Know Thyself - Or Not
November 25, 2013 2:06 PM Subscribe
FDA orders personal genetics company 23andMe to stop selling tests. 'Last Friday, the Food and Drug Administration sent a letter to the most popular personal genomics service, 23andMe, ordering the company to stop selling its genetic testing kits. According to the FDA's letter, the company has been advertising that its tests offer diagnostic information for a variety of human conditions, placing them in the category of a "medical device" and thus within the agency's jurisdiction. Accordingly, the FDA has been working with the company since 2009 to get 23andMe's testing approved. Now, the FDA has apparently run out of patience.'
'This isn't the first run-in between the FDA and personal genomics companies. Back in 2010, the agency sent a series of letters informing these companies that they were selling medical devices (here's a typical example); in response, all but 23andMe left the market.'
'More generally, the argument focuses on whether we have an unfettered right to information about our own bodies when that information is medically actionable. Many people view genetic information as the most personal information available, given that it's a central part of who we are. But the FDA believes some regulation is necessary and is already working to ensure the accuracy of other diagnostic tests (such as blood glucose levels) because the medical implications are so significant.'
'A study in the American Journal of Human Genetics concluded that "there is insufficient scientific evidence" to prove the tests "are useful in measuring genetic risk for common diseases." Also in 2008, a U.S. Department of Health and Human Services advisory committee found "significant gaps in the U.S. system of oversight of genetic testing that can lead to harms."'
'This isn't the first run-in between the FDA and personal genomics companies. Back in 2010, the agency sent a series of letters informing these companies that they were selling medical devices (here's a typical example); in response, all but 23andMe left the market.'
'More generally, the argument focuses on whether we have an unfettered right to information about our own bodies when that information is medically actionable. Many people view genetic information as the most personal information available, given that it's a central part of who we are. But the FDA believes some regulation is necessary and is already working to ensure the accuracy of other diagnostic tests (such as blood glucose levels) because the medical implications are so significant.'
'A study in the American Journal of Human Genetics concluded that "there is insufficient scientific evidence" to prove the tests "are useful in measuring genetic risk for common diseases." Also in 2008, a U.S. Department of Health and Human Services advisory committee found "significant gaps in the U.S. system of oversight of genetic testing that can lead to harms."'
My spittle is totally en route to these folks. Glad to know that I should seek second opinions on any health-related interesting results before flipping my wig.
posted by damehex at 2:14 PM on November 25, 2013
posted by damehex at 2:14 PM on November 25, 2013
What if all I want is to find out how much of my genetic inheritance is Neanderthal? I promise I wouldn't self-medicate based on that information. Nothing beyond a little trepanning to unleash my latent psychic abilities.
posted by Iridic at 2:21 PM on November 25, 2013 [37 favorites]
posted by Iridic at 2:21 PM on November 25, 2013 [37 favorites]
I'm a 23andme subscriber (but really only for the genealogical component). I feel like the information provided by 23andme is never provided as a diagnosis, and they always present stuff as being low risk, average risk, or higher than average-risk. I can't see that someone who receives information telling them that they are >average risk for something can actually do anything about it without consulting a doctor first, so on some level what's the big deal? The doctor you consult in a panic about your 23andme results acts as the safety check here.
On the other hand, failing to provide documents to the FDA does seem like a pretty dumbshit move. I was hoping to talk my Dad into signing up so I could try and track down our mystery heritage on his side, but I'm guessing 23andme will be gone soon. Bummer.
posted by Joh at 2:21 PM on November 25, 2013 [15 favorites]
On the other hand, failing to provide documents to the FDA does seem like a pretty dumbshit move. I was hoping to talk my Dad into signing up so I could try and track down our mystery heritage on his side, but I'm guessing 23andme will be gone soon. Bummer.
posted by Joh at 2:21 PM on November 25, 2013 [15 favorites]
But how else will I know that I'm from northern Europe!?
posted by The Whelk at 2:22 PM on November 25, 2013 [40 favorites]
posted by The Whelk at 2:22 PM on November 25, 2013 [40 favorites]
Aren't there other genetics operations out there? There's 192 countries out there which aren't regulated by the FDA and would take on American customers. And, yeah, this is just more oversight of something innocuous that only serves to drive business out of the country.
posted by crapmatic at 2:26 PM on November 25, 2013 [2 favorites]
posted by crapmatic at 2:26 PM on November 25, 2013 [2 favorites]
But how else will I know that I'm from northern Europe!?
