CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Results: Between Jan 2017 and Apr 2018, 220 consecutive non-pregnant adults who met local criteria for referral to ACs were enrolled and randomised (mean age 35y; 67% female; 24% previous ART; median nadir CD4 366 cells/mm 3 ; median time on ART 18w). 88% of patients randomised to ACs attended the club visit on schedule. VL measures for the primary outcome were available on 214 participants (97%) with no differences between those retained versus lost to follow-up, overall and by arm. By 12m on ART, VL<400 cps/mL was observed in 89% of participants randomised to be referred to ACs versus 93% of participants randomised to be retained in the clinic (risk difference, -4.3%; 95% CI, -11.9% to 3.2%). The finding for similar outcomes between AC and clinic-based care was consistent across subgroups of age, gender, previous ART use and nadir CD4 cell count; in a binomial model adjusted for the same factors; and when the outcome was examined at cutpoints of VL<50 and <1000 cps/mL. Conclusion: These novel data suggest that referral of stable ART patients to community-based DSD may take place as early as 4 months after ART initiation in this setting with comparable virologic outcomes achieved at 12 months on ART versus clinic-based services. 1111 DIFFERENTIATED SERVICE DELIVERY FOR HIV CARE: THE FAST-TRACK EXPERIENCE FROM ZAMBIA Samuel Bosomprah 1 , Isaac Zulu 2 , Michael Herce 1 , Lloyd Mulenga 3 , Minesh P. Shah 2 , Izukanji Sikazwe 1 , Annie Mwila 4 , Helen B. Mulenga 1 , Muhau Mubiana 1 , Mwansa Lumpa 1 , Pamela J. Bachanas 2 , Simon Agolory 4 1 Centre for Infectious Disease Research in Zambia, Lusaka, Zambia, 2 CDC, Atlanta, GA, USA, 3 Government of Zambia Ministry of Health, Lusaka, Zambia, 4 CDC Zambia, Lusaka, Zambia Background: Differentiated service delivery (DSD) models are designed to lower barriers to HIV care for people living with HIV (PLWH). In 2017, we implemented a DSD model known as “Fast Track” (FT) within Zambia’s HIV program that provided PLWH “stable” on ART (defined as WHO stage I/II disease, on ART ≥ 6 months, and CD4+>350 or viral load suppression [VLS]) with expedited clinical services. We report clinical outcomes for FT patients during the first 2 years of implementation. Methods: We reviewed individual-level PLWH data from Zambia’s electronic health record, SmartCare. Patients 15–59 years were included in our analysis if they started ART any time from January 1, 2010 at any of 14 high-volume (>3,000 patients on ART) clinics in Lusaka. All patients in FT from its inception (January 1, 2017) through September 30, 2018 had their data reviewed to ascertain 6- and 12-month retention (i.e. any visit within 90 days of their 6- and 12-month post-ART initiation anniversaries) and VLS. To enable comparison, we reviewed records for all FT eligible patients who did not participate in FT during the same period at the same clinics. Using random-effects log binomial regression modeling, we estimated relative risk of retention in care for FT versus non-FT patients. Results: During the review period, 3,671 patients participated in FT and 83,764 did not. FT participants were more likely to be female (64.9% vs 62.3%), ≥35 years (70.9% vs 60.2%), and on ART ≥24 months (77.6% vs 73.6%) (all p<0.01); there was no difference in the proportion with WHO I/II disease (72.6% vs 72.4%). FT patients were more likely to be retained at 6- and 12-months and to achieve VLS at 6-months compared to non-FT patients (p<0.001) (Figure). After adjusting for clinic, age, sex, WHO stage, and time on ART, FT patients were 1.23 and 1.49 times as likely to be retained in care as non-FT patients at 6- and 12-months, respectively (p<0.001). Conclusion: We observed superior retention in care and VLS, and higher risk of care retention in adjusted analyses, among patients receiving FT versus non-FT services at ART clinics in Lusaka, Zambia. Due to limitations with routine data, we could not control for baseline CD4 and other unmeasured confounders. New DSD models, such as FT, hold promise for increasing care retention and VLS among stable ART patients in routine HIV treatment programs.

1112 DIFFERENTIATED CARE: TIME SPENT IN DIFFERENT ART DELIVERY MODELS IN RURAL ZIMBABWE Benedikt Christ 1 , Janneke van Dijk 2 , Wesley R. Mukondwa 2 , Cordelia Kunzekwenyika 2 , Ronald Manhibi 2 , David Tasunga 3 , Frédérique Chammartin 1 , Matthias Egger 4 , Marie Ballif 1 , for the International epidemiology Databases to Evaluate AIDS (IeDEA) 1 Institute of Social and Preventive Medicine, Bern, Switzerland, 2 SolidarMed, Masvingo, Zimbabwe, 3 Ministry of Health and Child Care, Harare, Zimbabwe, 4 Centre for Infectious Disease Epidemiology and Research, Cape Town, South Africa Background: Differentiated service delivery (DSD) may contribute to reaching the UNAIDS 90-90-90 targets as the number of people living with HIV (PLWH) on antiretroviral therapy (ART) increases. The implementation of differentiated ART delivery is part of the national DSD guideline in Zimbabwe, with the aim to meet the diverse needs of PLWH, to reduce the time spent at health facilities (HFs) and to decongest the health system. Methods: We assessed 26 rural HFs in Bikita District, Zimbabwe, in 2019. At each HF, one or two nurses involved in HIV service delivery, and consecutive PLWH attending the HF on the day of data collection were recruited. We collected data on the availability of various ART delivery models and the time that PLWH spend at the HF using standardized electronic data collection forms. We used descriptive statistics and linear regression analysis on log transformed time data. Results: We assessed 22 rural health centers, 2 rural hospitals and 2 district hospitals. Median numbers of staff and patients registered were 4 and 346 (rural health centers), 13 and 994 (rural hospitals) and 24 and 1152 (district hospitals), respectively. Twenty HFs (77%) had at least one or more differentiated ART delivery model in place. The most common model was the community-based ART refill group (CARG; 13 HFs), followed by facility-based fast track (8 HFs), family refill group (6 HFs) and facility-based club refill (1 HF). Time spent at the HF was assessed for 203 PLWH (68% female, 12% pregnant or breastfeeding, median age 43 years [interquartile range: 34-52]). Fifty-seven (28%) were enrolled in a differentiated ART delivery model (34 in a facility- and 23 in a community-based or family model). Table 1 shows mean times spent at the HFs and results frommultivariable regression. There was no evidence that PLWH enrolled in a facility-based model spent less time than those on standard care, while PLWH in community-based models spent 54%more time at the HF during their visit. Time spent was overall longer at rural health centers compared to hospitals, and shorter at HFs with a patient-to-staff ratio >100. Conclusion: Differentiated ART delivery models are available in most of the assessed HFs in rural Zimbabwe, and a considerable proportion of PLWH on ART are enrolled in a differentiated ART delivery model. However, the type of HF and patient-to-staff ratios were more important determinants of the time spent at the HF than the ART delivery model.

Poster Abstracts

CROI 2020 419