CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

1184 Novel Post-Hospitalization Care Model Decreases Mortality in People With HIV in Zambia: Pilot Study Cassidy W Claassen 1 , Chiti Bwalya 2 , Morley Mujansi 3 , Linah Mwango 4 , Kirsten Stoebenau 2 , Caitlin Baumhart 1 , Godfrey Muchanga 3 , Brianna Lindsay 1 , Mundia Mwitumwa 5 , Nyuma Mbewe 6 , Wilbroad Mutale 7 , Michael Vinikoor 8 1 University of Maryland, Baltimore, MD, USA, 2 University of Maryland - College Park, College Park, MD, USA, 3 Maryland Global Initiatives Corporation, Lusaka, Zambia, 4 Ciheb Zambia, Lusaka, Zambia, 5 University Teaching Hospital, Lusaka, Zambia, 6 Zambia National Public Health Institute, Lusaka, Zambia, 7 University of Zambia, Lusaka, Zambia, 8 University of Alabama at Birmingham, Birmingham, AL, USA Background: Despite progress in HIV epidemic control in Zambia, HIV-related mortality remains high. Deaths are often preceded by hospitalization, and post discharge mortality among people with HIV (PLWH) reaches 20-40% within six months due to individual, psychosocial, and systemic factors. We conducted a feasibility study of a community health worker (CHW)-led model to improve patient health outcomes and reduce mortality post-discharge. implementation science framework. Adults hospitalized with HIV in Lusaka and then discharged were enrolled and followed up for 6 months post-discharge. The control group received standard of care (SOC) with telephonic follow-up. The intervention group received a novel care package, based on formative qualitative work, consisting of a discharge summary card, CHW home visits within one week of discharge, and screening and referral for depression and alcohol abuse at 1-3 months post-discharge. A physician-clinical liaison officer team based at the discharging hospital oversaw the CHW home visits, which included psychosocial counseling, vital signs check, medication counseling, and outpatient follow-up. Home visit data were collected by CHWs using electronic devices. Results: Among 124 patients (median age, 41 years; 57.8% women; median CD4, 299 cells/mm 3 ) in the SOC group, 23 (18.6%) died within 6 months of discharge. From 18 August to 20 September 2023, 21 patients enrolled in the pilot intervention group. To date, 13 (61.9%) received at least one home visit (7 of these were within 1 week of discharge) and 5 two visits, 15 (71.4%) received a discharge summary card, and 12 (57.1%) were screened for behavioral health problems. At one month, 12 were alive, 1 (7.7%) had died (from extrapulmonary tuberculosis), and 1 (7.7%) was readmitted based on the findings of the CHW at a home visit. Acceptability among participants and caregivers has been high. Conclusion: A novel discharge model of care, involving enhanced discharge instructions, CHW home visits, and screening and referral for behavioral health problems, appeared to be feasible and acceptable in urban Zambia. Post hospital CHW visits have potential to reduce post-discharge mortality among PLWH in countries with generalized HIV epidemics such as Zambia. Focusing on the peri-discharge period can strengthen health systems as countries move into HIV epidemic control. Methods: A quasi-experimental feasibility and acceptability study was conducted at two tertiary hospitals in Lusaka, Zambia, using the PRISM Jacqueline E Rudolph 1 , Keri Calkins 2 , Corinne E. Joshu 1 , Bryan Lau 1 1 The Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 2 Mathematica, Princeton, NJ, USA Background: Medicaid serves as the largest single source of insurance for people with HIV (PWH) in the US, with approximately 40% of PWH covered by Medicaid. Not only does Medicaid represent a large, diverse population of PWH in US, it also represents some of the most vulnerable PWH, since eligibility is based on low income or disability. While the HIV care cascade is well characterized among all PWH in the US and among those linked to HIV care clinics, less is known about the state of the care cascade among PWH on Medicaid or how the cascade has changed over time. Methods: We analyzed data from 273,799 Medicaid beneficiaries with HIV, enrolled in 14 US states, 2001-2015. All beneficiaries identified as having HIV through claims records are aware of their diagnosis and linked to care; thus, we focused on the later steps of the care cascade. We estimated prevalence of 4 levels of the HIV care cascade: retained in care and adherent to ART; retained but not on ART; not retained but on ART; not retained and not on ART. Beneficiaries were considered retained in care if they had an office visit, viral load measurement, or CD4 cell count every 6 months. Adherence to ART in each month of follow-up was defined as having a medication possession ratio of at least 80%. Prevalence of each state in each month 2001-2015 was estimated using a non-parametric multi-state approach, accounting for death as a 1185 The HIV Care Cascade in Medicaid, 2001-2015

