CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

and counseling, and ARV dispensing. Rapid procedures were conducted by nurses and counselors comparable to clinic staff. Patients in the control arm followed standard clinic procedures (3-4 clinic visits over 2-4 weeks prior to ARV dispensing). Here we report ART uptake and early retention ≤ 1 month of ART initiation. Results: Enrollment was completed in August 2014. 172 patients were randomized to rapid and 181 to standard initiation. There were no important differences between arms in gender, age, or CD4 count at study enrolment. In both arms, 83% of patients newly diagnosed with HIV were already eligible for ART. In the rapid arm, 97% (139/143) of patients eligible for ART initiated treatment ≤ 1 month, including 73% on the same day as study enrollment and 19% ≤ 1 week. In the standard group, 57% (86/151) initiated ≤ 1 month (hazard ratio for ART uptake=1.69; 95%CI 1.47-1.95). All those who did not initiate ≤ 1 month in the rapid arm (n=5) were required to delay ART for TB treatment. Time used for treatment initiation, from study enrollment to dispensing, averaged 2.8 hours for rapid arm patients not requiring TB testing. Within 1 month of initiation, 86% (n=120) of rapid arm and 85% (n=112) of standard arm patients had attended the clinic for their first follow-up visit. Of 54 subjects with ≥ 6 months follow-up and documented viral load, 91% and 77%were virally suppressed in the rapid and standard arms respectively.

Outcomes and Relative Risks by Study Arm Conclusions: Immediate ART initiation reduces loss of patients between treatment eligibility and treatment initiation significantly and is feasible and acceptable in a public health clinic setting. It should be considered for adoption in high-volume clinics in the public sector in Africa. 1092 Outcomes of a Clinic-Health Department “Data to Care”Relinkage Intervention Joanna M. Bove 2 ; Matthew R. Golden 2 ; Shireesha Dhanireddy 2 ; Robert Harrington 2 ; Julie Dombrowski 2 1 University of Washington, Seattle, WA, US; 2 University of Washington, Seattle, WA, US Background: The effectiveness and best structure of programs to identify out-of-care persons with HIV and relink them to care is uncertain. We implemented and evaluated an HIV care relinkage intervention that uses both clinical and health department surveillance data. Methods: The intervention occurs in the Madison Clinic, a university affiliated, Ryan-White funded clinic in Seattle, WA. The program uses a clinic registry to identify persons who are potentially out of care. The registry collates data from electronic health records, intake forms, and case management databases. HIV+ patients are eligible for relinkage outreach if they completed ≥ 1 clinic visit in the last 1000 days but no visits for ≥ 12 mo. A linkage specialist (LS) investigates cases and conducts outreach with the goal of care relinkage. We analyzed LS outreach success over a 12 mo. observation period among patients identified for outreach on 11/1/12 (intervention cohort). We compared outcomes of this cohort to patients who would have met criteria for the intervention one year prior to its initiation (historical cohort). We used c 2 tests and logistic regression to compare the percent relinked and the mean time to relinkage between cohorts. Results: 753 patients were identified as out of care on 11/1/12. Matching with surveillance data determined that almost half (347 [46%]) of these patients had moved, transferred care or were incarcerated (ineligible). LS investigations determined that an additional 245 (33%) were ineligible. Of the 161 (21%) truly out of care patients, 40 (25%) relinked before LS contact. Of the remaining 121, 38 (31%) did not have contact information; 46 (38%) did not respond to contact attempts; and 37 (31%) were successfully contacted, of which 20 (54%) relinked. In all, 116 (15%) intervention cohort patients relinked to care in 12 months; 24 (21%) were among the LS attempted contacts. 48 (42%) of the patients who relinked were initially identified as ineligible, but transferred back to Madison or were released from incarceration. More patients in the intervention cohort than the historical cohort relinked (15% vs. 10% [RR=1.6 (1.2-2.1)]), and they had a shorter mean time to relinkage (4.8 vs. 6.3 mo.; p=0.001). Conclusions: This collaborative HIV clinic-health department relinkage intervention showed modest ability to identify and return out-of-care patients to care compared to historical controls and highlights the utility and efficacy of integrating clinical and surveillance data in relinkage efforts. 1093 HIV Partner Services Can Achieve Near-Universal Linkage to HIV Care David A. Katz 1 ; Julia C. Dombrowski 1 ; Susan E. Buskin 2 ; Amy Bennett 2 ; Elizabeth A. Barash 2 ; Matthew R. Golden 1 1 University of Washington, Seattle, WA, US; 2 Public Health - Seattle & King County, Seattle, WA, US Background: Timely linkage to care following HIV diagnosis is necessary for maintaining the health of persons living with HIV and realizing the secondary prevention benefits of antiretroviral therapy and behavior change. HIV partner services (PS) provide an opportunity not only to test exposed partners but also to facilitate timely linkage for persons with newly diagnosed HIV infection (index cases). Methods: In King County, WA, PS staff attempt to provide PS to all persons newly diagnosed with HIV infection. PS are designed to ensure that index cases link to care and that partners are notified and tested. PS staff do not close cases until they have verified that index cases have linked to care. Using HIV surveillance and PS data, we examined the impact of receiving PS on timely linkage to care, defined as first CD4 count or HIV viral load (VL) within 3 months of initial HIV diagnosis, using chi-squared tests and logistic regression. Results: From 2010-2013, 1043 persons aged 15 and older were newly diagnosed with HIV infection in King County, of whom 963 (92%) linked to care within 3 months and 999 (96%) within 1 year. Only 18 (1.7%) did not have ≥ 1 CD4 or VL reported to surveillance by 9/8/2014. Of 1043 new cases, 838 (80%) received PS, 250 (30%) of whomwere interviewed during a public health program called One-on-One, which provides patients with counseling and an initial clinical assessment, including CD4 and VL testing. PS recipients were more likely to link to care within 3 months than non-recipients (94 v. 84%, p<.001); this association persisted even when One-on-One clients were excluded from the analysis (93 v. 84%, p<.001). In multivariable analysis, receiving PS remained significantly associated with timely linkage to care (p<.001); sex, gender of partners, age, diagnosing provider type, and year of diagnosis were not associated with timely linkage. Men who have sex with men (MSM), who represented 814 (78%) of cases during the study period, were more likely to receive PS than non-MSM (82 v. 74%, p=.009) but had identical levels of timely linkage to care (92 v. 92%, p=.9). Conclusions: Identifying linkage to care as an explicit outcome for HIV PS can increase timely linkage to care among persons newly diagnosed with HIV infection and has achieved nearly universal linkage to HIV care in King County, WA.

Poster Abstracts

636

CROI 2015