Your excellent healthcare prov...oh, you moved.
posted by jaduncan at 2:27 PM on November 25, 2013 [27 favorites]
Your excellent healthcare prov...oh, you moved.
posted by jaduncan at 2:27 PM on November 25, 2013 [27 favorites]
They've always had language along the lines of "speak with a genetic counselor" to understand your genotyping results. Which struck me simultaneously as a standard CYA, but also passing the buck so goddamned hard in the face of the seriousness of some of the things you can discover.
Like, you want me to talk with my genetic counselor? Oh ok I'll just schedule that in between my appointments with my solicitor and my falconer.
posted by danny the boy at 2:27 PM on November 25, 2013 [92 favorites]
Like, you want me to talk with my genetic counselor? Oh ok I'll just schedule that in between my appointments with my solicitor and my falconer.
posted by danny the boy at 2:27 PM on November 25, 2013 [92 favorites]
But how else will I know that I'm from northern Europe!?
I really should do that National Geographic genetic test thingummy I got for Christmas last year.
posted by Artw at 2:27 PM on November 25, 2013 [3 favorites]
I really should do that National Geographic genetic test thingummy I got for Christmas last year.
posted by Artw at 2:27 PM on November 25, 2013 [3 favorites]
On the other hand before I ever did 23andme I talked with my GP about some genetic testing results and he was completely unhelpful. Like, you're so much on your own regarding your basic health in this country as it is, I guess maybe this isn't even a change.
posted by danny the boy at 2:29 PM on November 25, 2013 [8 favorites]
posted by danny the boy at 2:29 PM on November 25, 2013 [8 favorites]
But how else will I know that I'm from northern Europe!?
It's not the obvious you're looking for, it's the surprise ancestor that no one wanted to talk about.
posted by BWA at 2:34 PM on November 25, 2013 [2 favorites]
It's not the obvious you're looking for, it's the surprise ancestor that no one wanted to talk about.
posted by BWA at 2:34 PM on November 25, 2013 [2 favorites]
Penis enlargement is a medical procedure too. If the FDA can't do anything about the penis enlargement people (obviously there isn't), 23andMe should offshore themselves to wherever the p.e. companies hang out.
posted by jfuller at 2:34 PM on November 25, 2013 [5 favorites]
posted by jfuller at 2:34 PM on November 25, 2013 [5 favorites]
23andMe is playing a dangerous and stupid game. Simply put, they shouldn't be marketing their stuff as a medical device, and it's kind of sickening that they've been getting away with it so long. But they are wealthy and powerful, and the FDA has been treating them with kid gloves because the FDA is heavily into letting industry set the proper bars for themselves. 23andMe isn't even willing to meet that low bar for friendly cooperation, and they insist on marketing. Not much better than shady unregulated nutritional supplement companies, IMHO.
In stark contrast, companies that do actually useful things, such as Foundation Medicine, are actually establishing standards for their lab techniques and following them. They are working closely with the FDA to help understand the types of lab standards necessary. 23andMe is apparently just blowing off the idea of trying to control for errors in the lab.
Anyway, my disgust for 23andMe runs pretty deep. They're incredibly deceptive and not too useful scientifically.
posted by Llama-Lime at 2:34 PM on November 25, 2013 [29 favorites]
In stark contrast, companies that do actually useful things, such as Foundation Medicine, are actually establishing standards for their lab techniques and following them. They are working closely with the FDA to help understand the types of lab standards necessary. 23andMe is apparently just blowing off the idea of trying to control for errors in the lab.
Anyway, my disgust for 23andMe runs pretty deep. They're incredibly deceptive and not too useful scientifically.
posted by Llama-Lime at 2:34 PM on November 25, 2013 [29 favorites]
The $99 23andMe tests felt to me like papal indulgences in the religion of SCIENCE!
posted by 2bucksplus at 2:39 PM on November 25, 2013 [12 favorites]
posted by 2bucksplus at 2:39 PM on November 25, 2013 [12 favorites]
I have used 23andme, keeping firmly in mind the 'entertainment' aspect of it. As to accuracy, at least on the genealogical side, they seem pretty accurate from what I can tell knowing what I know of my family history, which I can actually trace a fair amount back - assuming nobody cheated, ha! - (99% European, 0.1% unassigned): vast majority Scandinavian, a small bit German, smattering of French and trace amounts of Southern European, whatever that is. Apparently I'm also part Neanderthal at 2.6%, though less than the average European at 2.7%.
