Including a diverse range of people in clinical research is crucial if we are to improve health equity. The need to improve isn’t up for discussion, the problem comes when we try to untangle how to do that effectively, and consequently, how to evaluate if what we’re doing is working. The big question of “How can we measure diversity in clinical research?” is what we’re going to tackle in this post.
Surely, we just need to collect better demographic data?
You may be thinking, “EASY! Collect some demographic data from your patient population!”. Sure, demographics help us to describe our patient populations but what if I told you even then, these data can fall short in helping to understand if a clinical trial is diverse? See, there tends to be a standard set of demographics that are reported in clinical research: age, sex, and ethnicity. Whilst these are important pieces of information to collect, in isolation they may not provide a representative measure of diversity.
Clara Health’s blog post puts this into perspective, emphasising that a multitude of demographic datapoints must be considered to truly understand who is included in clinical research, and whether our populations are as diverse as we would like. Some examples provided include: mobility issues, household size, and employment status. Collecting several pieces of demographic data allows us to get a fuller picture of who our patients are. People are messy and complicated, and it’s important that we allow for our demographic data to show some of that complexity. Reducing the various factors that contribute to a person’s identity to a simple “how long have you lived on planet Earth?” isn’t going to cut it.
What’s everyone else doing?
That probably sounds like a good conclusion to the blog – add more datapoints – but unfortunately, it’s not that simple. If you’re anything like me, you’re now probably wondering what demographic information you should be collecting. Sorry to throw another spanner into the works, but currently a standardised set of measurements doesn’t exist. There is some good news though! Collaborators across the Milken Institute, the Clinical Trials Transformation Initiative, and the Multi-Regional Clinical Trial Center of Brigham and Women’s Hospital and Harvard University recognised the lack of standardisation, and recently published a framework as a call to action: “Toward a National Action Plan for Achieving Diversity in Clinical Trials” (posted May 6, 2024).
A plan to move forward
In this framework, eight domains that must be addressed to achieve diversity, equity and inclusion in clinical research are outlined. Of particular importance to our conversation on demographic data collection is Domain G: Comprehensive and Consistent Data. The goal of Domain G (reported on p.34), is to “Establish a national (and international), interoperable, and accountable system for collection and sharing condition-specific demographic and non-demographic data”. One of the achievable actions that they provide, is establishing terminology formatting for metrics; a shared language for demographic data. Without established terminology, confusion can arise in the context of what a metric is truly measuring and how this compares to other clinical trial data. On the surface this may sound a bit pedantic but in researching and writing this blog post I came across layers of semantics in demographic measures. For example, income and household earnings are likely collecting the same information, but a slight difference in context can limit a researcher’s ability to understand:
- If they are achieving a goal of increasing diversity in their patient population,
- How measures of diversity in their own trial differs from other clinical trials,
- And how we can learn from each other to effectively improve our clinical trial accessibility and inclusivity.
What next?
Over 500 words in, and you may have noticed that I’m yet to give you a definition answer to the question, “How can we measure diversity in clinical research?”. The truth is, there isn’t a definitive answer. We need a collaborative effort between researchers and patient partners to understand what measures of diversity are important to the clinical trial at hand, considering both a patient and condition-specific perspective, even when set standards are published. Standards should be viewed as the minimum with additional data added to support clinical research involving specific groups (e.g., children and young people), and taking place in specific contexts (e.g., during a pandemic).
