How Ethical AI Is Advancing HPV Vaccine Uptake and Dental Health Disparities Research

By Lynette Hammond Gerido, Ph.D., M.P.H., M.B.A; Scott Emory Moore, Ph.D., RN, APRN; Hamid Reza Kohan Ghadr, Ph.D.; and Preeti Pushpalata Zanwar, Ph.D., M.P.H., M.S.
Posted Nov. 1, 2024

Clockwise from top left: Drs. Lynette Hammond Gerido, Hamid Reza Kohan Ghadr, Preeti Pushpalata Zanwar, and Scott Emory Moore

Addressing health disparities requires innovative approaches that blend expertise across disciplines. Through the Science Collaborative for Health Disparities and Artificial Intelligence Bias Reduction (ScHARe), NIH’s collaborative, cloud-based platform for population sciences research, interdisciplinary research teams leverage advanced data science methodologies and artificial intelligence (AI) to uncover how social determinants of health contribute to health inequities and to identify opportunities for data-driven policy interventions.

Each team is challenged to use best practices to make their analysis explainable and to mitigate the “black box of AI,” such as through:

• Model cards
• Concise documents that summarize how an AI model works
• Its intended use
• Possible limitations

These teams, co-led by experts in data science and health disparities research, represent a unique convergence of disciplines focused on solving public health challenges using advanced computational tools.

ScHARe research teams exemplify the future of health disparities research with members from diverse backgrounds—from undergraduate students to senior faculty, as well as scientists and practitioners from academic and industry settings collectively bringing multidisciplinary expertise at varying career levels. These multigenerational cross-functional research teams symbolize ScHARe’s success in fostering a paradigm shift in health disparities research to use big data and to train underrepresented populations, including women, in biomedical data science.

Using Big Data to Understand HPV Vaccine Uptake Disparities

One of the ScHARe teams, co-led by Dr. Lynette Hammond Gerido and Dr. Scott Emory Moore, focuses on disparities in HPV vaccine uptake among marginalized populations. By analyzing datasets such as the National Immunization Surveys, the team has used data-driven techniques to examine the distribution of vaccine access before and after 2020, when the COVID-19 pandemic began.

Their goal is to identify the social determinants influencing disparities in HPV vaccine reach and to better understand the individual factors that may underpin those disparities. The insights generated may inform policy aimed at reducing disparities and promoting equitable access to essential, relatively low-cost, preventive health care services.

Through their advanced analysis, the team aims to uncover patterns of inequality, offering a better understanding of how social, economic, and biological factors interplay to contribute to health care disparities. Their research will ultimately contribute to more informed public health strategies, with the potential to significantly reduce inequities in HPV preventive care and treatment impacting cervical cancer outcomes.

Using Machine Learning to Understand Dental Health Disparities

Another ScHARe team, co-led by Dr. Hamid Reza Kohan Ghadr and Dr. Preeti Pushpalata Zanwar, addresses disparities in dental care access, an issue that worsened during the COVID-19 pandemic. Their research utilizes the Medical Expenditure Panel Survey, a nationally representative dataset that includes information on health care use, costs, and insurance coverage.

Through the application of advanced machine learning techniques, their research aims to identify key determinants of dental care access, such as out-of-pocket costs, preventive health behaviors, and socioeconomic factors. This work is particularly relevant given the pronounced disparities in dental care access in the United States, disproportionately affecting marginalized communities.

In addition to exploring these disparities, the team also is addressing potential biases in AI models and their implications for health equity. Their analysis will contribute to the ongoing discourse on bias mitigation in AI, particularly in the context of health disparities, and they aim to offer methodological guidance on applying machine learning to public health data regarding oral health to inform equitable health care policies.

The ScHARe platform is a critical enabler of these research efforts. It provides access to over 260 health disparities and social determinants of health (SDOH) datasets, along with cloud-based computational tools that support large-scale, interdisciplinary analyses.

The collaborative efforts of these research teams highlight the transformative potential of combining AI, big data, and multidisciplinary expertise to address public health challenges. As their projects progress, they aim to generate evidence that will guide policy and practice, contributing to the broader goal of promoting health equity and reducing disparities across diverse populations.

Lynette Hammond Gerido, Ph.D., M.P.H., M.B.A., is an information scientist and assistant professor in the Department of Bioethics at the Case Western Reserve University School of Medicine. She is also scientific director of the Department of Bioethics’ Center for Community Health ANd Genomic Equity (CHANGE) and a Population and Cancer Prevention Program member at the Case Comprehensive Cancer Center. She partners with communities in her research and uses population data to understand trends in ethical, legal, and social implications of clinical research, public health campaigns, and consumer health technologies.

Scott Emory Moore, Ph.D., RN, APRN, AGPCNP-BC, FAAN, is an assistant professor at the Frances Payne Bolton School of Nursing at Case Western Reserve University. His research centers on the biological, psychological, and social factors that influence health outcomes among LGBTQ+ adults. His nursing experience includes emergency and trauma nursing, acute and chronic stroke care, complex chronic diseases, and care for LGBTQ+ and other marginalized populations.

Hamid Reza Kohan Ghadr, Ph.D., is a molecular and computational biologist and a healthcare data scientist serving as a principal scientific consultant at Akna Health. He previously served as an assistant professor at the Michigan State University College of Human Medicine. He has expertise in AI, cloud computing, and bioinformatics and his research advances biomedical innovation by applying cutting-edge technologies to addressing health care challenges.

Preeti Pushpalata Zanwar, Ph.D., M.P.H., M.S., is a health economist and health services researcher who serves as a lecturer in Applied Health Economics & Outcomes Research and Health Policy at Thomas Jefferson University. She researches disparities in preventive health care access, socioeconomic pathways to cognitive aging, inequities in vaccine uptake, outcomes of viral infections, and modelling differences in health care costs of multimorbidity.