CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Conclusions: FSW in Lilongwe have high HIV prevalence and proportion of newly diagnosed HIV infections, with a substantial proportion of HIV-infected FSW experiencing inadequate engagement in the HIV care continuum. Although the majority of FSWwho are on ART are virally suppressed, ART uptake and adherence are sub-optimal. To reduce ongoing transmission and improve health outcomes, increased FSW engagement in the HIV care continuum is urgently needed. Incorporation of universal testing and treatment strategies for all FSW in Malawi must be strongly considered. 852 New HIV Cases and ARV Retention in Pretoria: A Gender Project for High-Risk Women Wendee M. Wechsberg 1 ;William A. Zule 1 ; Irene A. Doherty 1 ;Tracy L. Kline 1 ; Jacqueline Ndirangu 1 ; Charles M. van der Horst 2 1 RTI International, Research Triangle Park, NC, US; 2 University of North Carolina, Chapel Hill, NC, US Background: HIV testing and treatment programs by the South African Government currently miss many drug using women and sex workers. Retention rates in antiretroviral therapy (ART) for women who test positive are poor. This presentation reports complete baseline data regarding the HIV treatment cascade fromwomen enrolled in a gender- focused NIDA-funded cluster randomized study in Pretoria, South Africa to increase HIV testing, treatment and retention in ART. Methods: We used natural boundaries (e.g. highways, rivers, etc.) to divide the city of Pretoria (635 sq. miles) into 14 zones. Outreach workers for the project recruited sex workers and other women who reported using alcohol or drugs at least weekly from all 14 zones. We enrolled 640 women who completed interviews, testing for HIV, pregnancy, and recent alcohol and drug use. Results: The most common biologically confirmed drugs of abuse were marijuana 32%, opiates 18%, and cocaine 15%, and 14% had a positive alcohol breathlyzer. HIV prevalence was 55% overall, and 68% among sex workers. Eleven percent of women reported this was their first HIV test. Of these, 52% of were newly diagnosed with HIV. Of the women who had been tested previously, 85 (15%) were newly diagnosed. Only 22% of HIV positive women were on ART. CD4 counts results were only available for 39%. Of women with a CD4 count, 36% had a count ≤ 350 which made them eligible for ART, but only 37% of themwere on ART. The major ART were structural (e.g. clinics far away, clinics ran out of ARVs, no CD4 tests, etc.) and individual barriers (e.g. food insecurity, belief in traditional medicine, addiction). Conclusions: More people live with HIV in South Africa than anywhere else in the world. Yet, there is still an unmet need for reaching, testing, treating and retaining high risk HIV positive women. South Africa must increase focused efforts for high risk groups to help them progress successfully through the HIV treatment cascade. This project is testing a woman-focused intervention to help HIV positive progress through the treatment cascade and achieve suppressed viral loads. 853 Intimate Partner Violence and Antiretroviral Adherence in HIV-PositiveWomen in Kenya Kate S. Wilson 1 ; KristaYuhas 1 ; Ruth Deya 1 ; Barbra Richardson 1 ; Linnet Masese 1 ; Jane Simoni 1 ;Walter Jaoko 2 ; R Scott McClelland 1 1 University of Washington, Seattle, WA, US; 2 University of Nairobi, Nairobi, Kenya Background: Intimate partner violence (IPV) is common in HIV-positive women and may be a risk factor for poor adherence. We examined whether recent IPV was associated with poor antiretroviral therapy (ART) adherence among HIV-positive women in Mombasa, Kenya, who reported trading sex for cash or in-kind payment. Methods: Women in our prospective cohort study were HIV-positive, ≥ 18 years, eligible for ART and reported transactional sex. Women completed monthly follow-up visits and were offered HIV care including ART according to Kenyan guidelines. Intimate partner violence in the past year was defined as experiencing ≥ 1 of 13 acts of physical, sexual, or emotional violence by the current or most recent regular partner (index partner). Exposure to IPV was updated yearly. Poor adherence was defined as >48 hours late for a scheduled monthly ART refill by clinic data. We have previously shown that late refill is associated with plasma and genital HIV detection and genotypic HIV resistance. Adherence in the past month using a