The industry said trust. The data says design.

I have spent years arguing that health equity is a data infrastructure problem.

Not a trust problem. Not a communications problem. Not a problem that gets solved by better leaflets, more diverse stock photography, or a patient advisory board that meets twice a year. A data infrastructure problem. Which means the solution starts with building the infrastructure to collect, analyse, and act on what communities actually say.

Today we published the first evidence from that infrastructure.

The Frequency of Exclusion is the first annual community landscape report from the Equity Engine community. 1,204 people from underrepresented communities across the United Kingdom completed our survey in March 2026. They told us about their experiences of healthcare, their relationship with clinical research, the barriers they face, the organisations they trust, and the conditions that would change their minds about participating in clinical trials.

I want to tell you what the data actually showed, because some of it surprised even me.

The finding I did not expect

The headline numbers are significant. 82% of respondents have felt their health concerns dismissed by a healthcare professional. 83% would be more likely to participate in a clinical trial co-designed with their community. Those findings matter enormously, and I will come back to them.

But the finding that stopped me was neither of those. It was the graduate confusion rate.

45% of all 1,204 respondents were told about a clinical trial by a healthcare professional and did not fully understand what they were being told. That number sits at 68% for people with undergraduate degrees, higher than those with GCSEs, higher than those with no formal qualifications.

The problem is not literacy. The problem is the language of clinical research itself, which has become so specialised that it excludes the very people it is trying to recruit, including educated professionals who encounter it for the first time. This is a communication failure happening at scale, right now, in GP surgeries and hospital outpatient departments across the UK. And it is entirely fixable.

The structural argument

The clinical research industry has spent a decade treating diversity as a trust problem. The framing goes: underrepresented communities distrust research institutions, and the solution is better messaging, more culturally sensitive outreach, improved consent processes.

This data challenges that framing directly.

The two most cited barriers to clinical trial participation in our survey are structural. Too many hospital or clinic visits (628 responses). Trials not designed for someone like the respondent (451 responses). Neither barrier can be addressed by better marketing. Both require redesigning how clinical trials are built and run.

That absence is what Equity Engine exists to address.

The geography of exclusion

One finding in this dataset demands a specific response.

76% of all respondents avoided or delayed seeking healthcare in the past 12 months. That figure is alarming enough. But the East of England sits at 94%, eighteen percentage points above the national average, in the region with the fewest clinical trial sites relative to population and the thinnest community infrastructure. The clinical trial participation pipeline does not exist for these communities. Not because they are unwilling. Because the system was not designed with them in mind.

Improving clinical trial diversity without addressing healthcare avoidance is building on sand.

The commercial signal

83% of respondents would be more likely to participate in a clinical trial co-designed with their community. 45% would be significantly more likely.

This is not a soft preference. It is a quantified behaviour change signal. If co-designing clinical trials with community organisations increases participation likelihood for 83% of underrepresented populations, the return on investment in community engagement is an operational question, and the answer is clear.

The organisations that move earliest on this will have a structural recruitment advantage as regulatory requirements tighten. FDA Diversity Action Plans are already required for most Phase III IND submissions. MHRA guidance is in development. The direction of travel across every major regulator is consistent: community engagement must be evidenced, not assumed.

What comes next

The Frequency of Exclusion is free to download and free to share. It is Vol. 1 of an annual series. Vol. 2 targets 5,000 respondents, a sample large enough to support robust analysis by region, ethnicity, income, condition type, and trust profile.

Every finding in this report will be updated as the community grows. If you want to be part of that, the survey is open at frequencyofexclusion.com.

If you work in clinical operations, medical affairs, regulatory affairs, or patient engagement, and you want to talk about what this data means for your programme, I am genuinely interested in that conversation. You can reach me directly.

The community already told you what needs to change. The evidence is now on the record.

If you want the practical framework for acting on findings like these — spotting where a protocol will struggle to recruit before it locks — our white paper, "Health equity risk in clinical research: how to see patient recruitment failure before it happens", is free to download on the Resources page.