Health Disparities in Delirium

Racial and ethnic minority populations in the US experience greater cumulative disease burden, as well as social and economic barriers, stressors, and limited advocacy/access to culturally informed healthcare. This increased risk burden is expected to be associated with an increased risk for delirium during acute care encounters. Previous studies on health disparity and delirium are limited and report equivocal findings regarding delirium incidence, possibly related to sample bias or non-validated measures. Risk for delirium during acute care in health disparity populations (HDP) that include Black African Americans (BAA) and Hispanic-Latinx (HL) has not been systematically studied using validated measures. We conducted a retrospective analysis utilizing our delirium program (ADAPT) registry that systematically assessed all hospitalized patients through their entire hospital stay for the years 2018-2019 (36K patients, 80% NHW, 11% HL, 9% BAA). The Confusion Assessment Method (CAM and CAM-ICU) and Richmond Agitation Sedation Scale (RASS) were used as screening assessments to identify delirium. We know from previous studies that negative CAM results in our environment have high specificity. The incidence of delirium between populations was compared using a chi-square test. Delirium incidence was higher in HDP (BAA combined with HL) compared to NHW in 71-80yo (16.0% vs 12.6%, p=0.003). Delirium incidence was not different in all other age groups compared; <65yo (p=0.191), 61-70yo (p=0.223), 81-90yo (p=0.644). Understanding the association, or lack thereof, between health disparities, ethnic and race-based risks for delirium is expected to provide important insights into more focused delirium assessment, prevention and mitigation strategies in these populations.

Enhancing Diversity in Oncology Clinical Trials: Findings of a Case Study in Improving Black Patient Representation in a Multiple Myeloma Trial

Background: Black patients are underrepresented in multiple myeloma (MM) clinical trials. Despite the promise of Real-World Data (RWD), little research exists on RWD's usage to address this health disparity. In collaboration with a large pharmaceutical partner, we used RWD from commercial datasets (ConcertAI's Electronic Medical Record and claims datasets) aimed at identifying sites with a large Black patient population. We recommended including these sites in a recent clinical trial of Chimeric Antigen Receptor T cell (CAR-T) therapy for MM patients. Methods: We used the following criteria to identify promising sites: (1) high Black patient density, (2) access to a CAR-T accredited parent organization within 100 miles, (3) a hematologist/oncologist who treats MM patients, and (4) a history of treating MM patients with a Proteasome Inhibitor (PI) and Lenalidomide (Len) in the first line of therapy. For (1), sites were ranked using the lower 95% confidence interval for the percent of Black MM patients at the site. For (4), only sites with at least five MM patients who received PI and Len were included. Our data sources were: ConcertAI's Electronic Medical Record (EMR) and claims datasets to link each patient to a site, and Google maps API to identify the CAR-T center nearest each oncology site. The patients in our data sets were not identifiable, and our research was conducted in compliance with the Health Insurance Portability and Accountability Act. After having identified and filtered promising sites, we curated individual candidates in the order of Black patient density. The purpose of curation was to validate a final list of sites. Results: We identified 17 promising clinical trial sites affiliated with 16 healthcare systems in the mid-west, mid-Atlantic, southeastern, and southwestern regions of the United States (table 1). Our RWD captured an average of 141 MM patients (range: 6-791) who were treated at the 17 sites from 2015-2020. Thirty-nine percent of the patients were Black (range: 13-67%). This percentage was three times the recruitment rate of black patients in MM trials in the US (13%). On average, the sites were 44 miles driving distance (range: 0.8-96 miles) from the closest CAR-T center, had eight hematology/medical oncology specialists on staff (range: 1-17), and had previous interventional trial experience (13 sites had experience with MM trials). All of the identified sites were community-based sites, and none of the sites were previously identified by our pharmaceutical partner. Conclusions: We demonstrated that RWD can be leveraged to identify clinical trial sites with a high potential for Black patient recruitment, thereby addressing a known health disparity problem within multiple myeloma (MM) clinical trials. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.