Advancing Equity Scholars Program
This initiative is designed to empower early-stage investigators in addressing health disparities. Scholars selected for this program receive funding of up to $30,000 to support 12-month pilot projects, covering expenses related to data collection, research assistance and other activities that bolster larger project proposals. In addition to funding, participants in the Advancing Equity Scholars Program (AESP) benefit from comprehensive support, including guidance from the HEAL-R Expert Bench, access to clinical data and statistical assistance, grant writing consultations, financial administration support and networking opportunities through HEAL-R seminars.
| Project | HEAL-R Scholar | Objectives |
|---|---|---|
| Understanding the Determinants of Maternal Morbidity and Mortality | Felix Pabon-Rodriguez, PhD | Disparities in severe maternal morbidity (SMM) and maternal mortality rates (MMR) disproportionately affect racial and ethnic minorities, driven by complex social, biological, and systemic factors. To address these inequities, this study leverages electronic health records and public health data to: (1) assess the impact of socioeconomic and biological factors, such as income, education, and chronic conditions, on SMM and MMR across diverse populations; (2) evaluate how healthcare system infrastructure and policies influence disparities in maternal health outcomes; and (3) identify and test integrated intervention strategies, including enhanced prenatal care, telehealth, and community health worker programs, to reduce disparities and promote maternal health equity. |
| FoodRx: A Practical Strategy for Integrating Nutrition into Cancer Care |
Rebecca Rivera, PhD, MPH, CPH |
Many cancer survivors face food insecurity, with one in four struggling to access adequate, healthy food compared to one in ten Americans overall. This issue is closely tied to poor cancer outcomes, yet traditional methods for assessing diet quality, an intermediary between food insecurity and health, are too time-consuming for routine clinical use. This project aims to: (1) assess the usability of Diet ID™ among cancer care stakeholders, exploring how providers and patients use the tool’s data and how it fits into clinical workflows; (2) validate the tool’s accuracy and reliability by comparing its results to a national gold standard dietary recall tool, while linking data on food insecurity and financial hardship to electronic health records. This work ultimately seeks to improve cancer care by integrating diet quality assessments into efforts to address health disparities and food insecurity among survivors. |
Health Disparities & Equity and Learning Health System Research Pilot Program Scholars
HEAL-R contributes funds to the Regenstrief Institute’s Health Disparities & Equity and Learning Health System Research Scholars pursuing research on health disparities and equity. This new program supports research aimed at developing, disseminating, or implementing innovative interventions to address health disparities at various levels. Applicants with expertise in clinical research, epidemiology, public health, implementation science, informatics or health services research are encouraged to apply, with the goal of catalyzing future funding applications to the NIH and other external agencies.
| Project | HEAL-R Scholar | Objectives |
|---|---|---|
| A primary care learning health system approach using data and team engagement to address social determinants of health in colorectal cancer screening to improve health equity | David A. Haggstrom, MD, MAS | Colorectal cancer (CRC) is the 2nd leading cause of cancer death in the US. Despite several effective screening modalities to detect and prevent CRC, there are continued racial, ethnic, and socioeconomic disparities in CRC screening, incidence and mortality. Given its whole-person, comprehensive care focus, a high-quality primary care infrastructure is needed to address these disparities, but without proper identification of patients at the greatest risk of not receiving screening and more personalized information on which test is the most likely to be completed, delivery of equitable CRC screening is unlikely to achieved. Furthermore, without input from key stakeholders on usability and workflow integration of this information, such estimates are unlikely to have their intended impact. To address these key issues, we will (Aim 1) develop and validate novel, machine learning predictive models to identify clinical characteristics and social determinants of health that predict patients who are eligible for CRC screening but at high risk of nonadherence and predict which of the various CRC screening modalities they are mostly likely to adhere to. We will then (Aim 2) develop a primary care-based learning health systems intervention informed by data, healthcare teams, and members of the community to increase CRC screening among these identified patients. By leveraging key partnerships with IU Health Primary Care, the Departments of Family and General Internal Medicine, and the central Indiana Patient and Family Advisory Committee, we will be well-positioned at the end of this pilot to gain extramural support to test the efficacy of this intervention. |
|
Enhanced health-related social needs data extraction to support action and impact |
Joshua R. Vest, PhD, MPH, FACMI |
Health-related social needs (HRSNs) may be recorded using structured data elements like problem lists, screening surveys, or diagnosis codes, but they are more commonly recorded in free-text. As such, clinical notes and documents represent a critical source of HRSN information. For this project, we seek to further develop natural language processing (NLP) algorithms to meet health system partners’ demands for better access to HRSN information. Finding Other Risks & Contexts Electronically (FORCE) is a set of NLP algorithms that identify seven HRSNs from clinical notes and documents. FORCE has been published in the literature, encompasses HRSNs required by CMS, and is already being used by multiple Regenstrief investigators. To enhance FORCE to meet user needs and support an NIH grant application, our team from CBMI and HDS will: 1) Map HRSNs extracted from clinical notes to structured data elements; 2) Establish performance of the FORCE NLP algorithm across key INPC patient populations; and 3) Develop and test a data collection instrument to identify end-user requirements for HRSNs extracted from clinical documents. This proposal will enhance the Institute’s and key partners’ ability to turn data into knowledge and action. It will enhance the competitiveness of an NIH proposal. By leveraging the INPC, this proposal is inclusive of socioeconomically disadvantaged and racial minority populations. The project is supported by the Indiana Health Information Exchange and Indiana University Heath. |