So, they got that part right, seems to me. What it says about the accuracy of other stuff, I have no idea.
posted by VikingSword at 2:40 PM on November 25, 2013
So, they got that part right, seems to me. What it says about the accuracy of other stuff, I have no idea.
posted by VikingSword at 2:40 PM on November 25, 2013
Does this mean I still metabolize caffeine and alcohol faster than average?
posted by The Whelk at 2:40 PM on November 25, 2013 [2 favorites]
posted by The Whelk at 2:40 PM on November 25, 2013 [2 favorites]
I was gonna use one of these services for my dog, Shakedown, as I'm curious about the breed makeup. Anyone used one?
How stupid do I come off writing this post?
posted by dobbs at 2:40 PM on November 25, 2013 [1 favorite]
How stupid do I come off writing this post?
posted by dobbs at 2:40 PM on November 25, 2013 [1 favorite]
On the other hand, failing to provide documents to the FDA does seem like a pretty dumbshit move
Disruptive! etc. etc.
posted by Jimbob at 2:52 PM on November 25, 2013 [4 favorites]
Disruptive! etc. etc.
posted by Jimbob at 2:52 PM on November 25, 2013 [4 favorites]
I use 23andme mainly for the genealogical side of things - I like having my health results but they weren't my main concern, although it was interesting to see the stuff that seems to correlate with what I know of family medical history. But I've connected with some unknown relatives and have learned more about my ancestral makeup and that's what I wanted to get out of it. It's more entertainment than anything else. You'd have to be pretty silly to look at an online genetic test and use it as a diagnosis - even if only because it doesn't take into account your lifestyle or circumstances.
There are a lot of warnings before you get access to your health reports on 23. I seem to recall having to click the "YES I know this may upset me and YES I know I should discuss the results with my doctor" button several times before finally getting to read them.
They don't market out here so I've never spotted anything dodgy about their marketing - in fact Australian authorities have made displeased noises about DNA testing for some time, along the lines of arguing that we should only be able to get it done by a doctor and they should control our access to the results, which I don't agree with either. I think it's all about control.
OTOH, as Joh said, stupid move to not comply with the FDA.
posted by andraste at 2:53 PM on November 25, 2013 [7 favorites]
There are a lot of warnings before you get access to your health reports on 23. I seem to recall having to click the "YES I know this may upset me and YES I know I should discuss the results with my doctor" button several times before finally getting to read them.
They don't market out here so I've never spotted anything dodgy about their marketing - in fact Australian authorities have made displeased noises about DNA testing for some time, along the lines of arguing that we should only be able to get it done by a doctor and they should control our access to the results, which I don't agree with either. I think it's all about control.
OTOH, as Joh said, stupid move to not comply with the FDA.
posted by andraste at 2:53 PM on November 25, 2013 [7 favorites]
Like, you want me to talk with my genetic counselor? Oh ok I'll just schedule that in between my appointments with my solicitor and my falconer.
I'd like to think you're consulting all of them for the same thing.
"Predisposition to brain tumors, m'lud? Righto. I'll prep a surgical hawk and send it over for a biopsy. You'll want to shave your hair and drape this rabbit haunch over your skull."
posted by Iridic at 2:59 PM on November 25, 2013 [35 favorites]
I'd like to think you're consulting all of them for the same thing.
"Predisposition to brain tumors, m'lud? Righto. I'll prep a surgical hawk and send it over for a biopsy. You'll want to shave your hair and drape this rabbit haunch over your skull."
posted by Iridic at 2:59 PM on November 25, 2013 [35 favorites]
"Like, you want me to talk with my genetic counselor? Oh ok I'll just schedule that in between my appointments with my solicitor and my falconer."
