CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

and intergenerational sex and for recent HIV as outcomes. As part of the analytic framework, we also assessed whether secondary or greater education was associated with HIV infection in young adults (aged 15-24). Results: Of the 112,964 enrolled adults aged 15–59 years, 23% lived in households reporting FI. FI was associated with migration (away for >1 month in past 1–3 years), and in older adolescents (aged 15–17 years), lower odds of current school enrolment (Figure). Higher educational attainment was associated with lower odds of prevalent HIV in men and women aged 15–24 years. FI was associated with intergenerational sex in women aged 15–24 years and, in all women, with transactional sex, and with a two-fold increase in recent HIV infection (adjusted odds ratio [aOR], 2.08; 95% confidence interval [CI]: 1.04–4.17). FI was not associated with lower odds of VLS, but migrants were less likely to be suppressed (aOR, 0.48; 95% CI: 0.35–0.67). Conclusion: FI could negatively impact the HIV epidemic both in the short- term, by increasing high-risk sexual behaviors and HIV infection rates in women, and in the long-term, by impeding educational attainment and increasing migration.

without. (P <0.0001 for all comparisons.) Ten or more DC Cohort participants lived in 20 Washington DC zip codes. Of the 270 PWH with incident STIs, 85.6% lived in 10 zip codes (See figure). Of the 270 participants with incident STI, at least one HIV VL was available for 254 (94.1%). Overall, 69 (27.2%) of individuals with incident STIs had an HIV VL ≥200 copies/ml. Of these 69, 72.5% resided in 6 of the 20 Washington DC zip codes. Conclusion: In Washington DC, 6 zip codes of residence accounted for 72.5% of the estimated HIV transmission burden among participants in the DC Cohort. Estimates of HIV transmission burden by zip code of residence allow for targeted, neighborhood-level interventions that may strengthen efforts to end the HIV epidemic.

1134 LATE PRESENTATION PERSISTS UNDER UTT IN SOUTH AFRICA: A NATIONAL COHORT STUDY Jacob Bor 1 , Matthew P. Fox 1 , Cornelius Nattey 2 , Brendan Maughan-Brown 3 , Mhairi Maskew 2 , Dorina Onoya 2 , Alana T. Brennan 1 , Till Bärnighausen 4 , H Manisha Yapa 5 , Sergio Carmona 6 , Wendy Stevens 6 , Adrian J. Puren 7 , William B. MacLeod 1 1 Boston University, Boston, MA, USA, 2 Health Economics and Epidemiology Research Office, Johannesburg, South Africa, 3 University of Cape Town, Cape Town, South Africa, 4 Heidelberg University, Heidelberg, Germany, 5 Kirby Institute, Sydney, NSW, Australia, 6 National Health Laboratory Service, Johannesburg, South Africa, 7 National Institute for Communicable Diseases, Johannesburg, South Africa Background: South Africa implemented Universal Test-and-Treat (UTT) in Sept 2016, extending eligibility for antiretroviral therapy (ART) to all people with HIV, regardless of CD4 count. However, the impact of UTT will be limited if people do not seek care early in infection. Methods: We constructed a national HIV cohort by linking the complete historical laboratory records of South Africa’s public sector HIV program, Apr 2004 – Mar 2018, enabling all patients to be tracked longitudinally. For each patient, we defined “CD4 at presentation” as first CD4 count in the database and “CD4 at ART start” as the first CD4 count within 90 days of ART workup (ALT/ HG/CRT) taken in advance of starting ART. (The same CD4 count would apply for each if a patient started ART at presentation.) We assessed the distribution of CD4 counts at presentation and at ART start in 2015 (pre-UTT) and 2017 (post-UTT) and assessed features of these distributions (e.g. median) over time, 2004-2018. We also estimated trends in numbers of patients entering care and starting ART (as proxied by a VL or ART workup) at different CD4 counts, and evaluated the impact of ART eligibility expansions (from<200 to <350, to <500, and to UTT) on these trends. Results: 12M patients had a first CD4 count through March 2018. In 2017, 48.2% of patients presented with CD4 >=350 cells/mm 3 , and just 29.3% presented with CD4 counts >=500 (newly-eligible with UTT) (Fig 1a). The shares were nearly identical in 2015 (pre-UTT): 48.3%>=350 and 28.8%>=500. Median CD4 at presentation increased from 229 cells/mm 3 in 2005 to 338 in 2015, but plateaued thereafter, reaching only 337 in 2017 (Fig 1b). Median CD4 at ART start increased from 173 cells/mm 3 in 2005 to 327 in 2015, with a marginal increase to 332 in 2017. With each ART eligibility expansion, the gap between median CD4 at presentation and ART start narrowed, disappearing under UTT (Fig 1b). Numbers seeking care did not increase with UTT. Although ART eligibility expansions led to short-run increases in numbers starting ART, each successive change in CD4 criteria affected a smaller share of patients, and as a result, the numbers starting ART have plateaued since 2011 (Fig 1c).

Poster Abstracts

1133 GEOGRAPHIC ESTIMATE OF SEXUAL HIV TRANSMISSION BURDEN IN ERA OF U=U: DC COHORT DATA Hana Akselrod 1 , Morgan Byrne 1 , Anne K. Monroe 1 , Matthew E. Levy 1 , Rachel Denyer 1 , Adam Klein 1 , Michael A.Horberg 2 , Amanda D. Castel 1 , Rupali K. Doshi 3 , Alessandra Secco 1 , Jose Lucar 4 , Leah Squires 5 , Stefanie Schroeter 5 , Debra A. Benator 5 , for the DC Cohort Executive Committee 1 George Washington University, Washington, DC, USA, 2 Kaiser Permanente Mid- Atlantic States, Rockville, MD, USA, 3 District of Columbia Department of Health, Washington, DC, USA, 4 University of Mississippi, Jackson, MS, USA, 5 Washington DC VA Medical Center, Washington, DC, USA Background: Washington, DC (DC) has the highest jurisdictional prevalence of HIV in the US. Sexual transmission is the primary driver of the HIV epidemic in DC, the US, and globally. The Undetectable = Untransmittable (U=U) campaign advances the goal of ending the HIV epidemic by promoting durable viral suppression and reducing sexual transmission. On the other hand, insights into geographic areas of high HIV transmission burden allow for focused and impactful interventions. We aimed to assess HIV transmission by zip code of residence in the DC Cohort, a city-wide cohort of persons with HIV infection (PWH). We define HIV transmission burden as the number of PWH with high- risk sexual behaviors as identified by an incident STI who also are at risk for transmitting HIV. Methods: We conducted an analysis of DC Cohort participants, ages ≥13 from April 1, 2016 to March 31, 2018. We assessed by zip code of residence, HIV transmission burden: the number of those with incident STIs (gonorrhea, chlamydia, and syphilis) and any HIV VL ≥200 copies/mL from the nine months prior to the day of STI diagnosis to 3 months post STI diagnosis (to approximate the U=U criteria for undetectable). Results: Of 3,467 participants, 270 (7.8%) had at least one incident STI. Compared to those without any STIs, those with ≥1 STI were younger (mean age 41.1 with vs. 54.1 years without STIs), male (91.5% vs. 64.5%) and MSM (79.6% vs. 31.9%). White race was more frequently represented among those with STIs (23.3%) compared to those without STIs (8.6%) and blacks were less frequently represented (66.7%with vs. 83.2% of those without STIs). Homelessness or temporary housing was more common among those with STIs, 18.9% vs. 9.1%

CROI 2020 428