CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

and wait at facilities. Multi variable logistic regression models were used to determine factors associated with accessing outpatient healthcare. Results: 191 of 3,524(5.4%) and 482 of 3,900(12.4%) respondents in South Africa and Zambia, respectively, accessed general outpatient care within the past 3 months. The total monetary expenditures per healthcare visit were USD 6.57 for South Africa and USD 11.59 for Zambia. The costs of accessing care were higher for Zambia compared to South Africa (see Table 1). The majority of expenditure for Zambia (65%) were costs incurred at the facility while for South Africa (54%) the majority of the expenditure were travel costs. Regression results for Zambia show that being HIV positive and on ART significantly increased the odds of accessing primary healthcare while for South Africa having been diagnosed with TB in the past 12 months was associated with significantly increased odds of accessing healthcare (Table 1). Wealth, use of recreational drugs, and employment status were found not to affect the odds of accessing health care in both countries. Conclusion: Patients incurred costs when accessing outpatient healthcare, despite the abolition of user fees.The odds of accessing healthcare were higher for people living with HIV & diagnosed with TB, suggesting financial burden for people with chronic diseases.

ZAMPHIA survey data were used to estimate national annual costs for scale-up and number of ART initiates. Results: 5,353 youth accepted the offer of assisted HIVST. The yield of newly diagnosed positive per person tested was 1.0% (56/5,353) for community- based HIVST and 3.2% (214/6728) for facility-based SOC. Just over half of those who newly tested positive through HIVST initiated ART (33/56) within three months. The average cost per client tested was $7.96 for HIVST and $3.18 for facility-based SOC. The total testing cost per new positive diagnosis was $580 and $80 in the HIVST and SOC arms respectively. The cost per new ART initiate increases to $978 for HIVST due to low facility linkage. An estimated 1,114,000 youth tested through currently available testing modalities in 2018, leading to 31,663 ART initiations for an annual cost of $3.6m. National HIVST rollout would reach an additional 310,000 youth annually, increasing the proportion of youth diagnosed by 6%, at an additional cost of $2.5m. Of these, a maximum of 2,192 additional youth would initiate ART. Conclusion: Community-based HIVST identifies youth who may not otherwise have tested for HIV, but is unlikely to be economically feasible at a national level. Other methods for improving youth HIV testing uptake, such as unassisted HIVST, index HIVST, or targeted community-based strategies should be evaluated and compared.

Poster Abstracts

1081 COST-EFFECTIVENESS AND NATIONAL IMPACT OF INDEX HIV SELF- TESTING IN MALAWI Ogechukwu Offorjebe 1 , Kathryn Dovel 1 , Frackson Shaba 2 , Kelvin Balakasi 2 , Risa M. Hoffman 1 , Sydney Rosen 3 , Brooke E. Nichols 3 , for the EQUIP Health team 1 University of California Los Angeles, Los Angeles, CA, USA, 2 Partners in Hope, Lilongwe, Malawi, 3 Boston University, Boston, MA, USA Background: Testing sexual partners of HIV-positive individuals (index testing) remains a high-yield testing strategy. The secondary distribution of HIV self-testing (HIVST) kits for index testing is highly acceptable in Malawi and promises to increase testing coverage. To assess the cost-effectiveness (CE) and feasibility of index HIVST, we modeled the cost per index partner tested positive and cost per newly confirmed positive (defined as a positive test at the health facility) for HIVST and for the current standard of care, partner referral slips (PRS), as well as the cost and impact of HIVST national scale-up. Methods: A decision analytic model was parameterized using data collected as part of a randomized trial comparing uptake of HIVST to PRS among partners of antiretroviral therapy (ART) clients at 3 district hospitals in Malawi. Clients were randomized 1:2 to standard PRS or HIVST (Oraquick HIV Self-Test: demonstration and distribution). Baseline and follow-up surveys with ART clients were conducted. CE was measured as the cost per newly confirmed positive (index partner) and was calculated for HIVST (including cost of HIVST kit ($2), counselling, confirmatory testing) and PRS (cost of referral slip, counselling, and standard facility-based testing) divided by the total number of positives newly aware of their status and facility-confirmed positives. Model outputs were applied to national facility-level data on number of HIV tests from PRS to determine potential national increase in new diagnoses and related costs for index HIVST scale-up. Results: The cost per index patient was $0.85 per PRS and $2.34 per HIVST provided and $3.08 and $3.17 per test completed, respectively. The cost per person newly aware of positive status was $19.27 for PRS and $16.14 for HIVST respectively. The cost per facility-confirmed positive was $84.53 for index HIVST due to low facility linkage. For national scale-up, 146,785 new positives were

1080 COST AND IMPACT OF COMMUNITY-BASED, ASSISTED HIV SELF-TESTING AMONGST YOUTH IN ZAMBIA Brooke E. Nichols 1 , Refiloe Cele 2 , Charles Chasela 3 , Zumbe Siwale 4 , Alimwi S. Lungu 4 , Lawrence Long 1 , Crispin Moyo 4 , Sydney Rosen 1 , Roma Chilengi 5 , for the EQUIP Health team 1 Boston University, Boston, MA, USA, 2 Health Economics and Epidemiology Research Office, Johannesburg, South Africa, 3 Right to Care, Johannesburg, South Africa, 4 EQUIP Health Zambia, Lusaka, Zambia, 5 Centre for Infectious Disease Research in Zambia, Lusaka, Zambia Background: Uptake of traditional facility-based HIV testing services among adolescents and youth is poor in many countries. HIV self-testing (HIVST) offers one strategy for increasing youth uptake. In order to assess scale-up feasibility, we conducted an economic evaluation of a pilot study that provided assisted, community-based HIVST for 16-24 year olds in Zambia. Methods: 30 clusters were randomly allocated 1:1 to either intervention or facility-based standard of care (SOC) in Ndola and Kabwe districts. In the intervention clusters, community-based HIVST was implemented through the use of roving teams in 15 clusters over a 6-mo period. These teams conducted community sensitization, counseling and assisted self-testing to 16-24 year olds, and facilitated linkage to care for those testing positive through escort to the facility or referral slips. We estimated the cost (staff salaries, community sensitization, equipment, materials and HIVST kits) per new ART initiate if implemented under routine care and compared this with the cost/initiate for SOC. National census population data, provincial prevalence rates, and 2016

CROI 2019 425