CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

1026 ACCESS TO HIV CARE CORRELATES WITH DEPRESSION SEVERITY AND RATES OF VIRAL SUPPRESSION Ann K. Avery 1 , Mallika Lavakumar 1 , Allison R. Webel 2 , James Alsop 3 , Doug Gunlzer 1 , Diana Gurley 1 , Steven Lewis 1 1 MetroHealth Medical Center, Cleveland, OH, USA, 2 Case Western Reserve University, Cleveland, OH, USA, 3 University of Minnesota, Minneapolis, MN, USA Background: Depression is among the most common, yet unaddressed, problems identified in people living with HIV (PLWH). Under-diagnosis and under-treatment of depression in PLWH contributes to negative health outcomes. The collaborative care model (CCM) has been shown to improve both depression outcomes and co-morbid medical outcomes in primary care but there is limited data on its use in HIV care settings. The CCM includes routine screening for depression with the PHQ-9, measurement-based care and care management for all patients scoring >=10. Using an implementation science framework, we rolled out the CCM in our HIV clinic from June 2015 -June 2016. Methods: All patients with PHQ screening data were included. Patients with any score >= 10 entered the CCM. Data from June 2015 - Dec 2017 were analyzed to identify factors associated with greater severity of depressive symptoms at initial presentation. A multiple linear regression model was used to regress first PHQ-9 score for patients in CCM on a set of demographic, clinical, access-related characteristics to determine correlates of depression at baseline. A generalized estimating equations approach was used to evaluate if subjects in CCM compared to subjects not in CCM had higher HIV viral suppression over the subsequent 12 months after initial presentation. Results: 1473 patients were screened for depression between 6/29/15 and 12/31/17; 594 reported moderate to severe symptoms at least once (PHQ-9 >= 10). Patients who did not have a viral load documented in the year prior to the initial PHQ-9 score reported more severe depressive symptoms than those who had a viral load collected in the year prior (p=.004). Additionally, compared to patients with Medicaid, patients who were uninsured had more severe symptoms (p=.0003), while Medicare recipients reported less severe symptoms (p=.0393). The GEE approach did not demonstrate differences in achieving viral suppression over time between groups. However, the CCM group were 34% less likely to be virally suppressed at first PHQ-9 (OR 0.66 CI 0.52, 0.84). Additionally, patients in CCM but did not follow up for re-measurement within 1 year (n=180) were 65% less likely to be virally suppressed at first PHQ-9 compared to patients who never reported depressive symptoms (OR 0.35 CI 0.19, 0.64). Conclusion: Depressive symptoms were present in 1/3 of patients; interventions to engage PLWH reporting depressive symptoms should be given priority in efforts to improve HIV viral suppression rates. 1027 WITHDRAWN FACILITATED LINKAGE TO CARE FOLLOWING HOME-BASED HIV TESTING IN RURAL SOUTH AFRICA Kathy Baisley 1 , Janet Seeley 1 , Mark J. Siedner 2 , Olivier Koole 2 , Philippa Matthews 2 , Till Bärnighausen 3 , Theresa Smit 2 , Dickman Gareta 2 , Sphephelo Dlamini 2 , Kobus Herbst 2 , H. Manisha N. Yapa 2 , Collins C. Iwuji 4 , Hae-Young Kim 2 , Deenan Pillay 2 , Maryam Shahmanesh 5

1025LB CASCADE TRIAL: 24 MONTH OUTCOMES AFTER SAME-DAY HOME-BASED ART INITIATION Alain Amstutz 1 , Isaac Ringera 2 , Thabo I. Lejone 2 , Josephine Muhairwe 2 , Jennifer A. Brown 3 , Thomas Klimkait 3 , Tracy R. Glass 1 , Niklaus D. Labhardt 1 1 Swiss Tropical and Public Health Institute, Basel, Switzerland, 2 SolidarMed, Luzern, Switzerland, 3 University of Basel, Basel, Switzerland Background: The CASCADE trial, conducted in Lesotho, Southern Africa, has shown that offering same-day initiation of antiretroviral therapy (ART) to individuals found HIV positive during home-based testing resulted in significantly higher proportions engaged in care and virally suppressed 12 months after the testing campaign. At completion of the trial all patients not in care were traced and the protocol was amended to allow for a 24 months follow- up of study participants. Methods: CASCADE was a randomized clinical trial that assigned individuals recruited during a home-based HIV testing campaign to either the offer of same-day ART start (SD) or referral to a nearby clinic for preparatory counseling and ART start after ≥2 pre-ART clinic visits (UC). Consenting ART-naïve, HIV- infected individuals, ≥18 years, were enrolled. Methods and 12 month results were published previously (Labhardt et al. JAMA. 2018;319(11):1103). At 12 months those not active in care were traced by health workers and encouraged to return to care. At 24 months (range 22-28 months), engagement in care, viral suppression (<100 copies/mL) and reasons for disengagement were assessed among all trial participants. Trial registration: NCT02692027 Results: The care cascade and the status of patients at 24 months are displayed in Figure 1. Of 274 individuals randomized (137 SD, 137 UC), 64% (87/137) in the SD and 48% (66/137) in the UC group were active in care 12 months after testing positive (p=0.011), and 50.4% (69/137) vs 34.3% (47/137) had documented viral suppression (p=0.007). At 24 months, 64% (88/137) in the SD versus 59% (81/137) in the UC armwere in care (p=0.38) and 57% (78/137) vs 54% (74/137) had documented viral suppression (p=0.28). Among those active in care at 12 months, 11% (10/87) and 9% (6/66) were no longer in care at 24 months (p=0.63). Among those not in care at 24 months, 31% (15/49) and 38% (21/56) had been found through tracing but refused care. Most cited reasons were disbelieving in diagnosis/ART (N=6), discomfort taking medication (5), rejection of any contact with health system (4) and perceived ill-treatment by health professionals (3). Conclusion: After tracing of all participants not in care at 12 months, a significant difference was no longer observed between the SD and the UC arm regarding viral suppression and engagement in care at the 24-month follow-up. Both arms remained below the targeted 90% of people living with HIV receiving ART. One third of those not in care refused attending.

Poster Abstracts

CROI 2019 402