An FDA diversity action plan is shifting from voluntary planning to a statutory submission requirement for certain studies. Mandatory submission applies to covered studies only after the FDORA implementation trigger described by FDA: enrollment commencing more than 180 days after publication of final guidance. Under the Food and Drug Omnibus Reform Act of 2022, Congress gave FDA authority to require Diversity Action Plans for covered studies, and FDA's draft guidance explains how sponsors should prepare before that requirement becomes applicable. Covered programs need a plan that matches FDORA, FDA's draft guidance, and the intended-use population.
The shift matters because diversity planning has moved from values language into operational evidence. A credible plan needs numbers, rationale, methods, and accountability. Sponsors that treat it as a template exercise risk delayed feedback, weak enrollment performance, and avoidable questions during review.
What the FDA diversity action plan requirement covers
FDORA amended the Federal Food, Drug, and Cosmetic Act to require diversity action plans for certain clinical studies of drugs, biologics, and devices. FDA's June 2024 draft guidance describes the expected format and content, including enrollment goals, the reasoning behind those goals, and the measures the sponsor will use to meet them.
For drugs and biologics, the requirement generally applies to Phase III studies or other pivotal studies, excluding bioavailability and bioequivalence studies. For devices, FDORA covers devices requiring an investigational device exemption application. Devices not requiring an IDE, except studies described under 21 CFR 812.2(c), must develop a plan for relevant clinical studies, with submission tied to 510(k), De Novo, or PMA applications as applicable. The statute also gives FDA authority to waive the requirement in limited circumstances, but sponsors should treat waiver as an exception rather than a strategy.
The plan is intended to improve enrollment of populations that have historically been underrepresented in clinical research. That includes racial and ethnic groups, sex and gender groups, older adults, and other clinically relevant populations depending on the disease. The plan belongs with the likely users of the product, not with a generic diversity statement.
Why FDORA changed sponsor expectations
FDORA, enacted as part of the Consolidated Appropriations Act, 2023, is the operative legal source for the current Diversity Action Plan requirement. The result is a more concrete version of a message FDA had already been sending through prior draft guidance and trial-diversity initiatives: sponsors should plan for inclusion before enrollment begins and, for covered studies subject to the final-guidance timing, submit or develop the Diversity Action Plan according to the applicable FDORA pathway.
This affects development teams in three ways.
First, diversity goals now belong at the protocol and feasibility stage. If eligibility criteria exclude patients with common comorbidities, renal impairment, travel constraints, or prior therapies, no recruitment vendor can fix the enrollment pool later.
Second, site strategy has to be evidence based. Sponsors need to show why selected geographies and investigators are likely to reach the target population. That connects diversity planning directly to patient recruitment for clinical trials, because the same data that identifies eligible patients also identifies where representative enrollment is realistic.
Third, progress needs to be monitored during the study. If enrollment is skewing away from the plan, the sponsor should know early enough to add sites, adjust outreach, or rework screening workflows.
What belongs in a diversity plan template
A workable diversity plan template is short enough for teams to use and specific enough to withstand regulatory review. Six sections usually carry the weight.
- Study overview: product, indication, study phase, population, geography, and key eligibility criteria.
- Disease epidemiology: prevalence, incidence, severity, and outcome differences by relevant demographic groups.
- Enrollment goals: target percentages or counts by population, with a clear rationale.
- Operational strategy: site mix, recruitment channels, community partnerships, language access, transportation support, and retention methods.
- Monitoring plan: how enrollment will be tracked, how often it will be reviewed, and what triggers corrective action.
- Governance: named owners across clinical operations, biostatistics, regulatory, medical affairs, and patient engagement.
Good plans connect each enrollment goal to a source. A diabetes trial, for example, should tie enrollment goals for Black, Hispanic, or African ancestry participants to disease burden, treatment patterns, and the intended use population.
How FDA diversity requirements affect global trials
FDA diversity requirements clinical trials teams often ask whether non-US sites can help satisfy US diversity expectations. The answer is sometimes yes, but not automatically. A patient enrolled in Kenya, Ghana, Brazil, India, or Poland may add clinically useful population diversity, but the sponsor still has to explain relevance to the US intended-use population and to the disease biology.
