The Journey and the Destination: Process Measures for Diversity Action Plans

Yesterday was the last day to submit comments on the Food and Drug Administration’s (FDA) most recent draft industry guidance on diversity in clinical trials, Diversity Action Plans to Improve Enrollment of Participants from Underrepresented Populations in Clinical Studies (link). Here at CCS, we took this as an opportunity to highlight the need for process metrics for diversity in clinical trials. Underrepresentation in trials is a product of social systems that shape how trials are conducted and, in turn, who participates. You can read our submission below. If you want to learn more about how we can incorporate patient-centered process metrics into trials, please check-out our recent commentary in JAMA.

---

September 23, 2024

Comments on Docket #FDA-2021-D-0789: Diversity Action Plans to Improve Enrollment of Participants from Underrepresented Racial and Ethnic Populations in Clinical Trials – Draft Guidance for Industry

Dear FDA Review Committee,  

Core Clinical Sciences (CCS) appreciates the opportunity to provide comments on the draft guidance for industry, “Diversity Plans to Improve Enrollment of Participants from Underrepresented Racial and Ethnic Populations in Clinical Trials”. CCS is a boutique research consultancy that provides statistical, patient-engagement, and other methodological support to life sciences companies, academics, and non-profit organizations. We specialize in innovative and patient-centered clinical trial design and analysis.

We applaud the FDA for taking these important steps forward to improve representation in clinical trials. Based on this draft guidance, it appears likely that enrolment quotas based on race and ethnicity will be used as key indicators to track progress and outcomes of diversity plans. We agree that this is a good starting point, but these metrics need to be complemented with process changes and accountability throughout trial conduct. Focusing on enrolment targets puts current diversity plan initiatives at risk leading only to superficial change. These risks are particularly potent in a free-market economy where industry trial sponsors and contract research organisations are incentivized to find the most cost-effective means to reach enrolment quotas. These means may preclude the necessary process changes to meet the broader goals of this initiative, such as building trust in medical research and institutions, and promoting fairness.

Underrepresentation of racial and ethnic populations does not exist in a vacuum. Instead, it is the consequence of inequitable societal structures that have been built overtime. Meaningful change in trials will require changing how these structures work. Enrolment targets do not directly impact the processes that shape these social structures. It is possible for selective and exclusionary recruitment processes to persist even when targets have been met. For example, all participants of a given ethnicity or race may be recruited from a single site. In addition to indicating limited engagement with the broader community, this also introduces analytical challenges due to confounding of clinical site and the given race or ethnicity. To avoid this and improve trial outcomes (e.g., who participates) and processes (e.g., how they are recruited and retained), we urge the FDA to consider developing process measures that can be implemented now and continue throughout trial activity. 

To facilitate adoption of these process measures, existing risk-based monitoring infrastructure could be adapted. Risk-based monitoring is a continuous, real-time process for monitoring key indicators, such as safety outcomes. Because risk-based monitoring is routinely deployed at the site-, rather than trial-, level, it can identify risks before they become serious problems that threaten participant safety and data integrity. RBM could be adapted to support diversity initiatives using a blend of top-down and bottom-up engagement. For example, to improve accountability for representation in trials, risk-based monitoring could be used to track enrolment targets at individual sites. Key process indicators could be co-developed by sponsors and regulators to maximise cross-trial comparability, and by patients and other stakeholders to ensure that what matters most in terms of process is being prioritised. We provide further details on how these existing measures could be harnessed to support meaningful improvements in clinical trials operations and, ultimately, health equity throughout clinical development in our recent publication in JAMA.(1)

In sum, we propose the following:

1. Develop process measurements that can be implemented across trials for comparability

2. Require trial specific process measurements, co-developed with members of under-represented communities, as a component of diversity plans for clinical trials

3. Mandate reporting of site-level enrolment targets to mitigate against siloed recruitment strategies and reduce confounding between site and race or ethnicity

4. Encourage continuous monitoring of process and outcome measures related to representation in clinical trials throughout trial activity

Real change is needed to improve representation in clinical trials. Developing accountability mechanisms for process changes, as well as outcomes changes, is necessary to motivate and track improvements in the participation of underrepresented communities in clinical research.

Sincerely,

Jay Park, PhD

Scientific Lead and Founder

Core Clinical Sciences Inc. 

Rebecca Metcalfe, PhD

Senior Scientist, Advanced Epidemiology and Patient Centered Research

Core Clinical Sciences Inc.

References 

1.     Metcalfe RK, Park JJH. Diversity Action Plans in Clinical Trials. JAMA. Published online September 20, 2024. doi:10.1001/jama.2024.16009 (link)