competing event and for loss of Medicaid coverage using inverse probability of censoring weights. Analyses were conducted overall and by US state. Results: As shown in the Figure, 20% of beneficiaries with HIV were retained and ART adherent in 2001; this proportion reached a peak of 36% in 2013. The proportion not retained in care but ART adherent did not meaningfully change across follow-up (5-8%). In contrast, the proportion retained in care but not ART adherent decreased from 53% to 32%, and the proportion not retained in care and not ART adherent decreased from 20% to 13%. Death remained an important competing event in this era, with a cumulative incidence of 27% by 2015. Results differed by US state. Conclusion: Despite being linked to care, less than half of beneficiaries with HIV were classified as ART adherent across all of follow-up, likely indicating that many Medicaid beneficiaries with HIV were not virally suppressed during this time period. These findings were seen even in the post-2012 "Treat All" era. Future work will explore whether HIV care engagement improved between 2015-2021.

Poster Abstracts

1186 Prevalence and Predictors of Advanced Disease Among People Living With HIV in Masaka Region, Uganda Alex Daama 1 , Fred Nalugoda 1 , Asani Kasango 1 , Betty Nantume 1 , Grace N. Kigozi 1 , Robert Ssekubugu 1 , Absalom Ssettuba 1 , Joseph Kagaayi 1 , David Serwadda 1 , Joseph Kabanda 2 , Arthur G. Fitzmaurice 1 , Nelson Sewankambo 1 , Godfrey Kigozi 1 , Gertrude Nakigozi 1 1 Rakai Health Sciences Program, Kalisizo, Uganda, 2 Centers for Disease Control and Prevention, Atlanta, GA, USA Background: A large percentage (22%) of people living with HIV (PLHIV) present for care with advanced HIV disease (AHD), threatening the achievement of the 95-95-95 goals to end AIDS by 2030. For anyone over the age of five, AHD is defined as CD4 count <200 cells/mm 3 or with a current WHO stage 3 or 4 events. This study aimed to examine the prevalence and factors related to AHD among newly diagnosed PLHIV in Masaka Region, Uganda. Methods: A cross-sectional study was conducted from October 2021 through September 2022 among newly diagnosed PLHIV enrolled in care from 12 districts in Masaka Region. Data from electronic medical records (EMR) at health facilities were extracted for analysis. Variables included sex, age, marital status, location of facilities, and points of entry into care. Using a bivariable analysis, we determined the prevalence of AHD. A multivariable modified Poisson regression analysis was used to determine predictors with 95% confidence intervals (CIs). Results: Of 3,452 newly diagnosed PLHIV on ART included in this study, 2,254 (65.3%) were females. The prevalence of AHD was 15% (518). The results from multivariable modified Poisson regression revealed that participants aged 18-35 years had lower risk of AHD compared to those aged 5-17 years (aPR=0.45; 95% CI: 0.27, 0.78). Married individuals were at lower risk of AHD compared to unmarried participants (aPR=0.67; 95% CI: 0.57, 0.79). Male participants (269/1,198, 22.5%) had higher risk of AHD compared to females (249/2,254, 11.1%; aPR=1.85; 95% CI: 1.57, 2.19). Participants receiving ART services from urban facilities had higher risk of AHD compared to participants receiving ART from rural sites (aPR=1.61; 95% CI: 1.35, 1.93). Participants who were enrolled into care through HIV testing service outreaches (aPR=0.72; 95% CI: 0.58, 0.90) had lower risk of AHD while participants from other care points (aPR=1.28; 95% CI: 1.06, 1.53) had higher risk of AHD compared to the general outpatient point. Conclusion: The proportion of AHD in this cohort (15%) was lower than the national proportion of 22%. However, this work can be used to design interventions to address higher AHD prevalence among males, in urban facilities, and at care points other than HIV testing outreach sites and the general outpatient point. The figure, table, or graphic for this abstract has been removed.

CROI 2024 387