Don’t Let Bureaucrats Get Between You and Your Saliva Sample
April 14, 2017
For the past decade, spitting saliva in a tube for DNA processing has been all the rage for viewers of the Maurytelevision show, science geeks, hypochondriacs, and everyone in-between. Companies such as 23andMe have vigorously marketed their ability to use these genetic samples to identify customers’ ancestry and health risk. But the firm’s foray into health status prediction was too much for the Food and Drug Administration (FDA), which finally went nuclear on 23andMe after several warnings.
In 2013, the FDA barred 23andMe from sharing health related conjecture tailored to customers’ genetic markers. For example, my own 23andMe report, which gave me risk percentages of developing major diseases relative to the population, was suddenly devoid of any such projections after the FDA’s actions. In justifying the ban, the administration reasoned that any disease probability estimate was inherently misleading, since the vast majority of genes playing a role in the development of the illnesses are not known. In the government’s view, 23andMe was assigning an unreasonably-high weight to a handful of genetic markers they’d analyzed.
The administration slightly backtracked in 2015, when they allowed the testing company to alert their customers to the presence of “carrier genes” in user DNA. Genetic markers associated with cystic fibrosis and sickle cell anemia strongly predict the diseases in carriers’ children, and are thus particularly useful in pre-identifying health threats. The latest liberalization, announced a few days ago, allows 23andMe to identify ten diseases and/or conditions that may afflict the actual customer, as opposed to his/her progeny. The list of approved diseases, which includes late-onset Alzheimer’s Disease and Parkinson’s Disease, was approved by the FDA only after agency certification that identified markers were in fact strongly correlated with the diseases. That is, 23andMe’s commentary on customer DNA is misleading unless the FDA can confirm the DNA-disease linkages put forward by the company.
While the agency’s loosening of rules is a step in the right direction, the prevailing process is still far too strict. For starters, the FDA’s regulatory approach presupposes that customers will react in a very peculiar way to the withholding of information from them. If 23andMe shares subscribers’ DNA sequences with them but withholds any educated guesses about disease profiles, customers are unlikely to simply “rollover” and confine themselves to uncertainty. Rather, they’re likely to sleuth genetics discussion forums and pseudoscientific web pages for more information.
It’s not surprising that Google Trends for search terms like “DNA results” and “genetic markers” show small but noticeable growth starting in 2013 after remaining flat or declining for the previous two years. Among the beneficiaries of this internet traffic has been a low-budget web service called Promethease, where users can input their raw genetic data inputted via 23andMe and obtain a comprehensive medical literature review on their genetic markers. Search traffic around this website, which nearly tripled in the three months after the FDA’s initial clampdown on 23andMe, demonstrates a consumer willingness to work around cumbersome rules imposed by bureaucrats.
But not all user formats are created equally. In contrast to 23andMe’s pre-2013 organized guides exploring links between genetic markers and diseases, “low-tech” search engine alternatives offer a near-endless supply of conflicting information. While a deep dive into the literature benefits excites genetic enthusiasts, the average browser will have a more muddled experience. Research on “information overload,” for example, has shown that study participants have difficulty internalizing more than seven distinct bits of information at a time. Thus, memorization becomes extremely selective when making processing huge hunks of material.
And when it comes to reading up on genetic markers, this suboptimal research glut is bound to continue if current FDA regulations remain in place. 23andMe customers are bound to be misled by at least some conjecture, whether on the website itself or through an unregulatable “medical literature retrieval service.” Policymakers should, however, allow customers to choose the format by which that B-grade information is being provided. A cleaner, more selective visualization of the data offered by pre-regulation 23andMe ultimately does the best job in informing DNA test-takers of potential health risk. Bureaucrats at the FDA ought to respect these developments, rather than shackling genetic pioneers.
With a new agency head, the FDA has an opportunity to keep pace with technology, innovation, and consumer demand.