 841LB WITHDRAWN 842 RACIAL DISPARITIES IN HIV PREVALENCE AND COMPOSITION OF RISK NETWORKS, HPTN 037 Florence Momplaisir 1 , Mustafa Hussein 2 , Danielle Fiore 3 , Laramie Smith 4 , David Bennett 5 , Carl Latkin 6 , David Metzger 3 1 Drexel Coll of Med, Philadelphia, PA, USA, 2 Univ of Wisconsin-Milwaukee, Milwaukee, WI, USA, 3 Univ of Pennsylvania, Philadelphia, PA, USA, 4 Univ of California San Diego, La Jolla, CA, USA, 5 Drexel Univ, Philadelphia, PA, USA, 6 The Johns Hopkins Univ, Baltimore, MD, USA Background: HIV prevention interventions in the US have failed to reduce racial disparities in HIV prevalence. The purpose of this analysis is to evaluate individual and network factors associated with racial disparity in HIV prevalence among people who inject drugs using HIV Prevention Trial Network 037 data in Philadelphia. Methods: We measured racial consistency of risk networks (all members share the same race); network HIV prevalence by race of index participant; and network risk behaviors for drugs and sex among 232 index participants who regularly injected drugs and 464 network members. We then performed a logistic regression with a two level random intercept to evaluate the association between HIV status, individual and network characteristics. Results: Racial consistency was high among blacks and whites (79% and 70% respectively) while the majority of Hispanics were in racially mixed networks (racial consistency for Hispanics was 31%). HIV prevalence was 25% among networks of black index participants compared to 15% and 7% among networks of white and Hispanics index participants. Drug network risk behavior was significantly lower and sex risk behaviors similar in black compared to white and racially mixed networks: needle sharing was 23% in black, 48% in white, 46% in racially mixed networks (p<0.001). The number of unprotected sex events in the past week and number of sexual partners in the past month averaged around 2 in all networks. In our multivariable logistic regression, women (AOR 2.0, 95% CI 1.0-4.1), blacks (AOR 3.2, 95% CI 1.4-7.1), homeless individuals (AOR 2.2, 95% CI 1.1-4.5) and cocaine injectors (AOR 2.4, 95% CI 1.1-5.2) were more likely to be HIV positive compared to men, whites, housed individuals, and injectors who did not use cocaine (Table 1). Being in a network where members have multiple sex partners was negatively associated with HIV status (AOR 0.8, 95% CI 0.7-1.0). Conclusion: Despite having lower drug risk behavior, blacks were disproportionately HIV positive. HIV prevention interventions need to go beyond individual risks and consider social context and the composition of risk networks. 843 CONCURRENCY IN THE SEXUAL NETWORKS OF RACIALLY DIVERSE YMSM Patrick Janulis , Gregory Phillips, Brian Mustanski, MIchelle Birkett

Poster and Themed Discussion Abstracts

CROI 2017 364