CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

1003 Care-Cascade Status of Partners of Persons With New HIV Infections in North Carolina Anna B. Cope 1 ; Lisa Hightow-Weidman 1 ; JoAnn D. Kuruc 1 ; Jenni Marmorino 1 ; Steve Beagle 1 ; Philip J. Peters 2 ; Cynthia L. Gay 1 1 University of North Carolina, Chapel Hill, NC, US; 2 US Centers for Disease Control and Prevention (CDC), Atlanta, GA, US Background: HIV transmission risk is affected by HIV status awareness and care and treatment status. We aimed to describe the diagnosis, care, and viral suppression status of HIV-infected partners identified by persons newly diagnosed with HIV in North Carolina (NC). Methods: The STOP study is a multi-site, prospective study assessing methods to detect acute HIV infection (AHI). In NC, participants (age >12 years) at 3 sexually transmitted infection clinics were screened for HIV infection from 9/2011 to 10/2013. For newly diagnosed persons (indexes), partner services interviews elicited information about past sex partners within a specified period (AHI indexes=3 months; chronic HIV infection [CHI] indexes=12 months). The HIV status (HIV-infected, HIV-uninfected, status-unknown) and diagnosis (new or previously diagnosed), care (CD4 or viral load [VL] reported in NC surveillance databases), and viral suppression (VL<50 copies/ml) status were determined for reported partners. Results: Overall, 146 persons were newly diagnosed with HIV infection during the STOP study (20 AHI indexes; 126 CHI indexes). Index persons were predominately MSM (66%), young (median age 26 years), and black (86%). Index persons reported 791 sexual partners (80 by AHI indexes, 711 by CHI indexes). Over half of all partners were of unknown status (460 anonymous, 21 counselling-and-testing refusals, 45 testing-only refusals, 32 unlocatable). Of the remaining 331 partners, 129 (39%) were HIV-infected (24 newly diagnosed; 105 previously diagnosed). A total of 66 (63%) previously diagnosed partners had a reported VL/CD4 before the index diagnosis date; the last VL/CD4 for 30 (45%) of these partners was >6 months before the index diagnosis date, suggesting loss to care. Of those with a VL 6 months before the index diagnosis date (N=28), 19 were not virally suppressed (68%). AHI and CHI indexes reported a similar proportion of previously diagnosed partners; AHI indexes reported a higher proportion of virally suppressed, previously diagnosed partners (33% versus 4%; p=0.004). Half (N=80) of all indexes named ≥ 1 HIV-infected partner; 70 (48%) named ≥ 1 previously diagnosed partner.

Poster Abstracts

Conclusions: Previously diagnosed partners, many of whomwere not in care and virally suppressed, were prominent in networks of newly diagnosed persons. Prioritizing interventions to find previously diagnosed persons not in care and facilitate re-engagement and treatment could greatly impact HIV transmission. 1004 The Role of HIV Status Disclosure in Retention in Care and Viral-Load Suppression Latesha E. Elopre ; AndrewWestfall; Michael J. Mugavero; Anne Zinski; Greer Burkholder; Edward Hook; NicholasVanWagoner University of Alabama at Birmingham, Birmingham, AL, US Background: In the United States, retention in care is poor for patients infected with HIV and only 25% achieve viral load suppression. Knowledge that life expectancy and prevention of transmission to others is associated with adherence to anti-retroviral therapy suggests that better understanding of barriers to retention in care and effective viral load suppression is a priority. For newly diagnosed persons, the decision of whether and whom to disclose to is complex, with benefits weighed against perceived risks. The purpose of this study was to evaluate whether nondisclosure and selective disclosure of HIV status is associated with poor retention in HIV care and failure to achieve viral load suppression. Methods: This retrospective analysis evaluated the relationship of disclosure to poor retention in care (a gap in care > 180 days) and sustained viremia (viral load > 200 copies/ml) measured at 12 months after initiating HIV care. Participants must never have previously received HIV care and be older than 19 years of age. Primary analyses included disclosure status treated dichotomously (no disclosure vs any disclosure). Secondary analyses then evaluated nondisclosure and selective disclosure (disclosure to family only, friends only, significant other only) compared to disclosure to 2 or more groups (referent). Univariate and multivariable (MV) logistic regression models were fit including factors known to be associated with disclosure and the study outcomes. Results: From 2007-2013, 508 HIV infected patients presented to establish care, of whom 61%were African American, 53% had a CD4 + T lymphocyte count < 350 and 82%were men (60%men who have sex with other men). Of these, 65 (13%) reported nondisclosure and 258 (51%) reported selective disclosure. In primary MV analyses, nondisclosure was associated with poor retention in care (AOR 2.3; 95% CI 1.3, 4.1), but no relationship with sustained viremia was observed. In secondary MV analyses, the relationship between nondisclosure and poor retention in care was maintained (AOR 2.2; 95% CI 1.2, 4.3). Also, patients acknowledging selective disclosure to friends only (AOR 2.6; 95% CI 1.0, 6.5) or family only (AOR 2.9; 95% CI 1.2, 7.6) were more likely to have continuing viremia. Conclusions: Evaluating disclosure patterns among patients establishing HIV care may help predict inconsistent care and lack of viral load suppression. Further work is needed to evaluate why this relationship exists and to guide future interventions to improve these HIV-outcomes.

591

CROI 2015