September 1, 2016
Reflection on Race & Ethnicity in medical device studies and claims
In June 2016, the FDA distributed a draft guidance document on evaluation and reporting of age, race, and ethnicity data in medical device clinical studies. It was about the same time that I got introduced into the work of Prof. Roberts, an acclaimed scholar of race, gender and the law, who advocates how race-based medicine should be avoided. The contradiction these parties are trying to convince the reader about is immense and not only made me ask myself whether I put so called ‘race and ethnicity based medicine’ into practice, but more importantly, what effects this may have on the parties involved.
The collection of race and ethnicity data as part of subject demographics is common practice in medical device clinical trials across the world. This practice could be based on the shared thought of Taylor & Wright: genetic differences, biomarker differences, environmental differences, lifestyle choices and social factors – all items related to race and ethnicity in part or in whole – may directly or indirectly affect disease prevalence, pathophysiology and outcomes between populations and thereby the performance, safety and/or efficacy of the medical device under evaluation. One example is the mean aortic wall thickness, that has shown to be different between racial/ethnic populations. Your cardiovascular device may be specially designed for a particular wall thickness, providing excellent results for the one population, but limited or even negative results for the other.
Still, only 27% of the studies reviewed in the 2013 FDASIA 907 report contained a race or ethnicity subgroup analysis – an unexpectedly low number. The FDA guidance has therefore been established to encourage (considerations of) collection of relevant age, race, ethnicity and associated covariates; to outline recommended analyses of study subgroup data; and to specify FDA’s expectations for reporting age, race and ethnicity-specific information in summaries and labeling for approved or cleared medical devices. The guidance describes that if clinically relevant differences in treatment effect are anticipated, these should be considered in the study design and appropriated reported in the device labeling. This can be achieved by enrolling representative proportions of age, race and ethnicity subgroups.
Prof. Roberts advocates the opposite of what is requested by the FDA. She states that ‘race based medicine supports a new biopolitics of race that threatens to make health and other social inequalities even worse’. Using BiDil drug as an example, Roberts argues that claiming a drug only available for a certain race population is based on scientific flaw, where proof of superiority of the drug in one population group has not been shown. Additionally, she elaborates on how race based medicine provides extra marketing opportunities for the pharmaceutical industry, not directly supporting the demands of the health care users. She mentions how the FDA supported market access to the Afro American population only when the drug was tested for safety and effectiveness in the Afro American population only, while the same considerations are not being put into practice if studies have been performed in a white population only.
So what does this mean for our medical device trials and claims? Personally, I believe that race and ethnicity could have their influence on study results and device safety, performance and effectiveness. As a result, I agree on the FDA’s considerations to make products only available for the population tested for. However, the words by Prof. Roberts made me aware on how this is not a generally accepted method, and that care should be taken when asking subjects for race and/or ethnicity data, analyzing study results, and claiming market access. Medical device manufacturers should be aware of the pros and cons, and the related responsibilities, to analyze their product for race/ethnicity or not, and establish their device claims accordingly.
By Anne Leijsen