Disintermediation: Why Genomics is More Like Insulin Injections than Napster

Some are comparing the shake-up of business models in the genomics industry by DTC companies with the disruption of the music industry via P2P services like Napster, who radically changed how music was distributed. Seemingly overnight, the music industry felt they had been bamboozled and marginalized. Their role as the primary distributors, gatekeepers, and fee collectors of music had been challenged — and by a teenager! This made them very unhappy, to say the least.

Like Napster, the consumer genomics industry is a force of disintermediation. Access to personal genetic data no longer requires a trip to a clinic, or any face-to-face interactions with physicians or genetic counselors, as they have in the past. This makes the traditional medical genetics community very unhappy. Like the music industry, they do not want to be marginalized as distributors, gatekeepers, and fee collectors of genetic knowledge.

Some would argue that the analogy goes further: the ultimate fate of Napster will be the fate of consumer genomics companies. Napster was shut-down by court order, and later re-emerged with a more traditional model of distribution that re-inserted the intermediaries they famously had marginalized via P2P. Some believe its only a matter of time before consumer genomics companies are shut-down and are required to re-insert physicians and face-to-face counseling.

I disagree. The future of the consumer genomics industry is more likely to follow the path of insulin injections, than music. The reason is purely economic. To be sure, if diabetics needed insulin injections only once a year, rather than several times a day, physicians would still be in control of the syringe.

Insulin is both life-sustaining and life-threatening, depending on the dose. The risk that a diabetic patient might inadvertently kill themselves with the wrong dose of insulin is real. However, the fact that diabetics need injections on a regular basis, even several times a day for some, meant that doctors were out-of-the-picture. Self-injection of insulin by diabetics is so commonplace today, its easy to forget just how radical the practice of self-injection is, provided the dangers to the patient and the well-intended desire of the medical establishment to protect them.

We now accept the disintermediation of insulin injections without a second thought. Could it really be any other way? The same will be true of genomics. Although self-examination of biological data presents many risks to individuals, the expense of forcing people to go through gatekeepers and censoring data in the interest of protecting individuals from themselves will be too burdensome. An editorial in this month’s issue of Nature Biotechnology, sums this up nicely:

Faced with this huge expansion of data on ‘my’ health, it will simply not be affordable to maintain our dependence on medical gatekeepers, whether they are physicians or genetic counselors, without individuals taking a much greater responsibility for their own wellbeing. Whether the medical establishment likes it or not, it will be too cumbersome and too expensive to conduct personalized medicine if all diagnostic-to-therapeutic decisions depend on doctors…For personal genomics not to be stillborn, the medical community and regulators thus need to reevaluate their role as gatekeepers. Clearly, they need to be involved in the medical actions that might follow as a consequence of genetic or other diagnostic testing…But simply shutting down the whole direct-to-consumer gene testing enterprise because it departs from the traditional genetic testing paradigm of doctor ordered test will both retard progress and stifle investment in more advanced whole-genome sequencing technologies—technologies that have the potential to ultimately deliver the promise of genome-directed medicine.

Risks and responsibilities associated with self-management of genomic data will increasingly be transferred from physicians to individuals, as they were with insulin injections (and home pregnancy tests). The medical community and regulators will play incredibly important roles in genomics, but serving as gatekeepers to biological data is not one of them.

Editorial. “In need of counseling?Nature Biotechnology. 26(7):716. July, 2008.

Comments

12 Responses to “Disintermediation: Why Genomics is More Like Insulin Injections than Napster”

  1. Mr. Gunn on July 9th, 2008 3:02 pm

    Right on! This is exactly what I’ve been saying. All we need is a decent body of data from individuals of varying ethnic descent, and perhaps the coming healthcare reform, and people will wonder how this was even a debate in the first place.

    Yea for disruptive technology!

  2. David Hamilton on July 9th, 2008 7:09 pm

    At the risk of repeating myself (I just left the same comment over at Tim O’Reilly’s blog): It’s an interesting analogy. But don’t diabetics still require a prescription for their insulin? In other words, the disintermediation only went so far in that case, and I suspect the same will be true in personal genomics as well.

    Honestly, I don’t expect doctors to remain as “gatekeepers” for this information — that’s effectively a lost battle (although bad luck and/or bad decisions could mean it takes longer than many expect). But I don’t see what’s wrong with simpler, smarter rules that oversee testing procedures, set accuracy bounds, and limit the claims companies can make about their services (a la the ones hawking supplements, although that’s a more complicated problem).

