Inclusion of under served groups in trials

An audit at a UK primary care clinical trials unit

Rebekah Burrow is leading this research.

While working with a colleague, Melanie Carr, on ways to find more data for a clinical trial, we discovered a shared interest in making clinical trials more inclusive.

What is the problem?

“Under-served groups” are groups of people whom research and healthcare systems do not serve well enough. Often people in under-served groups are excluded from research. At the same time, people in under-served groups often need more, or different, healthcare. A few examples of under-served groups include people:

• under 18 years old
• from minority ethnic groups
• with less access to money, education, social support, and healthcare (“low socio-economic status”)

Clinical trials are a type of research. Clinical trials need to study people (participants) who are representative of real patients. When they do, we can work out how safe and effective the interventions (such as a medicine or a test) being trialled are, for real patients. Different groups of people (for example, men and women) sometimes respond differently to interventions. If doctors do not know how well an intervention works for an under-served group (for example, pregnant women and people) they may not offer it to them. People in under-served groups sometimes become more unwell than necessary.

Public organisations in the UK who spend money on research tell researchers to make their research inclusive of under-served groups. This is also an ethical principle that clinical trialists (researchers who carry out clinical trials) promise to uphold.

However, we do not know much about how inclusive clinical trials are. We also do not know much about how representative the participants are.

What research did we do?

This research is an audit. An audit is an independent examination of what has been done. This audit measures how inclusive trials are, in the Primary Care Clinical Trials Unit. Lucy Goddard joined the team to help us do this work. This is a Clinical Trials Unit at the University of Oxford, in the Department of Primary Care Health Sciences. There are about 50 Clinical Trial Units in universities and NHS organisations in the UK. In the audit we looked at whether:

• researchers deliberately excluded under-served groups
• researchers told potential participants who was allowed to take part
• researchers collected data on under-served groups who did take part

What did we find out?

We audited 19 clinical trials. The clinical trials looked at lots of different health conditions, including COVID-19, diabetes, urinary tract infections, and high blood pressure. The clinical trials were trialling interventions like medicines and tests.

All the clinical trials excluded some under-served groups. Trials often excluded people because of their:

• age
• sex or gender
• spoken language (ususally requiring English)
• capacity to consent for themselves (for example, or example, if someone is too unwell to understand the trial or decide whether to take part; sometimes a family member can consent for them)
• pregnancy
• being too ill for the trial (for example, having very severe COVID-19)
• having multiple health conditions (such as having a mental health condition alongside the condition being studied)

About half the time researchers told potential participants who was allowed to take part in “patient information leaflets”. Researchers were most likely to explain age and sex criteria.

Researchers often collected data about some under-served groups that could take part. These included participants’:

• age
• sex
• ethnicity

Researchers very rarely collected data about most under-served groups that could take part, including participants:

• socio-economic status
• disabilities
• pregnancy

What does this mean?

Many clinical trials were designed to carry out research to look after one under-served group (for example, people over 75 years old). Often this meant they had to exclude other under-served groups (for example, pregnant women and people). Some trials made special efforts to include other under-served groups, such as using simple language and pictures to explain the trial.

Some clinical trials excluded under-served groups when researchers could have found ways to include them (for example, people who do not speak English fluently).

These unnecessary exclusions are likely to have made trial participants less representative of real patients.

All trials could have collected more useful data to understand how well their participants represented real patients.

What are we doing about it?

The Institute for Cancer Research Clinical Trials and Statistics Unit have already carried out a very similar audit.

We have also:

• told the Primary Care Clinical Trials Unit about their results.
• published our work for other researchers to read
• asked other Clinical Trial Units to carry out the audit, with eight starting or completing the work
• set up a team in our department to focus on making our research more inclusive
• decided which under-served groups we are going to do better for, first
• started working out how to collect more useful data
• invited other Clinical Trials Units to review results together to identify patterns and learn from each other

We will apply for funding to repeat the audit together in a few years’ time to see what has changed, what has not, and how we can emphasise improving both clinical trials and the evidence they produce.

Patient and Public Involvement (PPI)

We talked to women with experience of pregnancy. We asked about their experiences while pregnant, including their access to treatments, information about treatments, and participation in research. We also asked about their opinions of, and interest in, research to increase inclusion in research during pregnancy. This webpage was developed with support from PPI contributors.