CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Results: PrEP use prevalence was 40% and 31% using SHS 2018 and NHBS 2017, respectively, and neither set of estimates differed significantly by race/ethnicity. Numbers of new HIV diagnoses for Black, Latino, and White MSM were 396, 515, and 234, respectively; HIV diagnosis rates for Black, Latino, and White men ages 13-59 were 105, 77, and 31/100,000, respectively. PEI varied markedly for Black and Latino MSM regardless of approach used (Black MSM: 1.7-3.9; Latino MSM: 2.3-3.3). Targets for Black MSM (range: 59%-100%) and Latino MSM (range: 77%-100%) varied by approach used to define PEI (figure), but would require substantial increases in current PrEP use prevalence to be met (Black MSM: 65-295% increase; Latino MSM: 131-235% increase). Conclusion: Applying a newly developed equity index to set local PrEP targets dramatically illustrates inequity in PrEP coverage for Black and Latino MSM, likely driven by both inequities in PrEP access and the large differential HIV burden in these populations. These findings illustrate the distance needed to travel to move beyond equality towards equity and should motivate intensive efforts to address racial disparities in PrEP scale-up.

2-1-1 PrEP reported missed doses at their last sexual encounter, and none had started postexposure prophylaxis. Conclusion: Our findings suggest that 2-1-1 PrEP is a desirable alternative for many patients. While missed 2-1-1 doses were infrequent in our setting, many individuals changed back to a daily dosing strategy or transitioned between daily and 2-1-1 dosing. With potential scale-up of 2-1-1 PrEP in the U.S., resources to support patients as they transition between dosing regimens are needed, as well as interventions to support the optimal use of 2-1-1 PrEP. 1007 USE OF HIV PREDICTION MODEL TO EVALUATE PrEP COVERAGE IN A LARGE HEALTH CARE SYSTEM Julia L. Marcus 1 , Leo Hurley 2 , Michael J. Silverberg 2 , J. Carlo Hojilla 2 , Jacek Skarbinski 3 , Stacey Alexeeff 2 , Douglas Krakower 4 , Jonathan E. Volk 5 1 Harvard Medical School, Boston, MA, USA, 2 Kaiser Permanente Division of Research, Oakland, CA, USA, 3 Kaiser Permanente Oakland Medical Center, Oakland, CA, USA, 4 Beth Israel Deaconess Medical Center, Boston, MA, USA, 5 Kaiser Permanente San Francisco Medical Center, San Francisco, CA, USA Background: Monitoring progress in scale-up of HIV preexposure prophylaxis (PrEP) requires tools for identifying populations who may benefit from PrEP. Our objective was to evaluate PrEP coverage and disparities in use among people at high risk of HIV acquisition in a large healthcare system, using a validated prediction model to estimate HIV risk. Methods: Our study population was all adult members of Kaiser Permanente Northern California (KPNC) as of January 1, 2018, excluding those with a prior HIV diagnosis as documented in the KPNC HIV registry. Using an HIV risk prediction model we previously developed and validated in our setting, we generated an HIV risk score for each member based on historical electronic health record data. We then used pharmacy fill data to assess recent PrEP use during January 1, 2018─June 30, 2019, and ever PrEP use during all enrollment history, by HIV risk score strata. Among members with very high risk scores (i.e., 3-year risk of incident HIV diagnosis ≥1.0%), we used chi-square tests to compare recent and ever PrEP use by demographic characteristics. Results: Among 3,281,965 members, recent PrEP use ranged from 0.02% to 40.4%, and ever PrEP use from 0.02% to 51.4%, among those with low and very high risk scores, respectively (Table). Of the 8,840 with very high risk scores, mean age was 38 years, 97.7%were male, 19.1%were Black, and 18.6%were Hispanic. Recent PrEP use among those with very high risk scores was higher among males than females (41.2% vs. 7.3%), higher among those aged 30-49 than 18-29 (44.8% vs. 33.8%), higher among those in the highest quintile of neighborhood-level socioeconomic status compared with the lowest (45.3% vs. 32.9%), and higher among Asian (50.8%), White (47.9%), and Hispanic members (42.4%) than Black members (14.1%; P<0.001 for all comparisons). Demographic differences were similar for ever PrEP use. Conclusion: HIV risk prediction models can be used to monitor progress toward PrEP scale-up and equity goals in healthcare settings. Of those identified by our model as being at very high risk of HIV acquisition, nearly 60% had not recently used PrEP and there were substantial disparities in use. Efforts are needed to increase PrEP uptake in insured populations, particularly among females, younger age groups, those with lower socioeconomic status, and Black individuals.