validated self-rating scale (very good/excellent vs. less than very good) and visual analogue scale (VAS) <80%were secondary endpoints. We used generalized estimating equations with a log link, exchangeable correlation structure, and robust standard errors to estimate relative risks (RR) and 95% confidence intervals (CI). Models were adjusted for age and education. Results: Overall, 247 women contributed 315.5 person-years to the analysis. Median age was 40 years (interquartile range [IQR] 35-45). Of 3,835 total visits, 568 (14.8%) were contributed by women who reported IPV by their index partner in the past year. Of 3,172 visits with refill data, late refill occurred at 86/480 (17.9%) IPV exposed and 693/2,692 (25.7%) IPV unexposed visits (RR 0.89, 95% CI 0.74, 1.08). The adjusted association was similar (aRR 0.85, 95% CI 0.69, 1.04). Results were similar when visits where women refused or had stopped ART were excluded, and when analyses were restricted to visits with an index partner. There was also no evidence that IPV was associated with increased risk of poor adherence by self-rating scale (data not shown). Conclusions: Intimate partner violence was common in this cohort of high-risk HIV-positive Kenyan women. However, we found no evidence that IPV was associated with poor ART adherence. While IPV should be addressed as an important women’s health problem, we may not expect reductions in IPV to be associated with improved ART adherence. 854 Re-engagement in Care Leads to Sustained Engagement and Viral Suppression Brittani D. Saafir-Callaway 1 ; Amanda D. Castel 2 ; Lena Lago 1 ; Christie Olejemeh 1 ; Garret Lum 1 ; Lawrence Frison 1 ; Michael Kharfen 1 1 District of Columbia Department of Health, Washington, DC, US; 2 The Milken Institute School of Public Health at George Washington University, Washington, DC, US Background: The HIV Care Continuum requires HIV+ people to be fully engaged in HIV care and treatment in order to achieve viral suppression. However, many people cycle in and out of care making it difficult for them to achieve this goal. We sought to use HIV surveillance and clinical data to determine whether identification and subsequent re- engagement of HIV+ out of care people results in improved retention in care and clinical outcomes. Methods: Through the District of Columbia Department of Health (DC DOH) Recapture Blitz, data on 1,015 HIV+ persons were received from 7 clinical providers who identified persons as being out of care for a 12-month period. Persons were matched to the DC DOH surveillance and services databases, and those without evidence of receipt of care based on surveillance data were sent back to clinics for investigation. Median CD4 and viral load, retention in care, and viral suppression at last lab after re-engagement were assessed 12 months following the Recapture Blitz. Results: From October 2012 to April 2013, 691 HIV+ persons had no evidence of being in care and were subsequently investigated. Viral load or CD4 results were available for 390 (56%) of those investigated. At their last known lab, the median viral load and CD4 among out of care persons was 200 copies/ml and 27 cells/ m l, respectively. Of the 691 persons investigated, 573 (83%) were contacted for re-engagement: 121 (21%) were in care elsewhere, 61 (11%) had moved to another jurisdiction, and 59 (10%) were re-engaged in care. As of April 2014, the majority of re-engaged persons remained in care (n=44, 75%); 32 (54%) were retained in continuous care and 12 (20%) were sporadically engaged in care. Fifteen people (26%) had no evidence of being in care as per surveillance records. Among persons retained in care at 12 months of follow-up, 25 (57%) were virally suppressed at their last reported viral load. The median CD4 results among persons retained in care increased significantly from 26 cells/ m l at last known lab to 458 cells/ m l after 12 months of follow-up (p<.0001) while the median viral load decreased from 1,800 to 85 copies/ml (p=0.1665). Conclusions: Through the Recapture Blitz, using a combination of surveillance and clinical data allowed not only for successful re-engagement in care but resulted in improved longer term health outcomes post-re-engagement. This analysis underscores the importance of re-engaging persons who have fallen out of care to improve overall rates of retention and viral suppression.

Poster Abstracts

515

CROI 2015