Abstract
The global burden of diabetes-across major forms such as type 2 diabetes, type 1 diabetes and gestational diabetes mellitus-disproportionately affects people of non-European ancestry, the majority of whom live in low- and middle-income countries. The heterogeneity of diabetes risks and phenotypes indicates that knowledge derived principally from European-origin populations may not be readily transferable to other groups. In this review our aim is to enhance the quality of diabetes research by championing the inclusion of diverse populations, ensuring clarity of population definition and encouraging exploration of population differences. We review the terminology used to define populations and make recommendations on the use of these terms. We argue that population membership by itself does not determine risks or response to intervention; rather, it is the confluence of genetic, environmental, sociocultural and policy factors that are causal and should be identified. We note that, while common diabetes forms are polygenic and populations are unlikely to harbour single genes that account for significant risk, environmental change that impacts lifestyle and biology demonstrably alters diabetes risk and provides opportunities for effective intervention. Similarly, while genetic variants are associated with adverse events, population group membership may sometimes not be a valid proxy for such variants, which has implications for healthcare equity. For most drugs used in diabetes there is little evidence that drug responsiveness materially differs by population grouping, although it is only recently that well-designed studies have been performed. In contrast, other population characteristics, such as sex, age and obesity, appear to alter glucose-lowering drug effectiveness and should be considered when prescribing. Inclusion of diverse populations in diabetes research, combined with a multidisciplinary approach, is essential if we are to combat the global burden of diabetes.