Advances in technology and the adoption of electronic medical records have generated an enormous amount of data about our health, providing researchers and practitioners with more insights into human illness and disease than they ever had before.
As public health scientists gather and probe these data and policymakers rely on them in making important decisions, it is critical that the information being generated and analyzed accurately reflect all individuals to reduce health disparities and create more equitable health systems.
Ensuring data equity is especially important given the increasing use of artificial intelligence (AI) in science and public health. AI is a powerful double-edged sword with the potential to both advance knowledge at lightening speeds and worsen existing disparities if not managed properly.
It is in this context that a global and diverse group of leaders from academia, government, health care, and private industry will gather at Yale on April 8-9 for a conference focused on The Role of Data in Public Health Equity and Innovation. The conference is organized by the Yale School of Public Health (YSPH), with support from the Whitney and Betty Macmillan Center for International and Area Studies.
Our hope is that this event is just the beginning of a longer trajectory of research and work with partners. We also hope to produce peer-reviewed papers from the roundtable discussions.
Kaakpema "KP" Yelpaala
The event is structured to go beyond knowledge-sharing and discussions of central themes surrounding data equity. Attendees will also be encouraged to work together to identify innovative approaches and potential solutions to the problem. The April 8 discussions are open to the Yale community; the April 9 meetings will be closed-door sessions.
“Through this conference, we are focusing on three thematic areas: 1) the depiction of social determinants of health in data; 2) community-based models for data; and 3) the effect of artificial intelligence on health data equity. We will also convene round table discussions in each thematic area to start to chart a collaboration-oriented path forward on the themes discussed,” said conference organizer Kaakpema “KP” Yelpaala, MPH ’06 (global health), senior fellow in public health (health policy) at YSPH and faculty director of InnovateHealth Yale (IHY).
“Our hope is that this event is just the beginning of a longer trajectory of research and work with partners. We also hope to produce peer-reviewed papers from the roundtable discussions.”
There will be two keynote addresses. One will be delivered by Bola Owolabi, director of the National Healthcare Inequalities Improvement Programme at the National Health Service (NHS) England. She will discuss the NHS England’s health equity strategy and the role of data in supporting national and local priorities.
A second keynote address will be given by Munmun De Choudry, associate professor at the School of Interactive Computing at the Georgia Institute of Technology, where she directs the Social Dynamics and Well-Being Laboratory. Her research lies at the intersection of computer science and social sciences, focusing on leveraging large-scale data analytics and machine learning to understand human behavior and mental health dynamics in online environments. She is renowned for her groundbreaking contributions to the fields of computational social science, human-computer interaction, and digital mental health, and will be discussing themes related to AI and health for the event.
More information about conference discussion panels and speakers can be found on the conference’s web page.
YSPH Dean Megan L. Ranney, MD, has identified data-driven leadership as one of the four pillars of the future of public health, along with inclusivity, innovation and entrepreneurship, and communication. Data equity – the process of ensuring that data is representative, accessible, and contextualized for all people – is an integral part of that mission.
“One cannot close health disparities without thoughtful engagement on gathering and analyzing data over time and how we measure differences between groups,” Yelpaala said. “It is for this reason that I deeply believe in Dean Ranney’s commitment to the role of data in public health leadership. From a health equity perspective, we have to consider how our policies and solutions are closing disparity gaps, neutral in their impact, or potentially deepening health inequities for various populations.”
Yelpaala is particularly concerned about the rising use of artificial intelligence in science and public health. He fears that existing health inequities may be exacerbated if AI algorithms have internal biases or if they are applied to non-representative datasets.
“All human beings have different biases, and we produce AI-driven solutions,” he explained. “In my view, there is a risk of replacing one form algorithmic bias with another if we are not clear about the deeper definitions and concepts of bias, fairness, equity, and ethics that are being used. I am happy to see the recent guidance on AI in health from the World Health Organization (WHO), but we still have much work to do.” He added that these concepts vary from group to group and country to country, so it is not a straightforward issue for global alignment, even with guidance from the WHO.
Equitable data comes in many forms. The World Economic Forum recently identified four classes of data equity:
- Representation equity, which seeks to enhance the visibility of historically marginalized groups within datasets while also accounting for data relevancy for the target populations.
- Feature equity, which seeks to ensure the accurate portrayal of individuals, groups, and communities represented by data, necessitating the inclusion of attributes such as race, gender, location, and income alongside other data.
- Access equity, which focuses on the equitable accessibility of data and tools across varying levels of expertise, with an emphasis on addressing transparency and visibility issues related to model construction and data sources.
- Outcome equity, which pertains to impartiality and fairness in results. Beyond developing unbiased models, it entails maintaining vigilance over unintended consequences that impact individuals or groups.
As part of the April 8 program, Yelpaala also will conduct a fireside chat, “Comparative Themes on Data, Health Policy and Health Equity in the U.S. and UK,” with two internationally renowned public health experts: Owolabi and Dr. Marcella Nunez-Smith, MD, MHS ’06, the inaugural associate dean for health equity research at the Yale School of Medicine (YSM), founding director of Yale’s Equity Research and Innovation Center (ERIC), and director of Yale’s Center for Community Engagement and Health Equity. She also served as a senior advisor to the White House COVID-19 Response and chair of the Presidential COVID-19 Health Equity Task Force at the Department of Health and Human Services.
Owolabi will share how NHS England has approached health equity issues through its Core20PLUS5 model. She will also discuss how the health care inequalities improvement program approaches data as a way to not only shape health equity strategies, but also track progress. In addition to her other responsibilities, Nunez-Smith is Yale’s C.H. Long Professor of Internal Medicine, as well as a professor of epidemiology (chronic disease), of public health (social and behavioral sciences), and of management. She has done extensive research on promoting health and health care equity for structurally marginalized populations. Her work emphasizes the importance of community engagement, supporting health care workforce diversity, and developing patient-reported measurements of health care quality. She will discuss health equity in the very challenging, highly complex, and sometimes deeply fragmented U.S. health system.
Yelpaala said he is “very enthusiastic” about bringing two international leaders in health equity together to share their insights and perspectives, and the other panels and keynote address during the conference.
“I am particularly keen for the insights that will build from their exchange of ideas,” he said. “Every health system has strengths and areas of challenge. I am looking forward to Marcella and Bola engaging on the similarities and differences between their contexts, including policy opportunities, data gaps, and how they see the future of closing health disparities in each country.”