CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

993 UCARE4LIFE: Mobile Texting to Improve HIV Care Continuum Outcomes for Minority Youth Carrie Jeffries 1 ; Polly Ross 1 ; Sabrina Matoff-Stepp 1 ; RenataThompson 1 ; Jennie L. Harris 2 ; Jennifer D. Uhrig 2 ; Laura Cheever 1 1 US DHHS, Rockville, MD, USA; 2 RTI, Research Triangle Park, NC, USA

Background: Minority youth living in the Southern U.S. face higher HIV rates and faster disease progression than non-minority youth. According to the Pew Institute, over 75% of youth use cell phones and the mean number of texts sent and received per day is 60. This study examined the utility of a text message program as an HIV self-management tool. Methods: A prospective, randomized two-group pilot study was conducted from Fall 2014 thru Summer 2015 to assess the promise of text messaging among minority youth to increase their retention in care and HIV medication adherence. Eligible participants included English-speaking HIV-positive youth aged 15 to 24, who had personal cell phones with text-messaging capability whom received their HIV medical care at one of three clinics located in Louisiana, Alabama, or North Carolina. A total of 341 unique, culturally appropriate text messages were delivered from domains such as treatment and appointment adherence, HIV basics, clinical visits, social support and risk reduction. Data were collected via medical record abstraction and pre/post-intervention assessment focus groups and behavioral assessment surveys. Results: A total of 164 youth were eligible and 146 voluntarily enrolled. The demographics of those enrolled were 86%male, 77% African American, 66% homosexual, and 85%were aged 21 thru 24. The intervention group received a mean number of 12 texts a week; the control received standard of care that did not include text messaging. T-tests, chi-squares, and McNemar statistical analysis were employed for data analysis. Post-intervention data showed that the viral load was significantly statistically lower for the intervention group compared to the control group at three (p=.039) and six months (0.003) post-baseline among youth who were non-adherent or new to ART at baseline. There was no statistically significant difference between retention in medical care for the intervention group versus control group. Six months post-baseline, the intervention youth demonstrated a statistically significant better understanding of the effects of substance use on their ability to remember to take their ART (p=.007) than among the controls. Conclusions: Delivering culturally appropriate, timely, tailored text messages to HIV positive youth is a promising new intervention. Youth who were non-adherent or new to taking ART benefited most with lower viral loads and understanding of the effects of substance use on their health. 994 Missed Opportunities: Adapting the HIV Care Continuum to Reduce HIV-Related Deaths Sarah Braunstein ; Rebekkah Robbins; Demetre C. Daskalakis New York City DHMH, Queens, NY, USA Background: With optimal care and treatment, persons with HIV (PWH) can lead healthy lives. However, HIV-related deaths among PWH remain common in New York City (NYC). We developed a novel continuum of HIV care that estimates care outcomes before death to identify opportunities to prevent HIV-related death. Methods: We selected PWH from the NYC HIV Surveillance Registry who died during 2007-2013, resided in NYC at death, and who died ≥15 months after HIV diagnosis. The primary analysis used CD4 and viral load (VL) tests from surveillance to measure outcomes during the “intervenable period,” the period from 15 months to 3 months before death. Outcomes included the proportion of patients who were: ever linked to care; retained in care; presumed to have ever started antiretroviral therapy (ART); clinically well-managed (VL≤1500 cc/mL); and virally suppressed (VL≤200 cc/mL). We stratified the continuum by underlying cause of death (COD) (HIV-related vs. other), and then the HIV-related continuum by sex, race/ethnicity, age at death, and transmission risk. Results: 11,187 PWH died during 2007-2013 and were eligible for analysis. Overall, 98% (N=11,007) linked after HIV diagnosis; 80% (N=8,992) were retained in care during the intervenable period; 66% (N=7,376) had ever started ART; 47% (N=5,217) were clinically well-managed; and 40% (N=4,518) were virally suppressed (VS). Half (47%) of all deaths were HIV-related (Figure). Retention in care was higher among PWH with HIV-related COD (83% vs. 78%), but VS was substantially lower (34% vs. 46%). The stratified HIV-related continuum revealed disparities in VS. Despite comparable retention rates, whites had higher VS (42%) than blacks (32%) and Hispanics (33%). Retention and VS increased with increasing age at death: 75% of persons <30 years were retained and 30% had VS, vs. 88% and 56%, respectively, among persons ≥60. By risk, men who have sex with men had the lowest retention rate (81%), and persons with a history of injecting drugs had the highest VS (37%). Conclusions: Although retention was high among NYC PWH who died during 2007-2013, VS was low, at nearly half that among persons living with HIV (64% in 2013). High retention coupled with low VS suggests the need to develop strategies to improve VS and address psychosocial and structural barriers to optimal clinical management. The HIV mortality reduction continuum is a novel framework for evaluating pre-death care outcomes among PWH and identifying opportunities for intervention.

Poster Abstracts

995 A Novel Practical Spatial Analysis of HIV Care Outcomes, Metro Atlanta, 2012-2014

Neela D. Goswami 1 ; Jeffrey McMichael 2 ; Deepali Rane 2 ; Jane Kelly 2 ; Julie Clennon 2 ; Carlos del Rio;Travis Sanchez 3 1 Emory Univ Sch of Med, Atlanta, GA, USA; 2 Georgia Dept of PH, Atlanta, GA, USA; 3 Emory Univ, Atlanta, GA, USA

Background: Spatial tools may facilitate strategic resource deployment to address local HIV continuum outcomes (linkage to care, viral suppression) in a targeted way, particularly when cases are visualized as absolute numbers per square mile. We aimed to characterize spatial (1) clustering of newly diagnosed HIV patients in metropolitan Atlanta, (2) clustering of their continuum outcomes with surveillance data collected over an 18-month period, and (3) overlap with tuberculosis (TB) and homelessness. Methods: We analyzed HIV cases diagnosed in 2012 and TB cases diagnosed 2009-2014 reported to Georgia Department of Public Health (GDPH) HIV/AIDS Epidemiology and TB programs, and homeless persons from Atlanta’s 2011 homeless count. With GDPH-reported CD4 and HIV viral loads (VL), “poorly linked to care” was defined by absence of CD4 or VL by three months, and “virally unsuppressed” by absence of a VL <200 copies/mL by 18 months from diagnosis. We evaluated for spatial clusters with K-function. Bernoulli spatial scan was applied to locations of co-infected persons to assess for clustering of excess co-infection ( ArcGIS 10.3.1, SatScan 9.4.2).

423

CROI 2016