CROI 2018 Abstract eBook

Abstract eBook

Oral Abstracts

145 WHO REMAINS UNTESTED FOLLOWING NEAR-UNIVERSAL (>95%) POPULATION HIV TESTING? Dalsone Kwarisiima 1 , Jane Kabami 2 , Norton Sang 3 , Kevin Kadede 3 , Atukunda Mucunguzi 4 , Katherine Snyman 5 , Tamara D. Clark 5 , Elizabeth A. Bukusi 6 , Teri Liegler 5 , Edwin D. Charlebois 5 , Laura B. Balzer 7 , Maya L. Petersen 8 , Moses R. Kamya 9 , Diane V. Havlir 5 , Gabriel Chamie 5 1 Makerere University Joint AIDS Program, Kampala, Uganda, 2 Makerere Univ and Univ of California San Francisco Rsr Collab, Mbarara, Uganda, 3 Kenya Medical Research Institute, Kisumu, Kenya, 4 Makerere University–University of California San Francisco Research Collaboration, Kampala, Uganda, 5 University of California San Francisco, San Francisco, CA, USA, 6 Kenya Medical Research Institute, Nairobi, Kenya, 7 University of Massachusetts Amherst, Amherst, MA, USA, 8 University of California Berkeley, Berkeley, CA, USA, 9 Makerere University College of Health Sciences, Kampala, Uganda Background: As HIV testing uptake increases in sub-Saharan Africa, with some settings exceeding the UNAIDS 90% target, HIV+ persons who remain unaware of their status are likely to contribute disproportionately to HIV-associated morbidity and transmission. We sought to characterize adults that remain untested for HIV despite living in communities with near universal testing coverage. Methods: Over 2 years, the SEARCH trial (NCT01864683) achieved 95.6% adult HIV testing coverage among 77,788 stable (≥6 months in community), adult (≥15) residents in 16 intervention communities in Kenya and Uganda with annual multidisease health campaigns followed by home-based testing for non- attendees, as previously described. We also ascertained vital status of residents. We sought to characterize stable adult residents enumerated by baseline census that remained alive and residing in community, but did not test with SEARCH over 2 years. We compared characteristics of non-tester residents to adults who tested at least once. Multivariate logistic regression was used to evaluate factors associated with never testing in adjusted analyses, accounting for clustering by household. Results: After 2 years, 74,342 (95.6%) tested for HIV at least once and 3,446 (4.4%) never tested. Of tested adults, 8720 (12%) out-migrated and 1203 (1.6%) died. Of non-testers, 1515 (44%) out-migrated and 126 (3.7%) died. Overall, 66,224 adult residents were alive, in-community after 2 years, of whom 1805 (2.7%) never tested. Of non-testers alive, in community, 1201/1805 (67%) were men, 678 (38%) were <25 years, and 159 (10%) lived alone. Of 1646 non-testers who did not live alone, 312 (19%) lived with an HIV+ household member. In multivariate models stratified by sex, odds of never testing among men were significantly increased in middle age (25-44), single men, greater months out of community at baseline, those with higher educational attainment, and those with no job or high-risk informal sector jobs. Among women, we observed similar associations, except the likelihood of never testing for women was greatest in 15-24 year olds, and greater among women with no job or formal jobs (Table). Conclusion: Following a 2-year testing intervention in the SEARCH trial, 2.7% of adult residents who were alive and living in the community remained untested. Among these non-testers, risk of never testing was significantly greater in middle-aged men and adolescent women, calling for continued outreach to these important populations at risk of HIV infection.

