CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

available, 87%were VS. Being >25 years old, employed or married predicted VS. Although not statistically significant, women who attended ≥2 clinics were slightly less likely to have VS (aRR 0.92 95% CI 0.82-1.03) and the distance they moved after transfer was not associated with VS. Conclusion: Some women never linked to care after leaving the integrated clinic. Those who did spread to a large number of different facilities and a quarter were not retained in care. Younger age was a shared risk factor for non-retention, raised viral load and mobility. HIV programs should facilitate clinic transfers when needed and targeted interventions supporting young postpartumwomen warrant consideration.

1089 LINKAGE TO CARE AFTER HIV TESTING IN THE COMMUNITY IN A HIGH HIV PREVALENCE SETTING Sarah Jane Steele , Bernadette Gergonne, Liesbet Ohler, Ellie Ford-Kamara, Gilles van Cutsem, Amir Shroufi MSF, Cape Town, South Africa Background: In 2012, Médecins Sans Frontières, in partnership with the Department of Health (DOH), implemented community testing using mobile, fixed site (M&FS) and door to door (D2D) models, in a catchment area covered by 9 MSF-supported DOH health facilities (HF) in KwaZulu-Natal, South Africa (2016 population 127,611). We assess linkage to care at an MSF-supported HF among people who test positive in the community. Methods: Routine data were analysed retrospectively (Jan 2013 – May 2017). Linkage was ascertained using two methods 1) All tests in the community were collected electronically and matched to facility-based data on individuals in HIV care (Tier.Net) using an algorithm based on name and date of birth; 2) referral slips provided after a positive HIV test in the community, and presented at a facility were encoded in Tier.Net. Time to linkage was available for individuals linked to care through matching, assessed by testing model and CD4 using (Interquartile range: 21.9-35.8). 695 (12%) were known to be in HIV care and were excluded from further analyses. Of those not previously in care, 2493 (47%, 2493/5309) were identified as ever linked: 966 (18%) by referral slips only, 924 (17%) by merging only and 603 (11%) by both. Among those linked to care and for whom time to linkage was available (1507), 10.8% (95%CI: 9.4%- 12.5%) linked to care on the same day as the community test, 42.3% (95%CI: 40.4-52.4%) within 1 week and 77.5% (95%CI: 75.4%-79.6%) within 6 months. Applying this to the 2493 known to be linked, we estimate overall linkage at 6 months to be 36% (1932/5309). Those with CD4<200 were more likely to link to care on the same day, as were those testing at M&FS but after six months of follow-up the differences no longer persisted (Table 1). Among linked, no differences in time to linkage were observed by sex or age. Conclusion: We found that only half of those testing in the community link to care, however, this is likely to be under ascertained. Males who linked to care were similar to females, but males were under-represented among those testing positive. A high proportion of those who link to care do so early. Healthy individuals testing in the community were just as likely to access care at 6 months as those who were less healthy. Kaplan-Meir estimates and applied to the overall number linked. Results: 6,005 individuals (67%, 4027 female) tested positive in the community. Median age among those testing positive was 27.3 years

Poster Abstracts

1088 POSTPARTUM MOBILITY AND TRANSFER OF CARE AMONG HIV+WOMEN ON ART IN SOUTH AFRICA Tamsin K. Phillips 1 , Kate Clouse 2 , Allison Zerbe 3 , Catherine Orrell 1 , Elaine J. Abrams 3 , Landon Myer 1 1 University of Cape Town, Cape Town, South Africa, 2 Vanderbilt University, Nashville, TN, USA, 3 Columbia University, New York, NY, USA Background: Women initiating antiretroviral therapy (ART) in pregnancy commonly require transfer of care postpartum. Transfer and mobility present a potential challenge to long-term retention in this vulnerable population. Methods: Working with a routine primary care cohort, we used electronic health data (HIV-related laboratory tests, ART dispensing and clinic visits from the National Health Laboratory Services and Western Cape Department of Health) to asses HIV care access and mobility in women who initiated ART in an integrated antenatal-ART service in Cape Town, March 2013-June 2014. Transfer out of the integrated clinic was required by all women postpartum. We investigated any linkage to care after leaving the integrated clinic and calculated the number of new clinics attended up to 30m on ART. Among women who did link, we used Poisson regression to explore predictors of i) retention: accessing care at least once at both 6-18m and 18-30m on ART, and ii) viral suppression (VS): HIV viral load ≤1000 copies/ml (>12m on ART). Results: Among 617 women, HIV care was accessed at 98 different facilities, with 11% of women moving out of Cape Town (Figure). Overall, 59%were retained; 21% never linked to care and 20%were lost after linking to a new clinic. Among 485 women who linked to care, 21% attended ≥2 (max 3) clinics. Women ≤25 years old or unemployed were more likely to attend ≥2 clinics (adjusted risk ratio [aRR] 1.10 95% confidence interval [CI] 1.02-1.18 and aRR 1.06 95% CI 0.99-1.12, respectively). Evidence of retention was found for 75% of women who linked (n=363). Those ≤25 years old or reporting unplanned pregnancies were less likely to be retained (aRR 0.87 95% CI 0.76-0.99 and aRR 0.86 95% CI 0.78-0.95, respectively). Among 338 retained women with viral load

CROI 2018 418