CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

896 YIELD OF HIV TESTING AND RE-ENGAGEMENT OF KEY POPULATIONS IN UGANDA AND KENYA Gabriel Chamie 1 , Norton Sang 2 , Dalsone Kwarisiima 3 , Jane Kabami 3 , Kevin Kadede 2 , Irene Bagala 3 , Atukunda Mucunguzi 3 , Tamara D. Clark 1 , Laura B. Balzer 4 , Edwin D. Charlebois 1 , Craig R. Cohen 1 , Elizabeth A. Bukusi 5 , Maya L. Petersen 6 , Diane V. Havlir 1 , Moses R. Kamya 7 1 University of California San Francisco, San Francisco, CA, USA, 2 KEMRI-UCSF, Kisumu, Kenya, 3 Infectious Diseases Research Collaboration, Kampala, Uganda, 4 University of Massachusetts Amherst, Amherst, MA, USA, 5 Kenya Medical Research Institute, Nairobi, Kenya, 6 University of California Berkeley, Berkeley, CA, USA, 7 Makerere University College of Health Sciences, Kampala, Uganda Background: The yield of key population outreach for HIV diagnosis and care re-engagement in settings that have surpassed the UNAIDS target of 90% of HIV+ persons aware of their status in sub-Saharan Africa is unknown. Methods: The SEARCH trial (NCT01864683) achieved >95% adult resident HIV testing coverage by 3 years with a hybrid approach (study census, multi- disease campaigns and home testing of campaign non-attendees) in 32 rural communities in Kenya and Uganda. After 3 years, 16 communities implemented 2-week mobile outreach, that included multi-disease and HIV testing services, for resident key populations considered high-risk for HIV (Table), and in-migrants (newly living in community since year 3 of trial). Known HIV+ persons were not retested, but referred to clinic if out of care by self-report or clinic database. We assessed HIV testing coverage and compared yield of new diagnoses including seroconversions (documented prior HIV-) and yield of known HIV+ out-of-care residents, across key populations. Results: HIV testing coverage of HIV-/unknown resident key populations was 16% (2,091/13,283) in Kenya, 14% (903/6,424) in West (W) Uganda, and 14% (1,830/13,555) in East (E) Uganda after the 2-week outreach. Yield of new HIV diagnoses among residents varied from 0% to 3.1% across key populations, and was highest among barmaids in Uganda (Table). Of 37 residents with newly-identified HIV, 29 (78%) were seroconversions. In-migrant testing yield was 19% (21/114) in Kenya, 17% (32/194) in W-Uganda, and 5% (13/287) in E-Uganda. Of 66 newly-identified HIV+ in-migrants, 40 (61%) reported prior HIV-/unknown status. The number needed to test to identify one newly- diagnosed HIV+ adult was 123 in Kenya, 69 in W-Uganda and 151 in E-Uganda among residents, compared to 12, 11 and 26 among in-migrants per region, respectively. Of HIV+ adult residents seen at mobile outreach, 28% (193/682) were out of care in Kenya, 21% (14/66) in W-Uganda, and 19% (10/53) in E-Uganda. Of all known HIV+, out-of-care residents within key population groups, 7% in Kenya and 3% in Uganda attended 2-week mobile outreach. Conclusion: Mobile, multi-disease outreach to key populations in SEARCH communities in rural Uganda and Kenya where HIV testing coverage was already high continued to yield new HIV+ diagnoses among residents, most of whom were seroconversions, and in-migrants. Outreach facilitated re-engagement of known HIV+ persons who leave care but remain willing to access mobile services.

895 RACIAL DISPARITIES IN BASELINE GENOTYPING IN THE ERA OF “ART FOR ALL” Sasinya Scott , Lisa A. Forgione, Lucia V. Torian New York City Department of Health and Mental Hygiene, Long Island City, NY, USA Background: Since 2007, federal guidelines for the care and treatment of persons with HIV have recommended genotyping at the initial care visit, both to establish a baseline and to guide antiretroviral therapy (ART). Previous studies indicated that patients were more likely to receive a baseline genotype if their CD4 at diagnosis reached the ART threshold in use at the time. Methods: We used laboratory data routinely reported to HIV surveillance to measure compliance among New York City physicians in two time periods – the “treatment threshold” era (2006-2012) and the “ART for all” era (2013-2017). We examined differences in baseline genotyping by provider type, patient demographics, risk factor, and clinical characteristics. A baseline genotype was defined as a genotype performed within 3 months of initial HIV diagnosis. Results: Baseline genotyping increased from 53% during the “treatment threshold” era to 63% during the “ART for all” era. The most important predictor of baseline genotyping between 2006 (39%) and 2012 (63%) was the CD4 count-genotyping was highest for people meeting the prevailing treatment threshold. In 2013, the year after guidelines recommended ART regardless of CD4, genotyping rose to 73%. After 2013, there was a steady decrease in percent genotyped overall, but relative increases in patients with CD4>500 and those receiving care at community-based organizations and free-standing clinics. Baseline genotyping in 2017, the last year for which data are complete, was 66%. Patients were more likely to receive a genotype if they were MSM, between the ages of 20-39 at diagnosis, white or Asian, acutely infected, CD4 <350, attended private providers or providers affiliated with hospitals and were diagnosed after 2012. Black race was independently associated with a 47.3% (95% CI 0.40, 0.69) lower likelihood of receiving a baseline genotype in the “treatment threshold” era, regardless of age, risk factor, neighborhood poverty level, clinical status, provider type, and year of diagnosis, and a non-significant 20.1% (95% CI 0.50, 1.26) lower likelihood of being genotyped in the era of “ART for all”. Our analysis was not able to account for the temporal changes in cost, reimbursement, turnaround time, guidance on interpretation, or other issues that may have affected provider decision to test. Conclusion: Five years into the era of “ART for all,” substantial inequity in baseline genotyping remains. Strategies to increase testing of black people are needed to improve quality of care.

Poster Abstracts

CROI 2019 349