Failing to collect, analyse, and report ethnicity data in clinical research leads to healthcare inequalities

Collection of ethnicity data in clinical research has been poor historically, but is essential to ensure that interventions are effective and safe across diverse populations. These data are needed to monitor variations in the prevalence of certain conditions in different ethnic groups1 as well as differences in drug metabolism, treatment efficacy, and safety profiles.2 The covid-19 pandemic exemplified how disease can disproportionately affect ethnic minority populations.3 Clinical trials must reflect the heterogeneity of the global population.

Studies have shown that genetic, cultural, environmental, and socioeconomic factors linked to ethnicity can influence the manifestation of diseases and responses to treatment.45 Furthermore, ethnic groups are heterogeneous. For example, crude umbrella categories such as South Asian (ie, of Indian subcontinent ancestry) are often used in research, however, large variations exist in cultures, behaviours, diseases, and risks within this “group” (ie, Indian, Pakistani, and Bangladeshi).6

Catalysed by the covid-19 pandemic, adequate ethnic representation in clinical trials is now being seen as a scientific necessity and not just a social justice imperative. For example, recent trials of organ protective therapies such as sodium-glucose co-transporter 2 inhibitors (SGLT2-Is) and glucagon-like peptide 1 receptor agonists (GLP1-RAs) have highlighted substantial ethnic and racial differences in cardiorenal effects in people with type 2 diabetes. Consistent benefits have been observed among white and Asian populations, and a consistent lack of benefit in black populations.7

Failing to collect, analyse, and report ethnicity data in clinical research can lead to healthcare inequalities, such as misdiagnoses, inappropriate dosing, and reduced treatment efficacy. Without these data, clinicians and regulators lack the information needed to tailor care appropriately, potentially perpetuating inequalities in health outcomes.

In June 2024, the US Food and Drug Administration (FDA) published draft guidance that aimed to improve enrolment of participants from underrepresented populations in clinical studies, including ethnic and racial groups.8 However, in January 2025, US president Donald Trump issued an executive order curtailing equality, diversity, and inclusion programmes, without any consultation. The draft guidance was removed from the FDA website9 but was reinstated by 11 February 2025 following a federal court ruling. As of July 2025, the link to the guidance on the FDA website has been removed, and no further details provided.

UK regulators are taking steps to improve diversity in clinical research. In March 2024, the Health Research Authority (HRA) and the Medicines and Healthcare products Regulatory Agency (MHRA) jointly released draft guidance requiring the development and submission of inclusion and diversity plans for all new clinical trials.10 This offers a framework for trial sponsors and researchers to set out how they will ensure that people from diverse ethnic and social backgrounds are represented. It includes expectations for community engagement, removing barriers to participation, and encourages demographic data transparency.

This guidance marks a significant shift toward embedding equity and inclusion into the design and execution of research in the UK. The HRA and MHRA guidance is clear on the reasons for inclusion of underserved populations. These include improving our understanding of a disease and the product or intervention under investigation; ensuring that the data generated reflect the diversity of the population expected to use the medical product or intervention; and potentially identifying effects on safety or efficacy outcomes that may be associated with or occur more or less frequently within these populations.10

Implementing this guidance requires that information on the diversity of participating populations is adequately captured and reported, but collecting such data presents challenges. Ethnicity, for example, is difficult to define and measure, as it encompasses notions of race (identification of self, based on physical appearance) biology, religion, and culture. Further barriers include lack of trust in institutions and researchers, and the limited transparency around data use.11

To support researchers and health care professionals, recent recommendations from the UK Health Data Research Alliance Ethnicity Coding Working Group were developed, including calls for development of a national framework and guidance to ensure ethnicity is captured and recorded accurately and consistently.11 A further recommendation included training for frontline and research staff involved in collecting ethnicity data, including cultural competency training to ensure sensitivity and confidence in data collection.

Alongside HRA and MHRA guidance, the National Institute for Health and Care Research recently mandated inclusive research practice as a condition for research funding.12 This is an important step as regulators and funders have a crucial role in directing the future of research conduct, applicability, and implementation. Diversity in clinical trials and collecting ethnicity data as well as other protected characteristic data is vital to achieving health equity and scientific validity. While UK research regulators and funders are making progress, recent developments in the US highlight the worrying obstacles presented by mixing political ideology with diversity related policy. This calls for urgent action by all stakeholders including governments, regulators, sponsors, researchers, and journal editors to ensure research is applicable to and benefits all.