Everything else aside, genetic councilors are actually a real thing, whose services are very much worth considering for a lot of people - especially those in populations at risk for genetic disorders.
posted by Blasdelb at 3:02 PM on November 25, 2013 [7 favorites]
Everything else aside, genetic councilors are actually a real thing, whose services are very much worth considering for a lot of people - especially those in populations at risk for genetic disorders.
posted by Blasdelb at 3:02 PM on November 25, 2013 [7 favorites]
I was gonna use one of these services for my dog, Shakedown, as I'm curious about the breed makeup. Anyone used one?
How stupid do I come off writing this post?
It's a little controversial - some services are more well-regarded than others and the accuracy really depends on your dog (some breeds are harder to differentiate than others, which can lead to random-seeming results). After reading some articles I'd probably choose not to do it - I'm not confident that knowing a dog's heritage is that important for preventing diseases, and I don't think I'd be convinced by any results I got.
posted by muddgirl at 3:03 PM on November 25, 2013 [2 favorites]
How stupid do I come off writing this post?
It's a little controversial - some services are more well-regarded than others and the accuracy really depends on your dog (some breeds are harder to differentiate than others, which can lead to random-seeming results). After reading some articles I'd probably choose not to do it - I'm not confident that knowing a dog's heritage is that important for preventing diseases, and I don't think I'd be convinced by any results I got.
posted by muddgirl at 3:03 PM on November 25, 2013 [2 favorites]
I was gonna use one of these services for my dog, Shakedown, as I'm curious about the breed makeup. Anyone used one?
In sent a test in for my dog last summer and the results seemed right on to me. I never thought about him having mainly a chow genetic base, but when I got the results it suddenly made perfect sense.
I'd say go for it.
posted by Benway at 3:05 PM on November 25, 2013
In sent a test in for my dog last summer and the results seemed right on to me. I never thought about him having mainly a chow genetic base, but when I got the results it suddenly made perfect sense.
I'd say go for it.
posted by Benway at 3:05 PM on November 25, 2013
Coincidentally, I fired my falconer shortly after getting my 23andMe results, which confirmed some long-held suspicions as to my genetic heritage.
My new chauffer complains a lot less and I really enjoy being able to see where we are going.
posted by griphus at 3:05 PM on November 25, 2013 [20 favorites]
My new chauffer complains a lot less and I really enjoy being able to see where we are going.
posted by griphus at 3:05 PM on November 25, 2013 [20 favorites]
My wife and I once met with a genetic counselor. I think it was because we were considered at a slightly higher risk for birth defects with the third kid because of my wife's age. It was a brief, though informative appointment and was covered by insurance. She wasn't some odd, semi-mythical creature.
(The only thing I remember from the appointment is that she was really interested when I told her in response to some question that my paternal grandfather and I both have multiple accessory spleens. In that instance, you could see why she became a genetic counselor.)
posted by Area Man at 3:07 PM on November 25, 2013 [3 favorites]
(The only thing I remember from the appointment is that she was really interested when I told her in response to some question that my paternal grandfather and I both have multiple accessory spleens. In that instance, you could see why she became a genetic counselor.)
posted by Area Man at 3:07 PM on November 25, 2013 [3 favorites]
Yeah - holy hell, I always viewed it as entertainment, not medical gospel. I felt they were pretty transparent -- from a user perspective -- of how this was all more or less "new" research and really "today you have deadly cancer, tomorrow we might figure out it's just nail fungus." edge of qualified research. Was enjoying the heck out of it as a geneological tool too. Hope they get some smarter people there to shape up the shop.
posted by cavalier at 3:09 PM on November 25, 2013 [1 favorite]
posted by cavalier at 3:09 PM on November 25, 2013 [1 favorite]
If you are in a population which might be susceptible to one of the disorders that 23andMe lists on their website, like sickle cell anemia or breast cancer, then your doctor (or a genetic counsellor!) can already order (targeted, scientifically validated) genetic test and discuss the impact of the results with you. What benefit does 23andMe provide by cutting out the middleman and marketing unscientific medical tests directly to the consumer? It does seem like this falls in the purview of the FDA and they were pretty reasonable in asking 23andMe to back up their marketing with science.
posted by muddgirl at 3:12 PM on November 25, 2013 [5 favorites]
posted by muddgirl at 3:12 PM on November 25, 2013 [5 favorites]
This yoinks a potential Xmas present out from under the tree - if I order now, am I throwing cash down the holiday well?
posted by robocop is bleeding at 3:15 PM on November 25, 2013 [4 favorites]
posted by robocop is bleeding at 3:15 PM on November 25, 2013 [4 favorites]
I just did, robocop. Two kits. Guess we'll see.