Global enrollment can help where disease burden is high outside the traditional North American and Western European site network. Many infectious disease, cardiovascular, oncology, and metabolic indications have major African patient populations that remain underrepresented in registrational evidence. The scientific case is clearest when sponsors can connect the geography to disease prevalence, genetic variation, standard of care, or unmet need.
That is where the broader argument for diversity in clinical trials becomes operational. Representative enrollment reduces uncertainty about how a product performs across the people who will actually use it.
Common mistakes sponsors make
The most common mistake is writing the plan after the protocol is already locked. If inclusion and exclusion criteria have already narrowed the eligible population, the plan becomes a justification document rather than an operating plan.
The second mistake is using national census demographics instead of disease epidemiology. FDA's direction is clear: enrollment goals should reflect the population affected by the disease or condition, not a generic population benchmark.
The third mistake is vague recruitment language. "Community outreach" is not a strategy unless the plan names partners, locations, materials, languages, staff responsibilities, and timelines. "Digital recruitment" is not a strategy unless the plan explains how patients will be identified, screened, referred, consented, and retained.
The fourth mistake is failing to budget for the plan. Transportation, translation, decentralized visits, community engagement, and additional sites cost money. If they are not in the budget, they usually do not happen.
Another mistake is separating diversity planning from statistical planning. Enrollment goals need biostatistics input because subgroup analyses, pooling strategies, and interpretability depend on sample size and endpoint design. A plan that names ambitious population targets but cannot explain how those data will be interpreted is incomplete.
Sponsors should also avoid assuming that digital recruitment automatically improves representation. Online advertising can expand reach, but it can also bias enrollment toward patients with smartphones, literacy, private insurance, or proximity to research centres. Digital channels work best when paired with community-based referral, multilingual materials, and site staff who understand local barriers to participation.
Building a defensible operating model
Sponsors should treat the FDA diversity action plan as a living operating model. That model starts with data: disease burden by group, patient concentration by site catchment area, standard-of-care variation, and likely screening failure drivers.
From there, teams can pressure-test each site. Does the site see enough eligible patients? Does it serve the population named in the enrollment goal? Does it have staff who can support consent in relevant languages? Has it enrolled similar patients before? If not, what support would make it credible?
Kapsule sees this issue often in African market-entry and trial-feasibility work. Sponsors may know that a country has high disease burden, but they still need patient-level signals, facility context, and ethics-ready data pathways before they can make a country or site part of a regulated evidence plan.
The operating model needs a review cadence. Many sponsors track overall enrollment weekly but review demographic mix too late. A better approach is to review enrollment against diversity goals from the first month of recruitment, with pre-specified triggers. If a group is under-enrolling, the team should know whether to change referral channels, activate backup sites, add patient navigation, revise outreach materials, or revisit overly restrictive criteria.
Documentation matters as much as action. Sponsors should preserve the rationale and implementation record, because FDORA and FDA guidance focus on goals, rationale, and how the sponsor intends to meet those goals. Meeting minutes, site-level dashboards, recruitment vendor reports, and protocol-deviation reviews can all support that record.
This evidence trail also helps internal governance. Senior leaders can see whether the team invested early enough, whether assumptions were realistic, and whether the same mistakes are recurring across programmes.
Practical takeaways for sponsors
Start the diversity plan before protocol finalization. Use epidemiology to define goals. Test eligibility criteria against real patients. Select sites with data, not only reputation. Track enrollment by demographic group from first patient in. Build corrective action into the plan instead of treating it as a rescue move.
The best FDA diversity action plan helps the trial team make better decisions early enough to change the result.
Kapsule provides access to structured, de-identified health records covering over 75 million patients across 14 African countries. Contact our team to discuss how African patient data can support diversity planning, country selection, and recruitment feasibility.
This article is intended for informational purposes only and does not constitute legal, medical, or regulatory advice. Readers should obtain independent professional counsel for their specific circumstances.