Researchers call for reform in the use of racial data in biomedical research

A recent study published in Synthese challenges the default collection and reliance on ethno-racial data in biomedical research, arguing that these practices often lack scientific or ethical justification. The paper critically examines the widely accepted “fair subject selection” requirement, which mandates proportional representation of racial groups in clinical trials, and suggests that this approach might inadvertently perpetuate scientific inaccuracies and social stereotypes.

The authors, Tomasz Żuradzki from Poland’s Jagiellonian University in Kraków and Joanna K. Malinowska from Adam Mickiewicz University in Poznań, propose an alternative framework that emphasizes hypothesis-driven research and transparent justification for population stratification.

The current practice and its flaws

Biomedical research in the United States often requires researchers to stratify human participants in research by their racial and ethnic categories, aiming to address health disparities and ensure fair representation. However, this study highlights several fundamental problems with these requirements.

First, racial categories, as defined by U.S. regulations, are social constructs rather than biologically meaningful classifications. This reliance on race risks conflating complex social, environmental, and genetic factors into oversimplified groupings, which may lead to spurious correlations and unfounded generalizations about health outcomes.

The authors argue that the current practice of mandating ethno-racial data collection often leads to post hoc subgroup analyses that lack prior hypotheses. This approach increases the likelihood of false positives and reinforces unsupported claims about racial differences in health outcomes. Furthermore, such analyses may divert attention from more relevant variables, such as socioeconomic status, environmental exposures, or specific genetic markers.

An ethical and methodological reassessment

The study emphasizes that fairness in participant selection should not equate to a blanket requirement to include racial categories in all research with human participants. Instead, the authors advocate for a nuanced understanding of fairness that considers the specific aims and contexts of individual studies.

For instance, while it may be ethically justifiable to prioritize underrepresented populations in certain contexts, researchers should avoid imposing arbitrary racial classifications unless they are clearly linked to hypotheses about relevant biological (including genetic), environmental or social mechanisms.

The authors also critique the epistemological basis of racial stratification in research. They assume that prediction is usually epistemically superior to accommodation in scientific inquiry. This means that hypotheses generated before data collection, based on evidence of mechanisms, are more likely to yield reliable and actionable insights than those derived retrospectively from observed patterns. In contrast, the widespread practice of accommodating data through subgroup analyses often prioritizes regulatory compliance over scientific rigor.

A proposal for change

To address these issues, the authors propose a practical framework for population stratification in biomedical research. They recommend that researchers who intend to use racial categories should:

• Justify stratification choices: Provide a clear rationale for using specific racial categories, grounded in hypotheses about mechanisms linking these categories to health outcomes.
• Pre-register hypotheses: Include stratification plans and associated hypotheses in pre-analysis plans to enhance transparency and accountability.
• Move beyond default categories: Encourage flexibility in population categorization, allowing researchers to define subgroups based on study-specific evidence rather than adhering to predefined racial classifications.

This approach mirrors existing practices like pre-analysis plans in other research fields, which are used to mitigate biases and enhance methodological transparency.

Implications for research and policy

The proposed framework has significant implications for both researchers and regulators. For researchers, it underscores the importance of integrating ethical and epistemological considerations into study design, ensuring that population stratifications are not only scientifically justified but also socially responsible. For regulators, the framework calls for a shift from mandating blanket inclusion of racial data to fostering critical evaluation of stratification practices on a case-by-case basis.

Moreover, the study highlights the need for closer collaboration between research ethics and the philosophy of science. By bridging these disciplines, researchers can better address the methodological and normative challenges of population stratification in biomedical research.

A path forward

While acknowledging the potential value of ethno-racial data in specific contexts, the authors caution against its routine use without adequate justification. They argue that moving away from rigid racial stratifications can lead to more precise and equitable biomedical research. This nuanced approach recognizes the complexity of human diversity and the limitations of current practices, advocating for research designs that prioritize both scientific rigor and ethical integrity.

This study invites regulators, researchers, and ethicists to critically reflect on the role of racial categories in medical research and to adopt practices that are both methodologically sound and socially just.

Provided by
Jagiellonian University