    I’m even sort of sympathetic to the notion that regulation might force companies to offer independent genetic counseling to customers, since I think as you travel down the bell curve from early adopters, you’re going to hit a whole bunch of potential customers who are going to have trouble understanding exactly what their readouts are telling them, and for whom that misunderstanding might easily be tragic. No one would have to accept the counseling in this scheme, but at least it would be available for those who think they need it.

    In my view, there’s a lot to like about this sort of libertarian paternalism, since if behavioral economics has taught us anything, it’s that many people consistently make bad decisions even with perfect freedom. Done properly, soft constraints on those decisions strike me as a terrific idea.

    Whether it’s realistic is an entirely different question, of course. Then again, you could say the same thing about the notion of a completely unregulated genomics market.

  3. jasonbobe on July 9th, 2008 7:39 pm

    @David: “don’t diabetics still require a prescription for their insulin? In other words, the disintermediation only went so far in that case and I suspect the same will be true in personal genomics as well.”

    Insulin is an imperfect analogy (aren’t they all?). Rx is required as injectable drug! That’s why, in my opinion, the analogy actually works quite well. We allow individuals to inject themselves with insulin, but we entertain restrictions on access to data? As you say, this is a “lost battle” from the get-go.

    Yes, physicians will also still play an important role in genomics too. I do not dispute that, it should be encouraged and supported.

    @David: “But I don’t see what’s wrong with simpler, smarter rules that oversee testing procedures, set accuracy bounds, and limit the claims companies can make about their services (a la the ones hawking supplements, although that’s a more complicated problem).”

    YES! Simpler, smarter rules are much desired. I’m heading that direction, these last two posts are merely foundation for getting into details.

    Best to start with the easy issues. Access to my personal biological data seems like a no-brainer to me, and good place to put a stake in the ground. The issues are complicated, as you know: CLIA, the definition of medical advice, professional liability, nuances of NY/CA state regulations, etc. Time permitting, I hope to knock out a few more essays over this next week that gets deeper into the issues. And, of course, a purely unregulated genomics market ain’t going to happen (and I’m not recommending it).

    I would love hear your recommendations for simpler, smarter rules. Here is an early draft of mine, feedback would be great.

    P.S. The hucksters in genomics are annoying and a nuisance (as they are in every other domain), but they should not be the driving force in policy. I think you would agree with that.

  4. Steven Murphy MD on July 9th, 2008 8:17 pm

    @Jason
    “And, of course, a purely unregulated genomics market ain’t going to happen (and I’m not recommending it).”

    Isn’t that what got us here in the first place?

    Also, people don’t just go out and start shooting insulin. They get education from a trained professional. No, it is NOT OPTIONAL. It is mandatory. End of story. They also get follow up and even have rules given to them by trained specialists….to call this GWAS novelty akin to insulin therapy is sort of appropriate….There is a whole lot of education and training that goes into self injection of insulin. And glucose interpretation is a whole lot easier than SNPs and CNVs buddy….you know that as well as I do. As for NYS….well, you got me there. I don’t think the feds will be as restrictive.

    I look forward to your treatises

    -Steve
    http://www.thegenesherpa.blogspot.com
    P.S. Do you have DM1? You should see how many people come in for DKA from inappropriate lack of Insulin use…..

  5. David Hamilton on July 9th, 2008 8:50 pm

    We allow individuals to inject themselves with insulin, but we entertain restrictions on access to data? As you say, this is a “lost battle” from the get-go.

    Well, in a way, any regulation of genomic-information providers amounts to some sort of restriction on access to data. I see a lot of people in this argument — not you, necessarily — making sweeping claims about having an absolute right to their own biological information, which strikes me as so much eyewash. If they mean anything, clinical standards and accuracy reporting will weed out some providers, thereby “restricting” access to data in some way (on cost, if nothing else). I don’t see anything wrong with that, but I suspect not everyone interested in this subject would agree.

    The insulin analogy, which I agree is imperfect, still illustrates that you can open up access to a medical technology while retaining gatekeepers. No one gets insulin who isn’t a diagnosed insulin-dependent diabetic under the care of a doctor, after all. What’s more, no one can make insulin without satisfying the requirements of the FDA. (There’s also the fact that disintermediation happened in that field largely as the result of technological advances and the fact that the alternative condemned diabetics to short, miserable lives.)

    Seems plausible to me that something similar could work out in personal genomics, and your draft chart is a great place to start. As I’ve said, I don’t see the point of getting a physician’s permission for access to my genetic data, but I see a lot of room for restrictions on who gets to offer the scans and under what circumstances, as well as on what happens to the data afterward — I’d definitely be amenable to guidelines on data retention and privacy, for instance, and perhaps a default counseling mandate. (You could even call it the Genetic Counselors’ Employment Act, although there again, I suspect we’re going to need a lot of them before much longer.)