Poster Abstracts

1006 IMPLEMENTATION OF ON-DEMAND PrEP IN A LARGE INTEGRATED HEALTH CARE SYSTEM J. Carlo Hojilla 1 , Julia L.Marcus 2 , Rachel Herbers 3 , Charles B. Hare 3 , Leo Hurley 4 , Michael J. Silverberg 4 , Derek Satre 5 , Jonathan E.Volk 3 1 Kaiser Permanente Division of Research and University of California, San Francisco, CA, Oakland, CA, USA, 2 Harvard Pilgrim Health Care Institute, Boston, MA, USA, 3 Kaiser Permanente San Francisco Medical Center, San Francisco, CA, USA, 4 Kaiser Permanente Division of Research, Oakland, CA, USA, 5 University of California San Francisco, San Francisco, CA, USA Background: Data describing real-world implementation of on-demand (2-1-1) HIV preexposure prophylaxis (PrEP) are limited. In this study, we report on the experiences of early 2-1-1 adopters in Kaiser Permanente San Francisco (KPSF). Methods: KPSF started offering 2-1-1 PrEP in February, 2019. We abstracted data from the electronic health record, including demographics, reasons for selecting 2-1-1, and self-reported challenges, adherence, and persistence. These were collected by clinicians at baseline and at follow-up visits using standardized notes. We examined data descriptively and, for those with 3-month follow-up data, we used Fisher’s exact tests to compare demographics between those who did and did not continue using 2-1-1 PrEP. Results: As of August 2019, there were 2338 active PrEP patients in KPSF, with 251 (11%) using 2-1-1 PrEP. At baseline, median age of individuals using 2-1-1 was 43 (range 18-78); most were white (57%), MSM (99%), and had previously used daily PrEP (76%). In total, 179 patients (71%) had 3-month follow-up data available at the time of analysis. Of these, 23 patients (13%) reported challenges with using 2-1-1 PrEP, including difficulty planning sex in advance, adherence to the dosing schedule, and side effects. Ninety patients (50%) used 2-1-1 PrEP as their sole dosing regimen in the last 3 months; 35 (20%) opted to stay on daily PrEP despite their initial interest in 2-1-1 dosing; 34 (19%) used a combination of 2-1-1 and daily dosing; 4 (2%) used a different intermittent dosing regimen; and 16 (9%) had discontinued PrEP, primarily because of loss of insurance or change in sexual risk. We found no differences in use of 2-1-1 PrEP at the 3-month follow-up by age or race/ethnicity (P>0.05). Of the 90 who reported using only 2-1-1 PrEP at the 3-month follow-up, the majority (90%) cited infrequent sex as their reason for opting against a daily regimen. Only 3 (3%) individuals using

1008 POPULATION-BASED ESTIMATES OF PrEP ACCESS-TO-NEED IN OREGON, 2012-2016

TimW. Menza 1 , Jeff Capizzi 1 , Lea Bush 1 1 Oregon Health Authority, Portland, OR, USA

Background: PrEP is an important HIV prevention modality. Population-based metrics of PrEP uptake and access are critical to the evaluation of public health efforts to increase PrEP use. Methods: Using the Oregon All Payers All Claims administrative dataset, we determined the number of unique individuals at least 16 years of age starting

CROI 2020 378