52. PrEP Adherence and Discontinuation at a Pharmacy-Supported PrEP Program in Atlanta, GA

Abstract Background Pre-exposure prophylaxis (PrEP) is a highly effective biomedical strategy to decrease Human Immunodeficiency Virus (HIV) acquisition. Effectiveness of oral PrEP is linked to medication adherence. In 2018, Grady Health System (GHS) launched a PrEP program to increase PrEP access among un- and underinsured individuals living in metro Atlanta, Georgia. The purpose of this study is to determine PrEP medication adherence, PrEP discontinuation rates, and associated individual factors of patients enrolled during the first 18 months of the program’s implementation. Methods A single-center, retrospective chart review was conducted on patients enrolled in the GHS PrEP program between June 1, 2018 and February 29, 2020 who received more than one monthly PrEP prescription. Adherence was estimated using the medication possession ratio (MPR). The primary outcome was mean adherence to PrEP. Secondary outcomes include rate of high percent adherence (MPR > 80%), median time of engagement in care, PrEP discontinuation rates, rates of PrEP re-engagement, and individual factors associated with PrEP discontinuation and low adherence. Results This study included 154 patients, 70.8% of them were Black, 62.3% were cisgender men, 59.1% were uninsured, and the mean age was 34. The majority of patients identified as men who have sex with men (51.9%). Mean PrEP adherence was 89.2%; 77.3% of patients demonstrated a high rate of adherence. No individual or social factors were associated with low adherence, but younger age was associated with higher rates of PrEP discontinuation (p< 0.0061). At the end of the follow up period on October 30, 2020, 53.8% of patients were active in the program and 12.7% of those who discontinued had re-engaged with the program. The average length of program engagement was 9.8 months. Table 1. Baseline socio-demographic characteristics (N=154) Table 2. PrEP Adherence and Discontinuation at the GHS PrEP Program from 2018 to 2020 (N=154) Table 4. Multivariate analysis of individual factors associated with PrEP discontinuation and low adherence Conclusion Mean PrEP adherence at a safety net PrEP program in Atlanta was high and PrEP discontinuation rates were comparable to other PrEP clinics nationwide. We found no association with individual factors previously linked to lower adherence, including Black race, younger age, and insurance status. Program-related factors that may have impacted these findings need to be investigated. Other future areas of research include strategies to optimize engagement in care in younger patients. Disclosures Bradley L. Smith, Pharm.D., AAHIVP, Gilead Sciences, Inc (Advisor or Review Panel member)


PrEP Adherence and Discontinuation at a Pharmacy-Supported PrEP Program in Atlanta, GA
Hiba Yacout, PharmD 1 ; Bradley L. Smith, Pharm.D., AAHIVP 1 ; Shelbie Foster, PharmD 1 ; Meredith Lora, MD 2 ; Laris Niles-Carnes, MPH 1 ; Ziduo Zheng, MS 2 ; Suprateek Kundu, PhD 2 ; Judah K. Gruen, MD 2 ; Valeria D. Cantos, MD 2 ; 1 Grady Health System, Macon, Georgia; 2 Emory University, Atlanta, Georgia Session: O-11. Disparities in HIV PrEP and Continum of HIV Care Background. Pre-exposure prophylaxis (PrEP) is a highly effective biomedical strategy to decrease Human Immunodeficiency Virus (HIV) acquisition. Effectiveness of oral PrEP is linked to medication adherence. In 2018, Grady Health System (GHS) launched a PrEP program to increase PrEP access among un-and underinsured individuals living in metro Atlanta, Georgia. The purpose of this study is to determine PrEP medication adherence, PrEP discontinuation rates, and associated individual factors of patients enrolled during the first 18 months of the program's implementation.
Methods. A single-center, retrospective chart review was conducted on patients enrolled in the GHS PrEP program between June 1, 2018 and February 29, 2020 who received more than one monthly PrEP prescription. Adherence was estimated using the medication possession ratio (MPR). The primary outcome was mean adherence to PrEP. Secondary outcomes include rate of high percent adherence (MPR > 80%), median time of engagement in care, PrEP discontinuation rates, rates of PrEP re-engagement, and individual factors associated with PrEP discontinuation and low adherence.
Results. This study included 154 patients, 70.8% of them were Black, 62.3% were cisgender men, 59.1% were uninsured, and the mean age was 34. The majority of patients identified as men who have sex with men (51.9%). Mean PrEP adherence was 89.2%; 77.3% of patients demonstrated a high rate of adherence. No individual or social factors were associated with low adherence, but younger age was associated with higher rates of PrEP discontinuation (p< 0.0061). At the end of the follow up period on October 30, 2020, 53.8% of patients were active in the program and 12.7% of those who discontinued had re-engaged with the program. The average length of program engagement was 9.8 months.

Conclusion.
Mean PrEP adherence at a safety net PrEP program in Atlanta was high and PrEP discontinuation rates were comparable to other PrEP clinics nationwide. We found no association with individual factors previously linked to lower adherence, including Black race, younger age, and insurance status. Program-related factors that may have impacted these findings need to be investigated. Other future areas of research include strategies to optimize engagement in care in younger patients. Background. Since the availability of antiretroviral therapy, mortality rates among people with HIV (PWH) have decreased; however, this trend may fail to quantify premature deaths among PWH. We assessed trends and disparities in all-cause and premature mortality by sex, HIV risk factor, and race, among PWH receiving care at the Vanderbilt Comprehensive Care Clinic from January 1998 -December 2018.
Methods. We examined mortality trends across calendar eras using person-time from clinic entry to date of death or December 31, 2018. We compared mortality rates by demographic and clinical factors and calculated adjusted incidence rate ratios (aIRR) and 95% confidence intervals (CI) using multivariable Poisson regression. For individuals who died, years of potential life lost (YPLL) were obtained from the expected years of life remaining by referencing US sex-specific period life tables at age and year of death; age-adjusted YPLL (aYPLL) rates were also calculated. We examined patient factors associated with YPLL using multivariable linear regression.
Results. Among the 6,531 individuals (51% non-Hispanic [NH] White race, 40% NH Black race, 21% female) included, 956 (14.6%) died. Mortality rates dramatically decreased during the study period (Figure). After adjusting for calendar era, age, injection drug use, hepatitis C virus (HCV), year of HIV diagnosis, history of AIDS-defining illness, CD4 cell count, and HIV RNA at clinic entry, only female sex (aIRR=1.32, 95% CI: 1.13-1.55 vs. males) but not NH Black race (aIRR=1.02, 95% CI: 0.88-1.17 vs. NH White race) was associated with increased mortality. In contrast, aYPLL per 1,000-person years was significantly higher for both female and NH Black PWH (Table 1). In adjusted models including CD4 cell count, HIV RNA, HCV, and year of clinic entry, higher YPLL remained associated with NH Black race, female sex regardless of HIV risk factor, and younger age at HIV diagnosis (Table 2).

Conclusion.
Despite marked improvement over time, sex disparities in mortality as well as sex and race disparities in YPLL remained among PWH in care in this cohort. YPLL is a useful measure for examining persistent gaps in longevity and premature mortality among PWH.
Disclosures. Peter F. Rebeiro, PhD, MHS, Gilead (Other Financial or Material Support, Single Honorarium for an Expert Panel)