CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

1.00-1.89, p=0.05). Getting care at a hospital was associated with a reduced hazard of return (aHR: 0.55, 95%CI: 0.35-0.86, p=0.01) (Table 1). Conclusion: Despite in-person peer educator tracing and encouragement to return, fewer than half of disengaged patients did so. Interventions which improve facility access and target young people may reduce treatment interruptions. New approaches to facilitate re-engagement and improve HIV program success should be explored.

Poster Abstracts

1032 PATTERNS AND PREDICTORS OF RETURN TO CARE AMONG DISENGAGED HIV PATIENTS IN ZAMBIA Laura K. Beres 1 , Ingrid Eshun-Wilson 2 , Sandra Simbeza 3 , Sheree Schwartz 1 , Aaloke Mody 2 , Kombatende Sikombe 3 , Jake Pry 3 , Paul Somwe 3 , Carolyn Bolton Moore 3 , Njekwa Mukamba 3 , Steph Topp 4 , Nancy Padian 5 , Izukanji Sikazwe 3 , Charles B. Holmes 6 , Elvin Geng 2 1 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 2 University of California San Francisco, San Francisco, CA, USA, 3 Center for Infectious Disease Research in Zambia, Lusaka, Zambia, 4 James Cook University, Townsville, Australia, 5 University of California Berkeley, Berkeley, CA, USA, 6 Georgetown University, Washington, DC, USA Background: Loss to follow-up from HIV treatment has been widely documented, but re-engagement in care – a critical dimension of long-term success – has not been examined closely. To inform interventions to reduce treatment interruptions, we sought to characterize re-engagement after peer- tracing and predictors of return. Methods: We traced a stratified, random sample of patients from 64 Zambian health facilities who had at least one facility visit between 1st August 2013 – 31st July 2015 but were lost to follow-up, defined as an unknown care status and >90 days from last visit. Among patients reporting disengagement, tracers encouraged return to care and administered a survey recording potential return predictors: reasons for disengagement, internalized and anticipated stigma, household violence, alcohol use, past retention support, wealth, role in household, mobility, disclosure, marital status, education, proximity to facility, and requirements for return. Using electronic medical records (EMR) linked by patient identification number, we extracted visit dates subsequent to tracing to estimate the proportion with a return visit, demographic characteristics, and HIV and ART history. We used Kaplan-Meier methods to estimate cumulative incidence of return and Cox proportional hazards models to identify predictors of return. A combination of theory and univariate association significance was applied to determine the final model. Results: Of the 2,769 ‘lost’ patients traced, 603 reported disengagement, 571 had follow-up EMR visit data, and 38.0% (95%CI: 33.7-42.7) had a return visit by the end of the study. Median follow up time was 595 days (IQR: 214-667, max: 836). Proportions returning at 30, 180, and 365 days were: 11.2% (95%CI: 8.9-14.1); 24.5% (95%CI: 21.2-28.3); and 29.4% (95%CI: 25.9-33.4). Significant predictors of care included age >50 years (aHR: 1.89, 95%CI: 1.04-3.46, p=0.04) and reporting the most stigma (aHR: 1.73, 95%CI: 1.06-2.83, p=0.03) with residence in facility catchment suggestive of increased return (aHR:1.37, 95%CI:

1033 LONG-TERM OUTCOME IN HIV-1 INFECTED ADULTS WITH ADVANCED IMMUNODEFICIENCY Guoqing Zhang 1 , Kouakou Adade Aka Digbo Micheline 2 , Peter A. Minchella 1 , Kouadio Leonard Ya 2 , Andre Tehe 2 , Mireille Kalou 1 , Karidia Diallo 1 , Judith Hedje 2 , Heather Alexander 1 , Clement Zeh 1 , G. Laissa Ouedraogo 2 , Christiane Adje-Toure 2 1 CDC, Atlanta, GA, USA, 2 CDC, Abidjan, Côte d’Ivoire Background: Prior to the President’s Emergency Plan for AIDS Relief (PEPFAR) Program, patients initiating antiretroviral treatment (ART) in resource-limited settings frequently had very advanced immunodeficiency. In this study, we examined CD4 cell count recovery and viral suppression among patients with very low baseline CD4 count in Côte d’Ivoire. Methods: We identified 1,883 HIV-1-infected adults who initiated ART at clinics in Abidjan between August 1, 1998 and July 31, 2004 and had follow-up data between August 1, 2004 and July 31, 2015. Among them, 474 had CD4 cell count less than 50 cells/μL at ART initiation. We retrospectively analyzed baseline CD4 count, CD4 recovery, and viral suppression data collected from patient laboratory records (1998-2015) in the CDC Projet Rétrovirus Côte d’Ivoire (RETRO-CI) laboratory information system. Results: After 10 years of follow-up, 231 of 474 patients (48.7%, 95% confidence interval [CI]: 44.3%-53.2%) with baseline CD4 count less than 50 cells/μL remained in care at the end of the study period, between August 2014 and July 2015. Their CD4 cell count increased from a median of 18.9 cells/μL at baseline to 517.5 cells/μL, with a sharp increase from 95.4 to 456.0 cells/μL occurring during the first six years of PEPFAR in the country (2004-2010). At their last visit, 50.6% of them had a CD4 count over 500 cells/μL, the proportion was comparable to that of patients with higher baseline CD4 counts of 50-99.9, 100- 199.9, and 200-349.9 cells/μL, but was lower than that of patients with baseline CD4 count over 350 cells/μL (Bonferroni-adjusted P values <0.01). Logistic regression (adjusted for age and year of ART initiation) showed that females were more likely to achieve CD4 recovery (adjusted odds ratio is 2.56, 95% CI: 1.76-3.72). Among the 231 patients, 58.0% (n=134) had viral load (VL) results recorded between August 2014 and July 2015, and 87.3% (117/134) of them had

CROI 2019 405