CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

(CI: 5.4-7.1) more than if started on ART at the facility. Compared to adults >25 years, the odds of attrition among 0-9, 10-19 and 20-24 years was 1.6 (CI: 1.2-2.1), 1.9 (CI: 1.4-2.4) and 2.2 (CI: 1.8-2.6), respectively. Compared to those on ART for 0-12 months, odds of attrition for those on ART 13-24 months and >24 months was 0.39 (CI 0.34-0.45) and 0.21 (CI: 0.18-0.23) respectively. VL suppression were significantly lower amongst attritions, regardless of reason for attrition; Odds Ratios 0.32 (CI: 0.24-0.42), 0.40 (CI: 0.27-0.60) and 0.18 (CI: 0.09-0.38) among TO, LTFU and dead respectively Conclusion: In a Kenyan urban population, age, being a transfer-in and duration on ART were strong predictors of attrition. Attritions had lower VL uptake and viral suppression rates compared to active patients. Patients transferring-in on ART had particularly poor outcomes suggesting vulnerabilities that should be considered in their care. The process of patient transfer also merits review Kenneth Masamaro 1 , Jacques Muthusi 1 , Maureen Kimani 2 , Irene Mukui 2 , Evelyne Ngugi 1 , George Mgomella 1 , Tai Ho Chen 1 1 US CDC Nairobi, Nairobi, Kenya, 2 Ministry of Health, Nairobi, Kenya Background: Scale up of antiretroviral treatment (ART) in Kenya included three major guideline changes during 2003–13, prior to adoption of ART eligibility for all HIV-infected persons in 2016. We hypothesize that ART scale up reduced delays in ART initiation for eligible patients. Methods: We conducted a retrospective chart review of HIV-infected patients aged ≥15 years enrolled in 50 health facilities in Kenya with ≥50 adults on ART during 2003–13. Primary outcomes included proportion of eligible patients with delayed ART initiation (defined as ART initiation >1month after enrolment) and median time to ART initiation. Guideline periods (AP) were adjusted to include a six month implementation period: Jan 2003–Jun 2006 (AP1, with ART recommended for patients with CD4 count <200 cells/mm 3 ), Jul 2006–Dec 2010 (AP2, ART for CD4 <250), and Jan 2011–Sep 2013 (AP3, ART for CD4 <350). We calculated weighted proportions and 95% confidence intervals (95% CI) as well median and inter-quartile range (IQR). We used Pearson chi-square to test for differences in proportion and Kruskal-Wallis statistics to test differences in the distribution of time to ART. Results: Of 3152 sampled patients, 2103 (66.7%) were women; median age at enrolment was 36.5 years (IQR 30.6, 44.3) for men and 31.5 years (IQR 25.4, 38.8) for women. Among this cohort, 1624 (51.5%) patients were eligible for ART initiation at time of enrolment. During AP1, 65/184 of eligible patients (35.0%, 95% CI 27.3–42.8%) were started on ART <1 month after enrolment compared to 305/847 (36.7%, 95% CI 31.7–41.8%) and 271/593 (44.3%, 95% CI 37.9–50.7%) in AP2 and AP3, respectively. Median days from enrolment to ART initiation in eligible patients were 75 (IQR 21, 271), 56 (IQR 20, 229), and 40 (IQR 15, 179) in periods AP1, AP2, and AP3, respectively (p<0.001). Patients in national referral facilities were less likely to have delayed ART initiation than those in lower tier facilities during AP3, but not in earlier guideline periods. Patients with opportunistic infections or recent TB diagnosis were less likely to have delayed ART initiation. Conclusion: Median time from enrolment to ART initiation decreased during successive guideline periods from 2003–2013. However, most patients eligible for ART at enrolment into HIV care experienced delays (≥1 month) in ART initiation during 2003–13 in Kenya. Treatment programs should continue to monitor for and address delays in ART initiation. 1094 YOUTH-FOCUSED CARE IN AN ADULT CLINIC IMPROVES RETENTION FOR YOUNG ADULTS WITH HIV David C. Griffith 1 , Shanna Dell 1 , Jeremy Snyder 1 , Samantha Greenblatt 2 , Jeanne C. Keruly 1 , Allison Agwu 1 1 Johns Hopkins University, Baltimore, MD, USA, 2 Johns Hopkins Hospital, Baltimore, MD, USA Background: Young adults with HIV (YAHIV) are less likely to be retained in care or achieve viral suppression when seen in adult clinics. We assessed outcomes of YAHIV newly entering or transitioning from pediatric care into a youth-focused care model embedded in an adult HIV clinic. Methods: The Accessing Care Early (ACE) program for YAHIV is embedded in a large adult HIV clinic. Providers are internal medicine/pediatrics trained; the support team includes a nurse, social worker, and peer navigator. Eligibility for ACE includes age 18-30 years with ≥1 criteria: transfer from pediatric care, mental illness, substance abuse, or known adherence issues. Ineligible patients 1093 TIME TO ART INITIATION IN KENYA, 2003-2013

