FDA Diversity Guidance Drives EHR and AI Integration in Clinical Trials

The U.S. Food and Drug Administration (FDA) issued draft guidance in June 2024, emphasizing the need for "Diversity Action Plans" to ensure clinical trials adequately represent diverse populations. This guidance, affecting Phase 3 studies and others, builds upon a previous draft from 2022 under the Food and Drug Omnibus Reform Act (FDORA).

Systemic Protocol Development

Experts suggest that sponsors will likely need to rethink their protocol development process more systemically to comply with the near-final guidance. For large pharmaceutical companies that rely on global protocol templates with pre-established inclusion and exclusion criteria, this may involve rewriting those templates to adopt a more inclusive approach.

EHR Data and Self-Reported Identity

Justin North, Director of Product Management at TriNetX, highlights the importance of electronic health record (EHR) inputs and artificial intelligence (AI) in future Diversity Action Plan development. While claims data is used to assess potential participant populations, EHR data, especially self-reported race/ethnicity data, is more reliable, particularly for patients seen recently.

"As protocols become more complex with more nuanced exclusion criteria, it can be hard to understand how subtle changes to the protocol might improve representation," North says. "But if you’re working from an EHR source that has high fill rates of self-reported race/ethnicity data, you can tap into that to inform those inclusion decisions."

EHR data can uncover various factors beyond race and ethnicity, including comorbidities, sexual orientation, and pregnancy status, all of which are included in the FDA’s guidance, which broadened its scope from "underrepresented racial and ethnic populations" to simply "underrepresented populations."

AI's Role in Protocol Decision-Making

AI can analyze thousands of protocol permutations to identify criteria that significantly impact inclusion. North emphasizes that AI should enable, not dictate, decisions.

"For a single human to try to evaluate all the complexities of a protocol to understand how it’s impacting inclusivity, that’s just not realistic… and it’s going to remain a huge burden across the industry," he says. "What AI can do for us is run thousands of permutations of a protocol at once and identify which criteria could have the most impact on inclusion."

Individualized Approaches to Inclusivity

Diversity Action Plans should be individualized, recognizing that "diversity" is not a uniform benchmark. North notes the importance of understanding inclusive protocols for specific diseases.

"You need to understand not just ‘What is an inclusive protocol overall?’ but rather, ‘What is an inclusive protocol for this disease?’" he says. "Sponsors will need to establish more epidemiological benchmarks for specific indications to define the representative population according to the need—because that definition isn’t going to be the same for every therapeutic area or indication."

Rare diseases, which may disproportionately affect certain populations, require a reflective approach to diversity planning. The FDA allows waivers of the Diversity Action Plan requirement in limited circumstances.

Resources for Representation

Addressing the historic lack of diversity in clinical research requires systemic changes to study protocols, starting with global templates. Diverse datasets and scalable automation are essential to meet this challenge.