CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

respectively). Log-transformed TFV-DP concentrations were analyzed using a mixed-effects model. Baseline statistics are presented as median (interquartile range). Model results are percent change [95% confidence interval] in TFV-DP for every significant change in the SDoH. All results are reported with no adjustment for multiple comparisons. Results: A total of 950 person-visits from 430 participants were analyzed, encompassing zip codes within the following Colorado counties: Denver, Arapahoe, Jefferson, Adams and Douglas. Baseline household income, Gini and TFV-DP concentration were $56,227 ($46,763, $70,369), 0.418 (0.391, 0.487) and 1721 (1181, 2441) fmol/punch, respectively. After adjusting for age, sex, race, estimated glomerular filtration rate, body mass index, hematocrit, CD4+ T-cell count, antiretroviral drug class and 3-month self-reported adherence, Gini was inversely associated with TFV-DP in DBS. For every 0.1 increase in Gini, TFV-DP concentration decreased by 9.2% [0.5, 17.1%; P=0.039]. Gini remained significant after adjusting for HIV viral suppression with the same 0.1 increase estimating a decrease of 8.7% [0.3 17.9%; P=0.042] in TFV-DP concentrations. No statistically significant associations were identified between TFV-DP concentration and the other SDoH (Table). Conclusion: Greater income inequality was associated with lower TFV-DP concentrations in PLWH on TDF, suggesting that adherence may be influenced by population level characteristics that exist in the presence of income inequality and impact individual level health outcomes. Future studies on the utility of this adherence biomarker to improve clinical care and adherence in marginalized PLWH are needed.

intake processes that facilitate rapid modification and initiation of ARVs should be routine in order move toward more rapid viral suppression among PLWH.

1087 USE CASE FOR NEAR POINT OF CARE HIV VIRAL LOAD: TARGETED TESTING AT LARGE FACILITIES Prakash Ganesh 1 , Tom Heller 1 , Boniface Chione 1 , Joe Gumulira 1 , Salem Gugsa 2 , Shaukat Khan 3 , Seth McGovern 3 , Angellina Nhlema 1 , Lyse Nkoma 1 , Jilian Sacks 3 , Clement Trapence 1 , Hannock Tweya 1 , Peter Ehrenkranz 4 , Sam J. Phiri 1 1 Lighthouse Trust, Lilongwe, Malawi, 2 University of Washington, Seattle, WA, USA, 3 Clinton Health Access Initiative, Boston, MA, USA, 4 Bill and Melinda Gates Foundation, Seattle, WA, USA Background: Point-of-care (POC) technologies in resource-limited settings can circumvent challenges of centralized laboratory testing, leading to an increased proportion of results used for clinical management. However, higher device costs and uncertain use cases for POC in routine care have inhibited scale-up. To address this gap, we investigated the feasibility and cost of targeted near-POC viral load (VL) testing in two large HIV clinics in Lilongwe, Malawi. Methods: VL testing using GeneXpert was targeted for patients suspected of treatment failure or returning to care following a previously elevated VL (>1,000 copies/ml). Descriptive analysis of retrospective clinical and cost data are presented. Results: During the 12-month study period, 2813 near-POC VL tests were conducted. 1511 (54%) tests were for onsite patients for whom results and reason for test were documented: 53% (794/1511) of tests were to confirm a previously high VL, and 31% (462/1511) were due to clinical indications. Overall, 61% (926/1511) of patients had a high VL, of whom 78% (719/926) had a recorded clinical action: 77% (557/719) switched to second line ART and 15% (194/719) were referred for intensive adherence counseling. Nearly 79% (567/719) of patients received a clinical action on the same day as testing. The ‘all-in’ cost was $33∙71 for a valid near-POC VL test, compared to an international benchmark for a centralized VL test of $28∙62. Conclusion: Targeted, near-POC VL testing was feasible and consistently enabled prompt clinical action. According to national data, fewer than 3% of patients are on a second-line regimen, suggesting that routine VL results are likely underutilized; near-POC VL may be an attractive modality to ensure patients are on optimized regimens. The difference between the ‘all in’ cost of near-POC VL and centralized testing of $5∙09 could be further reduced in an optimized national program combining targeted near-POC testing and centralized testing.

Poster Abstracts

1086 OPTIMIZATION OF HIV CLINIC INTAKE PROCESS TO REDUCE TIME TO VIRAL SUPPRESSION Milana Vachuska 1 , Kari Abulhosn 1 , Craig Ballard 1 , Jennifer Blanchard 1 , Lucas Hill 1 1 University of California San Diego, San Diego, CA, USA Background: This study describes a novel clinic intake process to more rapidly initiate antiretrovirals (ARVs) and compares mean time to initiate ARVs as well as time to viral suppression from before and after the advent of this new intake process. Methods: In April 2018, the UC San Diego Owen Clinic developed a new intake process that included an initial visit with a multidisciplinary team to improve access while providing the opportunity for rapid initiation or optimization of ARVs at first visit using clinical pharmacy services. Prior to this new process, patients only met with licensed vocational nurses (LVNs) at their initial intake visit. We conducted a retrospective study comparing time to initiate ARVs as well as time to viral suppression before and after this new intake process was implemented. We also evaluated clinic retention rates within both 1 month and 6 months of initial visit. Predictors of lack of retention in care were also evaluated. Results: We included 379 patients in the analysis. Table 1 shows demographic data, psychosocial data, and baseline virologic data. In the new intake cohort, there were significant reductions in mean time to initiate ARVs (54.2 days vs. 7.8 days, p=0.0002) and mean time to viral suppression (217.9 days vs. 75.9 days, p<0.0001). There was no significant difference in the proportion of patients retained either short or long term. Although not statistically significant, after logistic regression there was a trend that black patients were more likely to fall out of care long term (p=0.0535). Conclusion: We observed significant reductions in time to initiate ARVs and time to viral suppression in the new, pharmacist driven intake cohort. Similar

1088LB 3- VS 6-MONTHLY DISPENSING OF ART IN COMMUNITY ART GROUPS: A CLUSTER-RANDOMIZED TRIAL Geoffrey Fatti 1 , Nicoletta Ngorima-Mabhena 1 , Ashraf Grimwood 1 , Eula Mothibi 1 , Charles Chasela 2 , Risa M. Hoffman 3 , Owen Mugurungi 4 , Tsitsi Apollo 4 , Kudakwashe Takarinda 4

CROI 2020 409