If I hadn't seen this post, I probably would have put it off forever. Go fig.
posted by tigrrrlily at 3:18 PM on November 25, 2013
If I hadn't seen this post, I probably would have put it off forever. Go fig.
posted by tigrrrlily at 3:18 PM on November 25, 2013
I've been wondering if 23andme would get in any regulatory trouble ever since they started their latest ad campaign on podcasts (you know, the one that goes "helps you and your doctor identify health areas to keep an eye on")
I know it's an entertainment product (and lets be honest, I'm totally annoyed I didn't hop on that bandwagon before they got told off by the FDA) but I don't think you can make the assumption that that's how most people saw it. Talking to my family and friends it doesn't seem that way. I guess they try and make that clear on their webpage, but the advertising certainly doesn't.
posted by quaking fajita at 3:21 PM on November 25, 2013 [1 favorite]
I know it's an entertainment product (and lets be honest, I'm totally annoyed I didn't hop on that bandwagon before they got told off by the FDA) but I don't think you can make the assumption that that's how most people saw it. Talking to my family and friends it doesn't seem that way. I guess they try and make that clear on their webpage, but the advertising certainly doesn't.
posted by quaking fajita at 3:21 PM on November 25, 2013 [1 favorite]
I think the thing isn't so much that people don't know genetic counselors exist, it's that they (or at least me) have no idea how the hell you'd go about speaking to one. (Well, I'd either ask a doctor or, inspired by AskMe, phone a synagogue.)
posted by hoyland at 3:21 PM on November 25, 2013 [4 favorites]
posted by hoyland at 3:21 PM on November 25, 2013 [4 favorites]
If you are in a population which might be susceptible to one of the disorders that 23andMe lists on their website, like sickle cell anemia or breast cancer, then your doctor (or a genetic counsellor!) can already order (targeted, scientifically validated) genetic test and discuss the impact of the results with you. What benefit does 23andMe provide by cutting out the middleman and marketing unscientific medical tests directly to the consumer?
How would you know you are in a population which might be susceptible to one of those disorders? In some cases obviously because you had a lot of relatives who have the same disorder, but sometimes not, in which case its notifying you to check it out if you want. And then you would go see your doctor, who would order the proper test for you.
posted by Joh at 3:22 PM on November 25, 2013 [1 favorite]
How would you know you are in a population which might be susceptible to one of those disorders? In some cases obviously because you had a lot of relatives who have the same disorder, but sometimes not, in which case its notifying you to check it out if you want. And then you would go see your doctor, who would order the proper test for you.
posted by Joh at 3:22 PM on November 25, 2013 [1 favorite]
I think the thing isn't so much that people don't know genetic counselors exist, it's that they (or at least me) have no idea how the hell you'd go about speaking to one.
And also that most people would not already have one on retainer, and that anyone who already HAD a genetic counselor would logically have no use for an outside service like 23andme.
posted by elizardbits at 3:27 PM on November 25, 2013 [5 favorites]
And also that most people would not already have one on retainer, and that anyone who already HAD a genetic counselor would logically have no use for an outside service like 23andme.
posted by elizardbits at 3:27 PM on November 25, 2013 [5 favorites]
I always thought of it like... I am curious about whether I am a CF carrier because there are indications that there might be some in the family history, and other indications that the respiratory problems I've had my whole life are more common for CF carriers, but I'm only curious. I was thinking about paying for this when I had some cash just to see. There's nothing it'd really be useful for, and I'm not intending to have biological children. If I wanted that kind of genetic counseling, I would absolutely go to a professional. On the other hand, I don't want to pay what it would cost to do that just to assuage my curiosity, so I don't. Unfortunately, I do have to agree that I don't think this is a distinction the whole population is capable of making.
posted by Sequence at 3:28 PM on November 25, 2013 [1 favorite]
posted by Sequence at 3:28 PM on November 25, 2013 [1 favorite]
How would you know you are in a population which might be susceptible to one of those disorders?
