Don't Worry About the Government (and Genomics)

For the past week or so, the wonkosphere has been ablaze over the fact that the California Department of Public Health issued "cease and desist" orders to 13 companies that offer genetic testing in the state unless they comply with regulations that require such tests to be ordered and overseen by doctors. A typical reaction is that of Wired deputy editor Thomas Goetz, who argues that "it's insulting and a curtailment of my rights to put a gatekeeper between me and my DNA."
California, of course, isn't alone in cracking down on the nascent business of "personal genomics," in which any number of companies -- some reputable, some less so -- offer to scan an individual's DNA for genetic variations that reflect higher or lower risks of developing particular diseases. Back in April, New York issued warning letters to 23 gene-testing firms, including Google-backed 23andMe and Navigenics, telling them they needed a permit to offer services in the state. For some reason, though, it took California stepping into the fray to really get genomics enthusiasts exercised.

But all this outrage is misplaced. While it's easy -- and, let's admit it, fun -- to ridicule hidebound state bureaucrats, it's also true that most personal-genomics startups have gamed the system where regulation is concerned. While their businesses are built on the assumption that people will find their genetic information interesting and useful, companies like 23andMe also insist that their offerings amount to "information services" with no real medical utility.

If that's true now, it won't be for long, so a collision between companies offering "democratized" access to genomic information and health authorities was probably inevitable. It's just as well that we get it out of the way sooner rather than later, because anyone who thinks this field will remain unregulated is deluding themselves.

Goetz, for instance, may be right to argue that the state has no business telling people how or where they can access their most personal genetic information, but I'm not so sure. Consider just three reasons:

• Accuracy and quality standards would boost confidence. For now, personal genomics is mostly a playground for wealthy, educated technophiles. If it's ever going to move into the mainstream, quite a few people are going to want reassurance that they're getting what they pay for and that the interpretations of their genetic findings are sound.
• Regulation can weed out snake-oil salesmen. Sure, the market can often do this as well (and possibly more efficiently), but it comes at a cost -- that of tolerating an anarchic, Wild West genomic sensibility that could easily tarnish the reputation of every company in the space. You don't have to look far today to find companies selling dubious genetics tests or using them to hawk "health supplements" and the like -- and that will only get worse over time, given relatively low barriers to entry and how little your average genomics customer will likely understand about what they're buying.
• Personal-genomics outfits may benefit from regulation. A single well-publicized tragedy -- say, a suicide related to an incorrect genomic prediction of early-onset Alzheimer's disease -- could deal a serious setback to the whole field. Should such a situation ever arise, personal-genomics outfits would almost certainly be better off sheltered by smart rules than out on their own making abstract arguments about risk and uncertainty.

Finally, there's also a deeper philosophical issue at stake, which is that in medicine, personal autonomy is somewhat overrated. In particular, the phenomenon of medical overtreatment, which may result in tens of thousands of unnecessary deaths every year, stems in part from the heavy marketing of drugs and medical services to patients who may well not need them. (Doctors are almost certainly more to blame for most overtreatment, but that's a separate topic.) It's very likely that laissez-faire personal genomics could make the problem even worse by driving the "worried well" to seek unnecessary testing and treatment -- much of which would be expensive, and some of which might easily be harmful.

In other words, smart rules for genomics could very well establish the sort of "choice architecture" that Richard Thaler and Cass Sunstein suggest in their recent book Nudge -- one that helps the majority of people make the best choices about their health without constraining them overmuch. Admittedly, that's most certainly not what California's regulators are doing now. But it's not exactly an insult or a curtailment of anyone's "rights," either.

Update: I wouldn't have expected it, but apparently 23andMe's Linda Avey effectively agrees with me -- sort of -- on the need for smart regulation of personal genomics:

We agree that this evolving field of personal genomics is in need of proper regulatory oversight. While our mission to provide accurate and contextual information to our customers about their genetic information is aligned with the regulatory mandate to protect the public health, we also want to ensure that efforts to rein in our industry do not hamper the potential benefit of genetic knowledge to our health.