146 EXPANDED HIV TESTING ELIGIBILITY INCREASES DETECTION OF HIV INFECTIONS, WESTERN KENYA Rachel Joseph 1 , Paul K. Musingila 1 , Fredrick Miruka 1 , Lucy Ng’ang’a 2 , Hellen Muttai 1 , Kevin M. De Cock 2 , Felix M. Mulama 3 , Caroline O. Dande 4 , Polycarp Musee 5 , Stella Wanjohi 6 1 US CDC Kisumu, Kisumu, Kenya, 2 US CDC Nairobi, Nairobi, Kenya, 3 ICAP at Columbia University–Kenya, Kisumu, Kenya, 4 KEMRI-UCSF, Kisumu, Kenya, 5 Elizabeth Glaser Pediatric AIDS Foundation, Washington, DC, USA, 6 Centre for Health Solutions, Siaya, Kenya Background: Homa Bay, Siaya, and Kisumu counties in western Kenya have the highest estimated HIV prevalence (20–26%) in the country, and struggle to meet targets for HIV testing services (HTS). The Kenya Ministry of Health (MOH) recommends annual HIV testing for the general population. To increase access to HTS, seven high-volume facilities in western Kenya expanded HIV testing eligibility in March, 2017 to include those reporting a negative HIV test in the past 3–12 months, or in the past <3 months if unverified by records. We quantified HIV infections detected among clients meeting MOH and expanded eligibility criteria for HIV testing. Methods: We conducted a retrospective analysis of routinely collected program data fromMarch to July, 2017. Data from clients >10 years of age who received HTS in outpatient (OPD) and inpatient settings at the seven health facilities were included. Outcomes were meeting specific MOH and expanded HIV test eligibility criteria, and HIV test result. STATA version 14.2 was used to explore the data, and to test for differences in outcomes by client characteristics using bivariate analysis. Results: During the 15-week period 88,641 clients received HTS, of whom 79% (70,065) were screened in OPD, and 59% (52,475) were women. A total of 70,493 (80%) were eligible for testing and 97% (68,513) received a test. Overall, 26% (18,456) of clients tested met MOH eligibility criteria: 7% (4,921) had never been tested, 15% (10,247) reported a negative HIV test in the past >12 months, and 5% (3,288) met other criteria. The remaining 74% (52,037) met expanded criteria: 52% (35,274) reported a negative test in the past 3–12 months, and 23% (15,913) had an unverified negative test in the past <3 months. In total, 1.1% (740) of clients had a positive HIV test. Although the yield of positive tests was 2.4-fold higher among those meeting MOH criteria (1.9% vs. 0.8%; p<0.001), more than half of all infections (406) were found among clients meeting expanded criteria, the majority (77%) reporting a negative test in the past 3–12 months. Conclusion: The majority of HIV infections detected at facilities with expanded testing occurred among clients reporting a negative HIV test in the past 12 months, clients ineligible for testing under the current MOH guidelines. Expanding MOH HTS eligibility in high HIV-burden areas to include clients tested in the past 3–12 months could increase access to HTS and timely diagnosis, and accelerate epidemic control. 147 OPTIMAL HIV TESTING STRATEGIES TO INCREASE HIV DIAGNOSIS IN SOUTH AFRICA Leigh F. Johnson 1 , Gesine Meyer-Rath 2 , Craig Van Rensburg 3 , Caroline Govathson 3 1 University of Cape Town, Cape Town, South Africa, 2 Boston University, Boston, MA, USA, 3 University of the Witwatersrand, Johannesburg, South Africa Background: The UNAIDS goal of a 90% rate of HIV diagnosis by 2020 has prompted much innovation in community-based testing and self-testing strategies. However, it remains unclear which strategies are likely to have the greatest population-level impact and which strategies are likely to be most cost-effective. Methods: A mathematical model was developed to simulate the impact and cost-effectiveness of different HIV testing strategies in South Africa. The model is calibrated to historic data on levels of HIV prevalence and testing in South Africa. Existing strategies include general testing, partner testing and testing in antenatal clinics, STI clinics, patients with OIs, prisons, men seeking MMC and sex workers using PrEP. Potential new strategies include home-based testing, mobile testing, testing targeted to sex workers and MSM, testing partners of pregnant women and testing in family planning clinics, schools and workplaces. Self-testing is modelled in the context of home-based and antenatal partner testing. Results: Over the 2019-39 period, the strategies with the highest yield (numbers on new diagnoses per test) are expected to include testing partners of newly-diagnosed individuals, sex workers, OI patients, PrEP users and MSM,

Oral Abstracts

53

CROI 2018