J Med Ethics 32:724-728 doi:10.1136/jme.2005.014456
  • Research ethics

Non-reporting and inconsistent reporting of race and ethnicity in articles that claim associations among genotype, outcome, and race or ethnicity

  1. H Shanawani1,
  2. L Dame2,
  3. D A Schwartz3,
  4. R Cook-Deegan2
  1. 1Department of Biostatistics and Research Epidemiology, Henry Ford Hospitals, Detroit, MI, USA
  2. 2Duke University Center for Genome Ethics, Law, and Policy, Durham, North Carolina, USA
  3. 3National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA
  1. Correspondence to:
 Hasan Shanawani
 Department of Biostatistics and Research Epidemiology, Henry Ford Hospitals, 1 Ford Place, 5C-69 Detroit, MI 48202, USA; hshanawani{at}
  • Received 16 September 2005
  • Accepted 16 March 2006
  • Revised 14 March 2006


Background: The use of race as a category in medical research is the focus of an intense debate, complicated by the inconsistency of presumed independent variables, race and ethnicity, on which analysis depends. Interpretation is made difficult by inconsistent methods for determining the race or ethnicity of a participant. The failure to specify how race or ethnicity was determined is common in the published literature.

Hypothesis: Criteria by which they assign a research participant to racial or ethnic categories are not reported by published articles.

Methods: Methods were reviewed for assigning race and ethnicity of research participants in 268 published reports reporting associations among race (or ethnicity), health outcome and genotype.

Results: Of the 268 published reports reviewed, it was found that 192 (72%) did not explain their methods for assigning race or ethnicity as an independent variable. This was despite the fact that 180 (67%) of those reports reached conclusions about associations among genetics, health outcome and race or ethnicity.

Conclusions: More attention needs to be given to the definition of race and ethnicity in genetic studies, especially in those diseases where health disparities are known to exist.


  • Competing interests: None.

  • Contributors: HS generated the idea for this project and wrote the protocol that was further elaborated by the other two authors. He also obtained the data from the articles. All authors interpreted the data and analyses. HS wrote the final draft. LD, DAS and RCD commented on it critically.