    My main concern about your framework is that the line between medical and non-medical uses of genomic information is pretty hard to draw, since once an individual has their genomic sequence they can pretty much do whatever they want with it. We’ve already watched 23andMe skate along that line — I mean, current and potential future medical utility is one of the main reasons anyone is interested in looking at their genome in the first place, but since the start 23andMe has insisted their service has no such utility at all, specifically so they wouldn’t be regulated. I’m not sure things how would be any different under your proposed regime, since if the advantage of being unregulated is great enough, many companies are going to be tempted to cross that line and take their chances.

    As for the hucksters, I agree they shouldn’t be the driving force in policy, but I think they’ll inevitably be a driving force, because public agencies are usually driven to react to the excesses of the worst actors. Agreeing on rules that knock out or severely hamper the bad apples would be good for the field as a whole, although it would also be “uncompetitive” in a classical economic sense — not to mention a restriction on access to data.

  6. Steven Murphy MD on July 10th, 2008 5:39 am

    David and Jason,
    This is the difficulty with calling this non-medical……In addition, I am curious….which one of you is diabetic or has a diabetic relative?

    -Steve
    http://www.thegenesherpa.blogspot.com

  7. Jason Bobe on July 10th, 2008 11:30 am

    I’m going parse a few things from your remarks David, some good stuff in there.

    @David: “I see a lot of people in this argument — not you, necessarily — making sweeping claims about having an absolute right to their own biological information, which strikes me as so much eyewash.”

    I might be one of those people! Not sure, if we’re talking about the sames things here.

    I would like to separate data providers, e.g. such as labs who provide sequence data, from those who provide interpretations, such as medical advice, e.g. diagnoses, or educational information, such as WebMD.

    Yes, I do think restricting access to biological data such as basepairs should be avoided. I’m not clear on the arguments to the contrary, or their strengths. What are the conditions with which we should abandon personal autonomy? Because we’re worried that people will act inappropriately? Specific cases of injury?

  8. Jason Bobe on July 10th, 2008 11:50 am

    @Steve:
    “Isn’t that what got us here in the first place?”

    I’m not sure where “here” is, but, to my knowledge the California situation stemmed from 2 consumer complaints, one person thought the services should be cheaper (!!!), the other person claimed there were problems with validity of something. I’m not sure about the details of the claim, more info would be helpful.

  9. Steven Murphy MD on July 10th, 2008 2:57 pm

    Jason,
    I thought there were 12 complaints…….
    Also…no one answered my question about the diabetes?????? Are you guys having HIPAA/privacy issues??????

    LOL
    -Steve
    http://www.thegenesherpa.blogspot.com

  10. Jason Bobe on July 10th, 2008 10:19 pm

    @Steve: “Also…no one answered my question about the diabetes?????? Are you guys having HIPAA/privacy issues??????”

    You baffle me sometimes. I’m afraid to ask why you think HIPAA is relevant to a lack of response about diabetes.

    Jason

  11. Jason Bobe on July 10th, 2008 10:27 pm

    @David…

    Regarding the discussion on the rights of individuals to access their own biological data, and the conditions under which autonomy might ethically be restricted, I’ve found this paper very helpful to consider:

    http://www.ajph.org/cgi/content/abstract/98/1/15

    Still digging for more.

    Thanks,
    Jason

  12. Steven Murphy MD on July 15th, 2008 10:28 am

    Jason,
    I reviewed your rough draft. I think it is great. There is a big problem, that can only be solved one way…..

    Non-medical genomic screening cannot release any information on medical risk alleles etc. I.E. you get genome screened for ehnicity….All that you receive in interpretation IS ethnicity…..

    Otherwise you have what currently exists and that regulators are fighting hard against….Which is the release of any medical risk factors…..regardless of what disclaimer is out there. People don’t read disclaimers……

    So I propose you change your slide to exclude non-consensus medical data altogether and also exclude medical information altogether from your non-clinical algorithm.

    I just don’t see how any medical information regarless of whether it is valid or not will ever leave the legal grasp of the medical regulators. Medicine is one of the most heavily regulated fields….I don’t see that ever changing for the lesser. Even if there are disclaimers…

    -Steve
    http://www.thegenesherpa.blogspot.com

Leave a Reply




You may use HTML tags in your comment. Please be patient, comments may take a while to post.

Subscribe without commenting