Do you want to know the future?

That is the big question being considered in this long read about the direct-to-consumer genetics firm 23andme and one mother’s struggle to make sense of (and peace with) a $99 genetic profile she got of her 5 year old adopted daughter. I was interviewed for this story because of my work on the use of genetic markers as potential risk adjustor/underwriting variables for private long term care insurance and the fact that the author’s daughter was found to have the e4/e4 variant of APOE-4 which confers increased risk of Alzheimer’s disease.

The author of the story (it is published under a pseudonym) and I talked for a long time, and her anxiety about her daughter getting Alzheimer’s disease was palpable, a result to which she had locked onto. My advice to her was this:

But I appreciate the advice from Duke’s Don Taylor most. “It’s possible the best thing you can do is burn that damn report and never think of it again,” he said. “I’m just talking now as a parent. Do not wreck yourself about your 5-year-old getting Alzheimer’s. Worry more about the fact that when she’s a teenager she might be driving around in cars with drunk boys.”

It is not that AD is not a terrible disease, and goodness knows our nation’s long term care system is messed up, but you can only worry about so many things at once. There must be good uses of such genetic profiles, but it seems easier to get it wrong than right. And coming up with a coherent long term system is not important because a given person is at increased risk of AD, but because we know for certain that many of use now alive will contract the disease, it is just not clear who.

cross posted at freeforall.

Comments

  1. paul says

    Absolutely. One of the crucial questions for any medical test is “are you going to do anything differently depending on the result of this test?” As long as there’s no cure for Alzheimer’s — or even early treatment that one might administer in the absence of actual symptoms, “knowledge” like this isn’t really useful. It could even be anti-useful. Even for a disease with far worse consequences — Huntington’s chorea — many of the people at risk have decided that knowing the future is a net negative.

    The other part of this is that many of the elevated risks are nowhere near solid enough for individuals to make life choices based on them — maybe for human resources departments wanting to reduce the total exposure of their employee portfolio.

    And, as your conclusion implies, for public policy purposes it’s often better not to have this kind of information. If you think you can figure out who is more likely to be subject to some particularly expensive diseases and who isn’t, the temptation to cut your potential losses can be overwhelming.

    • Keith Humphreys says

      Also, the odds that we will know more about how to treat Alzheimers’ in a half century, when it might start to matter for this little girl, are very high.

    • Warren Terra says

      maybe for human resources departments wanting to reduce the total exposure of their employee portfolio.
      And, as your conclusion implies, for public policy purposes it’s often better not to have this kind of information. If you think you can figure out who is more likely to be subject to some particularly expensive diseases and who isn’t, the temptation to cut your potential losses can be overwhelming.

      My (admittedly vague, secondhand) impression is that we’ve actually got some very solid laws on the books to prevent the scenarios you describe. And with the ACA, much better laws than we had a couple of years ago.

      • J. Michael Neal says

        We’ve got a bunch of laws on the books but it can be almost impossible for any individual to demonstrate that their termination (or inability to get hired in the first place) had anything to do with violating them. One of the first rules of dealing with a Human Resources department is that they will never give you a straight answer to anything and are really good at documenting that they have made decisions for entirely different reasons than they really did.

        I’ve been told that I was turned down for a job with an insurance company because I was taking the actuarial exams and they were afraid I would leave the position I was interviewed for. Think about that.

  2. Brett Bellmore says

    Given even the current half-hearted rate of medical progress, it’s irrational to worry about your 5 year old having a predilection to get Alzheimer’s 50 or more decades from now. What she should be reassured by is the failure to detect any known problems which would crop up in the next twenty years.

  3. Warren Terra says

    Our knowledge of Alzheimers is really rather poor, but as it happens a vast proportion of the efforts that are made in this area are aimed at developing possible prophylactics – and if those efforts succeed (an admittedly big if, with some special problems relating to possible cause/effect problems with our understanding of Alzheimers, but it’s worth noting that people who are neither stupid nor profligate have found results in laboratory models to conduct some massively expensive studies of candidate drugs), knowing that your genetics say you’re at greater risk of Alzheimers becomes very important, potentially decades before any symptoms might appear. It’s not something to worry about with a five-year-old (that’s far, far too early to start watching for symptoms, and we’re at least a decade away from anyone recommending the drugs being studied), but it’s knowledge that’s worth having.

    More generally, this speaks to the problem with 23-and-me, and similar companies: assuming they do a great job of data collection and analysis (and I’ve heard good things about them) they will still struggle to understand the results for you, especially at the bargain-basement prices they charge. Anyone receiving this information for more than entertainment purposes really should get advice from a knowledgeable person.