CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

uptake in infrequently and never tested populations was three times higher in rural workplaces (Table 1) and especially high in never tested employees aged <25 and in infrequently tested employees aged 25-34. The average cost of distributing HIVST in rural agricultural workplaces and urban industries was 4.30 USD vs. 4.35 USD, with 56% due to the cost of the kits (incl. freight), followed by distribution staff (32%), sensitisation (5%) and travel (3%). Conclusion: HIVST distribution at the workplace leads to improved HIV testing coverage especially in underserved rural populations, at a similar cost to urban populations.

963 CHARACTERISTICS OF MSMWHO REGISTER FOR HIV SELF-TESTING IN SAO PAULO, BRAZIL Ricardo D. Vasconcelos 1 , Vivian I. Avelino-Silva 1 , Ivone P. De Paula 2 , Leda Jamal 2 , Maria Clara Gianna 2 , Flavio Santos 3 , Cristina Santos 3 , Robinson Camargo 3 , Eduardo Barbosa 4 , Gilvane Casemiro 5 , Maria Cristina Abbate 3 , Marly M. Cruz 6 , Aluisio C. Segurado 1 , for the A Hora é Agora - SP 1 Universidad de São Paulo, São Paulo, Brazil, 2 Centro de Referência e Treinamento DST/AIDS-SP, Sao Paulo, Brazil, 3 Secretaria Municipal da Saude de Sao Paulo, Sao Paulo, Brazil, 4 Centro de Referencia da Diversidade, Sao Paulo, Brazil, 5 Ministry of Health, Brasilia, Brazil, 6 Escola Nacional de Saúde Pública, Brasilia, Brazil Background: HIV testing is a critical step of both HIV care and prevention. Since 2015 WHO recommends HIV self-testing (HIVST) as an additional screening strategy to improve testing coverage among key populations. Prior to implementation of HIVST in the public health system in Brazil, the demonstrative study “A Hora é Agora” evaluated the acceptance, interest in use and logistics of distribution of free HIVST kits among men who have sex with men (MSM) in Curitiba and Sao Paulo, two state capitals in Brazil. We here analyze the characteristics and prevention attitudes of participants registered to undertake HIVST in Sao Paulo Methods: Between April-December/2018 potential participants were invited through social media and gay venues to complete a web-based anonymous survey on prevention attitudes, HIV infection risk and risk perception. We explored demographic and vulnerability characteristics associated with reported lifetime HIV testing using univariate analyses. We also compared participants with and without prior testing for their preferred testing strategy Results: 6,477 respondents who provided valid answers were included. All were MSM, with median age of 28 years (IQR 23-34); 54% self-declared as white and 68% had at least 12 years of schooling. Sexual orientation was homosexual for 81%. Fifty percent of the participants reported at least 1 episode of unprotected anal intercourse in the past 6 months; 25% reported illicit drug use in the same period. Despite a high-risk profile, the perception of risk for HIV infection in the next year was high for only 4%. 78% reported being previously tested for HIV, with factors such as facility working hours (53%), exposure of personal issues to a provider (34%) and gender identity/sexual orientation- related stigma (21%) cited as barriers for testing. Older age, higher education, illicit drug use and gay orientation were associated with higher percentage of lifetime HIV testing (p<0.001). Most participants (67%) reported not knowing of the availability of HIVST before enrolling in the study. The preference for HIVST was higher among participants who had never been tested (71%) compared to those with previous HIV testing (61%; p<0.001) Conclusion: In this study including high risk MSM, HIVST was the preferred testing strategy among participants who had never been tested. This shows HIVST may be an important tool to improve HIV testing, particularly among hard-to-reach key populations

Poster Abstracts

962 COST-EFFECTIVENESS OF HIV SELF-TESTING AMONG LONG-DISTANCE TRUCK DRIVERS IN KENYA Deo Mujwara 1 , Elizabeth Kelvin 2 , Gavin George 3 , Eva Mwai 4 , April D. Kimmel 1 1 Virginia Commonwealth University, Richmond, VA, USA, 2 City University of New York, New York, NY, USA, 3 University of KwaZulu-Natal, Durban, South Africa, 4 North Star Alliance, Burnt Forest, Kenya Background: Awareness of HIV status is critical for achieving UNAIDS targets, particularly for sub-populations at high risk of acquiring and transmitting HIV. These sub-populations require targeted, resource-intensive strategies for HIV test uptake, a challenge when resources are limited. We conducted a trial-based cost-effectiveness analysis on offering the choice of HIV self-testing (CHIVST) in a high-risk population—long distance truck drivers—in Kenya. Methods: We leveraged data from a randomized controlled trial of CHIVST (intervention, n=150) vs provider-administered testing—standard of care [SOC] (control, n=155). CHIVST included choice of SOC or clinic- or home-based self-test. Economic cost data (HIV test kits, medical supplies, labor, capital and overhead costs, patient time), including upper and lower bounds, came from the literature and reflected a societal perspective. Generalized Poisson and linear gamma regression models estimated the effectiveness (relative risk) and incremental costs (2017 I$), respectively, with incremental effectiveness calculated as the reciprocal of the absolute risk difference and reported as the number needed to receive CHIVST for an additional HIV test uptake. We reported incremental cost-effectiveness ratios, with 95% confidence intervals (CIs) calculated using Fieller’s theorem. Deterministic sensitivity analysis identified key cost drivers; non-parametric bootstrapping generated cost-effectiveness acceptability curves to assess uncertainty in the ratio. We determined cost- effectiveness according to a willingness-to-pay threshold of 3x GDP per capita for Kenya (I$9774). Results: HIV test uptake was 23%more likely for CHIVST vs SOC, with six individuals needed to receive CHIVST for an additional HIV test uptake (6.25, 95% CI 5.00-8.33). The mean cost per patient was more than double for CHIVST (I$26.56) compared to the SOC (I$10.47). The incremental cost-effectiveness of CHIVST was I$97.21 [95% CI 65.74-120.98] per additional HIV test uptake compared to SOC. Self-test kits and patient time were the main cost drivers, with findings robust even in a worst-case scenario of all upper bound economic costs. The probability of CHIVST being cost-effective at a given willingness-to- pay threshold approached one at a threshold of I$140 (Figure). Conclusion: CHIVST is a highly efficient use of resources for improving HIV test uptake among high-risk populations. Policies supporting CHIVST in these populations may expedite achievement of country-specific UNAIDS targets.

CROI 2020 361