.Back to listing

Be Careful What You Take As Fact: Commercial Test Results


While we were all on break, a Columbia University Master of Bioethics student, Kira Peikoff, published a fascinating article in the New York Times.  She wanted to assess the accuracy and precision of different commercial genomic testing companies, using her own DNA.  Peikoff submitted DNA samples to 23andMe, Genetic Testing Laboratories, and Pathway Genomics.   The results were worryingly scattershot: not only could the companies not agree on even a broad range of risk for certain common genomic conditions, they couldn’t agree if the risk they assessed indicated a problem worth consulting a physician about or significantly less than the average lifetime risk.  For example, G.T.L. assessed a 10.3% risk of Type 2 diabetes as “medium” while 23andMe thought a 15.7% risk of diabetes was “decreased”.

I should note that 23andMe’s DTC test is no longer returned with interpreted genomic health risks per an FDA order.  That being said, all three of these companies are profit-driven.  The “risks of disease” they present (or in 23andMe’s case, presented) for common diseases mediated by many genes and the environment  aren’t only marketing ploys, but neither are they a sufficient basis upon which to make major medical decisions.  What must be made abundantly clear to DTC consumers is the information the consumer is receiving is simply a best guess at an uncertain future…at the moment that guess is not all that great.  Moreover, by purchasing these companies’ products the consumer offers her data as proprietary data points for future uses.

With respect to health outcomes, there is not much evidence right now to suggest patients are giving much weight to these results: a study in the NEJM in 2011 found that among those receiving results from these companies, 90.3% reported no test-related distress. To me, this suggests that while consumers similar to Ms. Peikoff may not be aware of just how varied the returned results can be, they are probably not the suckers many outraged bioethicists take them for.

What is left, then, is the extent to which the public realizes it is surrendering its privacy by opting into these sometimes fun and fascinating yet often clinically inadequate test kits.   Readers, I’d love to hear your thoughts via email.