CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

all” guidelines should continue to be scaled-up to achieve 90-90-90 targets and reduce HIV morbidity and mortality.

earlier HIV diagnosis and linkage to care, critical for reaching 90-90-90 targets. These trends suggest that CD4 counts at entry into care and at ART initiation will continue to jointly increase over time with expanded implementation of effective test-and-treat strategies.

Poster Abstracts

1024 PATIENT-REPORTED REASONS FOR DECLINING IMMEDIATE ART INITIATION IN LUSAKA, ZAMBIA

1023 NARROWING THE GAP IN CD4 COUNT AT ENTRY INTO CARE AND AT ART INITIATION, 2005-2016 Jennifer S. Lee 1 , Richard D. Moore 1 , Stephen J. Gange 1 , Yuezhou Jing 1 , Marina Klein 2 , Peter F. Rebeiro 3 , Kathryn Anastos 4 , Amy C. Justice 5 , Michael A. Horberg 6 , Mari Kitahata 7 , W. C. Mathews 8 , Michael J. Silverberg 9 , Angel M. Mayor 10 , Keri N. Althoff 1 , for the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) of IeDEA 1 Johns Hopkins University, Baltimore, MD, USA, 2 McGill University Health Centre, Glen site, Montreal, QC, Canada, 3 Vanderbilt University, Nashville, TN, USA, 4 Albert Einstein College of Medicine, Bronx, NY, USA, 5 VA Connecticut Healthcare System, West Haven, CT, USA, 6 Kaiser Permanente Mid-Atlantic States, Rockville, MD, USA, 7 University of Washington, Seattle, WA, USA, 8 University of California San Diego, San Diego, CA, USA, 9 Kaiser Permanente, Oakland, CA, USA, 10 Universidad Central del Caribe, Bayamon, Puerto Rico Background: In March 2012, the US Department of Health and Human Services updated HIV treatment guidelines to recommend antiretroviral therapy (ART) for everyone infected with HIV, regardless of CD4 count, to reduce morbidity and mortality among those infected and prevent transmission to others. Our objective was to describe observed trends in CD4 count, at entry into care and at ART initiation, among patients enrolled in US-based clinical cohorts of the NA-ACCORD between 2005 and 2016. Methods: The study sample comprised treatment-naïve adults (aged ≥18 years) without a clinical AIDS diagnosis who presented for HIV care with a viral load >500 copies/mL (-180/+14 days) and a recorded CD4 count (-90/+30 days). A subset of the study sample initiated ART (defined as being prescribed a combination ART regimen) with a recorded CD4 count (-90/+30 days). For patients with >1 CD4 count collected during the 120-day window, we used the measurement obtained closest to the visit date of interest. We generated plots of median CD4 counts at entry into care and at ART initiation, by calendar year. We also calculated median number of days from entry into care to ART initiation, by calendar year. Results: We identified 28862 patients who entered care; of those patients, 23521 initiated ART. Median CD4 count at entry into care was 302 (IQR: 115–481) cells/μL in 2005, 360 (IQR: 174–545) cells/μL in 2012, and 370 (IQR: 211–565) cells/μL in 2016. Median CD4 count at ART initiation was 157 (IQR: 51–287) cells/ μL in 2005, 346 (IQR: 182–507) cells/μL in 2012, and 382 (IQR: 207–583) cells/ μL in 2016. Median number of days from entry into care to ART initiation was 70 (IQR: 20–546) in 2005, 29 (IQR: 12–74) in 2012, and 12 (IQR: 0–25) in 2016. Of patients who initiated ART after entering care in 2016, 31% initiated ART on day of presentation and 4% initiated ART ≥60 days later. Conclusion: Median CD4 counts at entry into care and at ART initiation have been trending towards convergence since 2005 and clinically equivalent since 2012, reflecting the reduction in time from entry into care to ART initiation and adoption of “treat all” in clinical practice in the US. Additionally, the increase in CD4 count at presentation over time indicates progress towards

Jake Pry , Jenala Chipungu, Carolyn Bolton Moore, Jacob Mutale, Helene Smith, Theodora Savory, Michael Herce Centre for Infectious Disease Research in Zambia, Lusaka, Zambia Background: Programs are focusing increased resources to meet the UNAIDS “second 90” treatment target. To help achieve this goal in Zambia, we developed a quality improvement tool to evaluate reasons people living with HIV (PLHIV) do not immediately link to care (LTC) and start ART. We designed the tool to be used in routine care settings to understand reasons for LTC and ART delays, and to improve individualized post-test counseling. Methods: We created a simple 1-page screening tool with structured items to capture three broad categories for failed LTC and ART delay: social, personal, and structural. We implemented the tool in three facilities, two urban and one rural, in Lusaka District over a three-month period. We administered the tool to all individuals who refused LTC and immediate ART. Individuals were allowed to choose as many reasons as relevant. Failed linkage risk was modeled using mixed effects logistic regression controlling for age, sex and testing point, and allowing random effect for clinic. Results: A total of 1,292 people with new HIV infection were identified across clinics, of whom 9.6% reported a refusal reason. Each respondent reported a median of three reasons (IQR:2-3). Of those who refused immediate LTC, 69.6% were female, with median age 30 years (IQR: 23-40 years). Females refusing LTC were younger on average at 28.5 years (IQR: 21-37 years) than their male counterparts at 34.5 years (IQR: 26-44 years). Of the 504 non-mutually exclusive responses, 87.3%were classified as personal, 62.7% as social, and 46.0% as structural. The two most commonly cited reasons for refusal were: “Clinics are too crowded” (12.3%), which was number one among females (13.8%), and “Friends and family will condemn me” (11.2%), which was most common among males (14.1%). Testing point was not significantly associated with LTC refusal (RR:1.15, 95% CI:0.03-51.51). Females were more likely to refuse (RR: 2.03, 95% CI: 1.69-2.45), as were PLHIV ages 20-24 years (RR: 5.18, 95% CI: 1.85- 14.50). Structural, personal, and social reasons for refusal differed significantly (all χ2 p<0.001) across testing points. Conclusion: The top refusal reason was associated with facility over-crowding, speaking to the importance of differentiated service delivery model scale-up to decompress busy clinics. Given the differences in refusal reasons observed across testing points, males and females, and different age bands, new, tailored LTC approaches warrant further study.

CROI 2019 401