Peer review of funding applications for health research in ethnic minority groups must improve

Many ethnic minority groups in the UK have worse overall health outcomes. It is widely agreed that we must do more to improve their health by increasing representation in research and tackling systemic and interpersonal racism in medicine and wider society.1 The short term costs of including minority groups in more research are mitigated by billions in longer term savings associated with better health outcomes, increased workforce participation, and lower medical spending.23 Major UK funders, including the National Institute for Health and Care Research (NIHR) and UK Research and Innovation, seek to increase representation of ethnic minorities in research.45

With this in mind, we should support and encourage quality research proposals centred on the health of people from ethnic minorities, but this does not always happen. Disagreements, sometimes legitimate, are ongoing about how to include these populations in research. In our experience, peer reviewers of funding applications repeatedly ask for further justification of a focus on participants from ethnic minorities—even after the worse health outcomes, evidence gaps, and need for cultural sensitivity are documented.

We experienced this in response to our research, which focused on the largest ethnic minority group in the UK, who had worse health outcomes for the studied health condition and were previously excluded from effectiveness trials to treat it. Despite this, one peer reviewer wrote that they felt we should include all ethnicities in the research and that they did not see a compelling need for the “narrow focus” of the research. This is at odds with the gaps in the evidence base and the need to improve health outcomes of this condition in the studied group. We see this as the academic equivalent of stating that “All Lives Matter” in response to the Black Lives Matter movement, which undermines the point of identifying and focusing on issues that disproportionately harm some ethnic minority groups. Another reviewer asked about what evidence supports the development of interventions for black and other ethnic minority populations. The need for developing specific interventions for ethnic minority communities has been clearly established.6 It is surprising that ethnicity could affect funding decisions when it is a legally protected characteristic.

At best, such comments are unhelpful because they do not provide suggestions to enhance the quality of research. At worst, they contribute to widening ethnic inequalities in health when they are used to dismiss quality research.

We call for a “review of the review” process. A working group should be established with wide representation across government, voluntary and community organisations, patients, academia, and healthcare to create a set of guidelines for evaluating funding applications that could take the form of checklists.7 This could complement existing approaches to improving ethnic minority representation in research,8 including in patient and public involvement and project staffing.5

These guidelines should cover several issues—for example, they could advise on what constitutes good practice when identifying a research group, including consideration of the diversity of the research group and reviewers. It is not sufficient to simply match a trial population to the served population, as the NIHR recommends, if the matching does not allow for the identification of harms in some population groups owing to small sample sizes.9 Caution should be exercised when assuming that problems are shared between—and within—ethnic, racial, or religious groups to avoid oversimplified cultural categorising and perpetuating inequalities.10

Reviewers’ judgments differ about what to do when sample sizes for ethnic groups are small, and guidelines should clarify this. With small sample sizes, it can be hard to tackle certain questions. These are important given that ethnicity is socially constructed and ethnic differences in health may be attributable to other factors.11 This is a fundamental issue at the heart of epidemiology, which must practically categorise groups, even when the basis for the division is not strong.

Funding criteria based on past success of research team members should be treated with caution and integrated with criteria about tackling inequalities. When the status quo has produced a situation where ethnic minority patients still have worse health outcomes, and we lack good evidence on how to improve their health, it must change. Reviewers often strongly disagree when evaluating the same funding applications,12 consistent with evidence about variability or “noise” in decision making.13 Such evidence questions the legitimacy of prioritising previously successful applicants and about the funding peer review process more generally.14

In our experience, comments on funding applications that include ethnic minority groups fail to recognise the importance of research focused on these groups, even when it is clearly justified. It is a matter of integrity to ensure the quality of research commissioning and peer review processes. More transparent standards can help to ensure quality and representation from ethnic minority groups, contributing to improving health and reducing health inequalities.