receive standard of care (SOC) in the general adult clinic. We performed a retrospective analysis of patients 18-30 years old entering ACE vs SOC from 2012- 2014. Multivariable logistic regression assessed retention, HIV viral suppression (VS) < 200 copies/mL, and the association between clinical services utilization (nurse visits and telephone calls, social work visits, psychiatry visits, and peer navigator communication) and retention and VS. Results: 137 patients entered care (2012-2014), 61 ACE and 76 SOC. In ACE 23% had perinatal HIV vs 3% in SOC; 39% of ACE transitioned from pediatric care compared to 5% in SOC. ACE YAHIV were more likely to have substance abuse, mental health disorder, and less education. Overall ACE YAHIV were less likely to be lost to follow up compared to SOC (16% vs. 37%, p<0.01). At 24 months 49% in ACE vs. 26% in SOC met the retention measure, (P<0.01). Adjusting for age, gender, race, HIV risk, viral load, CD4, mental health, and substance abuse, ACE was associated with retention in care (AOR 3.26 [1.23-8.63]). For those who met the retention measure, 60% (15/25) of ACE versus 89% (16/18) of SOC were virally suppressed (AOR 0.63 [0.35-1.14]). Adjusting for ACE vs. SOC, more frequent social work visits and nurse phone calls was associated with retention. Appointments were less likely to be missed if peer navigator confirmed via a bi-directional communication (OR 2.69 [1.64-4.42]). Conclusion: The youth-focused ACE program successfully identified YAHIV at high risk for attrition and viremia. Despite comprising higher risk YHIV, ACE had better retention compared to SOC for YAHIV in an adult clinic. Improved retention did not to lead to improved VS compared to the SOC, underscoring the challenges with adherence and need for additional interventions to optimize VS for YAHIV.

Poster Abstracts

1095 CAN FOOD HELP RETENTION IN HIV CARE: A COHORT STUDY OF ADULTS INITIATING ART IN HAITI Julie Reiff 1 , Stanislas Galbaud 2 , Akanksha Dua 1 , Jean Edouard Mathon 2 , Patrice Joseph 2 , Adias Marcelin 2 , Pierrot Julma 2 , Serena Koenig 3 , Pierre Cremieux 1 , Jean W. Pape 2 , Margaret McNairy 4 1 Analysis Group, Inc, Boston, MA, USA, 2 GHESKIO, Port-au-Prince, Haiti, 3 Brigham and Women’s Hospital, Boston, MA, USA, 4 Weill Cornell Medicine, New York, NY, USA Background: Attrition from HIV care is highest in the first 6 months after initiation of antiretroviral therapy (ART) in resource-poor settings. Food was distributed to adults after ART initiation for 6 months at GHESKIO in Haiti, with the goal to improve retention in care. We evaluated the association of food with early retention in care. Methods: This retrospective observational cohort study included routinely collected data from HIV-infected, non-pregnant adults ≥ 18 years who initiated ART fromMarch-December 2016. During this time, food was distributed in monthly packages of 10 lb rice, 5 lb beans, 1 litre cooking oil, 3 pasta boxes, 3 tomato paste cans, and 4 milk cans for 6 months per patient. Early retention in

CROI 2018 420