And so 23andMe has launched in the UK.
For those not familiar with it, 23andMe allows individuals to swab themselves and have their genome analysed, at a cost of £125. The company is offering to generate a report covering about a hundred traits, giving information on a range of potentially important to fun things: the list includes tests for the presence or absence of inherited conditions such as Tay-Sachs and Beta Thalassemia; risk factors relating to things like Alzheimer’s; how much DNA you have in common with Neanderthals; and earwax type.
To be honest, I’d’ve thought that by the time you’ve got £125 to spend on a test like this, you’d probably know all you’d ever want to know about your earwax, but… well, apparently there’s more. Joy.
Anyway: BBC Breakfast invited me to witter on about it the other day. I only got a couple of minutes, and so didn’t get to say much; shamelessly, I’m going to think aloud a little bit here. My basic starting point is that it’s hard to see why the test per se is too big a problem: all else being equal, who would begrudge a person information about himself? All the same, I think that there are questions that are probably worth asking. (NB: in what follows, whenever I mention 23andMe, the point should be taken to cover any company offering a similar service.) So, in no particular order…
1. Getting the information
One of the obvious points to consider about this service is that there is no way to ensure that people who make use of it will be all that well informed about what genetic information means, and what it means for them. There’s no provision for genetic counselling. Now, this can’t be forced on anyone; but I suppose that it might be reasonable to prefer that people are able to talk to someone who can explain what the information reveals as a part of the process – not as an extra that they’d have to seek out for themselves. For example, if you discover that you’re carrying a gene associated with condition C, and it’s a fairly obscure condition – lots of the things mentioned on the website are – then there’s a number of things seem to be important. What does C do to you? What does an increased risk mean, given that any given condition will kill you eventually provided that nothing else kills you first? What can be done about the risk?
One important thing to keep in mind here is the possibility of causing needless worry, and the corresponding possibility of false reassurance. Suppose that you’re found to carry a cancer gene. Even for something like BRCA-1, that doesn’t mean you’ll get cancer: it means that, over the course of your life, you’re quite likely to. But that does depend on a reasonably long life, and “quite likely” isn’t certainty. In that kind of case, prophylactic interventions might still be worth considering; but for other genes and other cancers, there is – one might worry – a risk that people could end up making quite significant alterations to their lives and bodies, possibly exposing themselves to surgical risk, when that’s simply not warranted.
It is, I guess, less likely that learning that you don’t have the gene for x will make you think you’re immortal – but having said that, the risk that a woman might decide that there’s no reason at all to check her breasts because she doesn’t have BRCA-1 is not negligible. It might be, indirectly, dangerous to be allowed to think that you’re in the clear.
2. Why test?
Following on from this, I wonder whether we might ask whether a lot of these tests are worth it anyway. As I said on the telly, if a test is worth doing, it’s worth doing; and that, to me, suggests that it’s the sort of thing that should be offered on the NHS – to targeted slices of the population, at least – anyway. And if the NHS doesn’t think that the test is worth the hassle… well, it might not be.
3. Use of information
I think that this probably unobjectionable enough, as far as it goes. But it seems to me to be doing much the same thing, and a bit less well, than UK Biobank. After all, UKB has access to medical records, and 23andMe, or similar companies, wouldn’t. And, of course, one has to wonder about their motive. It’s possible that the companies are acting wholly altruistically… but they are private enterprises, and the information is valuable. UKB doesn’t have the same motives. I might be hyper-cynical on this; but the question still seems to me to be worth asking.
This is a bit speculative, given the ABI’s moratorium on the use of genetic data in insurance policies – a self-denying ordinance that may not be renewed, but might be, or might be made a statutory requirement. All the same, it’s not crazy to wonder what impact widespread availability of genetic data would have on the insurance market. And one might wonder whether, in time, there would be a kind of expectation that people have had the tests. I can well imagine an economically libertarian future government allowing people to volunteer genetic information to insurers to get lower premia. But in that case, there’d be a tacit suspicion of people who don’t volunteer the info as having something to hide. They might find themselves priced out.
Whether that really matters is a different question. Ahem.
There’s a couple of worries here. The first is related to the fact that genes are weird – they challenge a lot of how we normally think about agency. Here’s the problem: we wouldn’t normally allow a person to look at a sibling’s medical records: that’d be a violation of the sibling’s privacy. But getting a genetic test on yourself means, in effect, getting the results of your identical twin’s genetic record without them ever even knowing it. It gives you slightly less good insight into the genome of other family members – but, especially if they’re closely related, it’s still a pretty good insight once you take into account how closely or distantly related they are.
The other worry has to do with how 23andMe confirms that the name of the address corresponds to the name of the person whose cells are being sequenced. That’s hard to guarantee; and that means that one person could reasonably easily get a genetic readout for another, without that other’s knowledge. Which leads to a worry about…
6. Parentage and reproduction
This isn’t particularly a point about 23andMe; but the point stands that someone making use of genetic sequencing could learn, accidentally, that their biological parentage isn’t what they expected. All they have to do is to send a sample from their parents, and one from themselves, and notice that they have a gene that doesn’t come from either… and… well: they’d learn something they hadn’t banked on. Granted that genes do spontaneously mutate, there could be pretty good grounds to deduce things about your ancestry.
Of course, this only really matters morally if you think that genetic parenthood matters. I’m not convinced that it does. And the fact that it matters to some doesn’t tell us that it matters period: Stoke City remaining in the Premiership matters to some people, but I don’t think it matters sub specie aeternitatis.
All the same: I guess that even if the genetic relationship doesn’t matter in its own right, the sense of deceit might should it turn out that at least one parent has been making a point of keeping information back from you.
One of the tests is for food preference. Really. And this brings me to a much bigger question, to do with the importance that we invest in genetics. We seem to think that genes are important – but it’s not clear what the direction of travel really is. Do we (correctly) think that they’re important because they are, or do they gather whatever importance they have simply because we take them very seriously. (My money’s on the latter, but I don’t have any detailed arguments that I could squeeze in here.)
And suppose you are carrying the gene that makes your pee smell funny after eating asparagus or Sugar Puffs – they test for the former, but not the latter… Well, really, who cares? If you want to know whether you’re going to get smelly wee after asparagus, you could just try eating some asparagus. It’s cheaper.
It won’t tell you anything about your genome – but is it really important that you know that it’s genetic? Really?
There’s probably a genetic explanation for that, too.