Reporting and Social Construction of Race in Alzheimer’s Disease Clinical Trials

I read with interest the work by my colleagues in the Diversity Professional Interest Area on the under-reporting of race in Alzheimer’s disease (AD) clinical trials.[1] This follows other recent work documenting a long-known phenomenon.[2] The affiliation of these and previous authors with centers which conduct majority of said clinical trials is encouraging for greater representation in future study designs, but a key chasm between practitioners in clinical trials and health disparities is perhaps more the reasoning behind inclusion than omission (which the authors documented) or potential exclusion.

Whereas observational studies are widely used to generate new hypotheses, clinical trials are largely designed to test very specific postulates. Balancing scientific and social responsibilities in these trials is not always straightforward. If a drug undergoing testing targets a rare disease which primarily occurs in an imaginary alien race, it makes as much logical sense to include all human races as it would to include large numbers of AD patients in a clinical trial for familial frontotemporal degeneration. If race is a completely social construction – W.E.B. Du Bois’ conceptualization that racial experience was assigned by the dominant group over one’s own in post-Civil War US – without biological correlates, what is the scientific rationale for including non-White adults? If there are hopes of identifying contemporary social determinants’ impact on response to intervention, how can representation accomplish this in a non-pragmatic trial while omitting appropriate measures for these social determinants (which are themselves underdeveloped in biomedicine)? If centuries of social construction had biological outcomes which can reasonably be expected to influence drug metabolism or response, mere representation mirroring the national make-up will also not achieve adequate power in any of the non-White populations. Designs short of intentional over-representation of non-White adults will then paradoxically diminish hopes for sufficiently-powered trials to detect response differences between individuals regardless of the epistemology of group-level differences.

Notwithstanding greater scientific rigor, investigators in Alzheimer’s disease need to be mindful that trials which over-enroll non-White participants risk distortion in intent and messaging. The African American Study of Kidney Disease and Hypertension (AASK) was the first to show race-related dissociation between predicted and observed renal clearance.[3] Greater muscle mass (along with diet, renal clearance, and others) was one possible explanation offered for the mismatch, yet it became an easy target for those looking for scientific racism in the media. This contributed to the recent elimination of race from glomerular filtration rate estimation – a practice known to introduce inaccuracy in Black adults – with the additional cost of introducing misdiagnosis in White adults.[4] Future corrections of this myopic practice will undoubtedly be necessary, but all the while funds allocated for these exercises can substantively reduce care disparities. Should Black health take a backseat to academic medicine’s discomfort in responding to misplaced vigilantism? I hope not.

How race is operationalized will continue to differ as time, borders, and society evolve. Reporting numbers of non-White participants is first and foremost good scientific practice because it is necessary to determine generalizability, and exclusion remains theoretical until there is mandatory report of screen failures’ characteristics. Reporting needs to be accompanied by a thoughtful re-evaluation of whether race as a social construction has meaningful biological correlates to limit rote generalization from predominantly White studies. Quantitative genetic and non-genetic traits – which have never contradicted the 99.9% genomic identity argument – remain the most valid reasons for over-enrollment and hypothesis-testing. Investigations into these traits have the most promise of centering the health disparities debate, and are more likely to achieve brain health equity than academic activism.