CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

247 (18%) had an incident STI. Being single, African American, active duty, carrying a condom, inconsistent condom use, and having a male or non-steady partner at last sexual encounter were significant on univariate analyses, while alcohol use and alcohol consumption were not. In multivariate analyses, STI risk increased with number of sex partners (p=0.001), history of sex with a man (OR 4.3 and 6.0 for PNA, p<0.01) and the odds of getting an STI increased as self-perceived risk increased (low, med/high, unsure) compared to no risk (ORs: 3.4, 4.4, 8.9, respectively, all p<0.001). Age and race remained significant. In a separate model, condom use, number of new sex partners and non-steady last sex partner were all independently associated with perceived risk. Conclusions: MSM behavior among those with HIV is higher than previously reported in a military setting. While we know that STIs continue to occur in our population (18%), not all risk behaviors were predictive of incident STIs. Further research to understand correlates beyond individual risk such as social networks are warranted. STI prevention efforts should focus on condom use and new partners as opposed to perceived risk. 933 Intentional HIV Seroadaptive Behavior Patterns Among Seattle MSM, 2008-2014 Sara N. Glick 1 ; Richard Burt 2 ; Christine Khosropour 1 ; HanneThiede 3 1 Univ of Washington, Seattle, WA, USA; 2 PH, Seattle & King County, Seattle, WA, USA; 3 PH, Seattle & King County (retired), Seattle, WA, USA Background: Seroadaptive behaviors are an HIV risk reduction strategy practiced by men who have sex with men (MSM) and include serosorting and modifying sexual behavior on the basis of viral load among HIV-infected MSM. Most previous estimates have measured serosorting behaviors without assessing whether they are a product of intentionality. Methods: We evaluated intentional seroadaptive behaviors among MSM using serial cross-sectional data from the 2008, 2011, and 2014 MSM cycles of the Seattle-area National HIV Behavioral Surveillance system (N=1,242). MSM were recruited using venue-based sampling. Local questions asked about intentional seroadaptive behaviors in the last year including condomless anal sex (CAS) due to knowledge of concordance with a partner’s HIV status (intentional serosorting), HIV-concordant CAS at last sexual encounter (serosorting proxy), and CAS due to undetectable viral load among HIV-infected MSM (intentional viral load sorting). We also asked about a decision to refrain from CAS due to knowledge of a partner’s HIV status. We stratified findings by self-reported HIV status and assessed differences in behaviors over time. We used multivariable logistic regression to identify factors differentiating MSM reporting seroadaptive behaviors from other MSM. Results: Self-reported HIV prevalence was 17%. Intentional serosorting in the last year was reported by 64% of HIV-infected MSM and 48% of HIV-uninfected MSM. Among HIV-infected MSM, STD diagnosis in the last year was associated with intentional serosorting (aOR=3.9, 95% CI: 1.3-12.0). Not having sex due to knowledge of a partner’s HIV status was reported by 21% of HIV-uninfected and 24% of HIV-infected MSM. There was no evidence of differences in intentional concordant CAS over time (Table). There was a significant increase in concordant CAS at last sex among HIV-uninfected MSM only. There was no significant change in intentional viral load sorting. Among HIV-infected MSM, 27% engaged in intentional viral load sorting in the last year, and this was associated with age (aOR=0.94 per year, 95% CI: 0.89-1.00), ≥10 male sex partners (aOR=4.2, 95% CI: 1.5-11.5), and methamphetamine use (aOR=0.3, 95% CI: 0.1-1.0). Conclusions: Intentional seroadaptive behaviors among Seattle MSM were common and stable from 2008-2014, and among younger MSM, viral load sorting is an emerging seroadaptive behavior. Assessing intentionality of serosorting resulted in a different temporal pattern than a measure based solely on reported behavior.

Poster Abstracts

934 Determinants of HIV Transmission Risk Among HIV-Infected Persons Engaged in Care Charitha Gowda 1 ; Dagan Coppock 1 ; Cristina E. Brickman 2 ; Pamela A. Shaw 1 ; Robert Gross 3

1 Univ of Pennsylvania, Philadelphia, PA, USA; 2 Univ of California San Francisco, San Francisco, CA, USA; 3 Univ of Pennsylvania Perelman Sch of Med, Philadelphia, PA, USA Background: HIV incidence in the US has not decreased over the past decade likely due to the inability of prevention efforts to reach those at risk of HIV transmission and acquisition. HIV-infected persons in medical care represent an accessible group in whom preventive efforts can have maximal success. We sought to identify factors associated with HIV transmission among HIV-infected persons engaged in care, including younger age, alcohol or illicit drug use, depression, low socioeconomic status, less education, and men who have sex with men (MSM). Methods: We used data from the University of Pennsylvania Center for AIDS Research Clinical Core Cohort to examine factors associated with HIV transmission risk. For each visit, HIV transmission risk was based on the presence of both HIV viremia >1500 copies/mL and reported unprotected sexual activity. We calculated the proportion at risk for HIV transmission and explored the incidence of changing HIV transmission risk within person over time. We used multivariable mixed effects logistic regression models to estimate the association between HIV transmission risk and our hypothesized risk factors, with random intercepts to account for the correlation between multiple study visits per person. Results: From 2/28/2007 to 5/28/2015, 1,883 HIV-infected persons contributed a median (IQR) of 3 study visits (2 – 5) for a total of 4,582 visits. Median time between consecutive visits was 11.6 months (7.3 – 17.8). 174 (9.2%) individuals were at risk for HIV transmission during at least once during the period, with 27 (1.4%) having multiple at risk intervals. In addition, 114 (6.0%) individuals changed HIV transmission risk state over time, of which 23 (20%) underwent multiple shifts. Factors associated with HIV transmission risk included younger age (OR [95% CI] per 10-year decrease=2.06 [1.65, 2.47]), illicit drug use (OR=2.40 [1.56, 3.68], depression (OR=1.99 [1.29, 3.06]), and education < 12 th grade (OR=1.87 [1.08, 3.25]). Alcohol use, sexual orientation and gender were not predictive of HIV transmission risk. Conclusions: Nearly 1 in 10 HIV-infected individuals engaged in care were at risk of transmitting HIV during the study period due to concomitant uncontrolled HIV infection and high risk sexual behavior. Behavioral interventions to decrease HIV transmission need to focus on patients in care who are younger, less educated, depressed, and actively using illicit drugs.

394

CROI 2016