Because you have access to health care and go to a doctor on an regular basis who does regular screenings (like my gynecologist, who screens for medical history markers for breast and ovarian cancer). If someone can't going to a doctor on a regular basis in the first place, selling unvalidated medical tests to them seems like it would be worse than just not screening them at all. They have no basis for the accuracy of the results, so now they're making a health care decisions based on junk data.
posted by muddgirl at 3:29 PM on November 25, 2013 [2 favorites]
Because you have access to health care and go to a doctor on an regular basis who does regular screenings (like my gynecologist, who screens for medical history markers for breast and ovarian cancer). If someone can't going to a doctor on a regular basis in the first place, selling unvalidated medical tests to them seems like it would be worse than just not screening them at all. They have no basis for the accuracy of the results, so now they're making a health care decisions based on junk data.
posted by muddgirl at 3:29 PM on November 25, 2013 [2 favorites]
I dunno. People can roll their eyes about 23 and me, but I think it points to an interesting cultural desire teamed with a lack of trust in doctors.
If there is all this genetic research going on, people are inevitably going to be curious which of those genes they have, even or especially for non medically–vital things. For instance, my tests showed I am pretty fast-twitch, muscle-wise. There was evidence to this previously: at swim meets I was routinely the first person off the blocks, fast reflexes, etc. It was cool to see it confirmed. I am not going to go to some genetic counsellor or doctor to get tests for fast-twichiness or caffeine metabolism — it isn't *that* important — but I am curious to know.
At the same time, doctors kinda suck in the customer-service department. Not all doctors, obviously, but many, many doctors, especially perhaps those available on vaguely crappy health plans, are overworked and often dismissive. Getting what feels like a fair hearing and personal care out of them can be so hard and so frustrating. So maybe they would miss something. That, too, is a reason I think people are so into this genetic testing. The FDA getting all "medical-device" about the tests does feel a little like medical professions trying to overguard their turf with law instead of with customer service, to me, but I do just have that turn of mind.
posted by dame at 3:30 PM on November 25, 2013 [12 favorites]
If there is all this genetic research going on, people are inevitably going to be curious which of those genes they have, even or especially for non medically–vital things. For instance, my tests showed I am pretty fast-twitch, muscle-wise. There was evidence to this previously: at swim meets I was routinely the first person off the blocks, fast reflexes, etc. It was cool to see it confirmed. I am not going to go to some genetic counsellor or doctor to get tests for fast-twichiness or caffeine metabolism — it isn't *that* important — but I am curious to know.
At the same time, doctors kinda suck in the customer-service department. Not all doctors, obviously, but many, many doctors, especially perhaps those available on vaguely crappy health plans, are overworked and often dismissive. Getting what feels like a fair hearing and personal care out of them can be so hard and so frustrating. So maybe they would miss something. That, too, is a reason I think people are so into this genetic testing. The FDA getting all "medical-device" about the tests does feel a little like medical professions trying to overguard their turf with law instead of with customer service, to me, but I do just have that turn of mind.
posted by dame at 3:30 PM on November 25, 2013 [12 favorites]
Our genetic counselor at summer camp let us stay up late and tell ghost stories.
posted by dr_dank at 3:39 PM on November 25, 2013 [5 favorites]
posted by dr_dank at 3:39 PM on November 25, 2013 [5 favorites]
The FDA is only asking them to stop advertising their service, so if you go and order the test directly, you're just fine.
Also, one of the deceptive things about 23andMe is that they give people the impression that they're getting their genome "sequenced". However, usually sequencing would require that the test reports adjacent bases. Instead 23andMe is doing genotyping chips, which can only report on the existence of known variants, and therefore can't find any novel, never before seen sequence. That's why I kind of chafe at their use of the term "sequence". The technology 23andMe uses is 15-20 years old now, and consists of putting short, known DNA sequences in a grid on a small glass slide. They then extract DNA from your sample, fragment your DNA into small chunks, label it with a due, then hybridized it on the glass slide. Your DNA fragments preferentially locate to matching sequence, so to genotype your DNA they can just measure the amount of DNA on each grid spot after hybridization. For example, if it's known that there are alternate alleles of
ATTAGCCGCGTAACG
ATTAGCCTCGTAACG
Then both sequences have a spot on the chip. But if you have
ATAAGCCTCGTAACG
And that's never been reported before, 23andMe will miss it.
Also, the whole connection between common disease and genetic basis is completely not understood scientifically, and though we have very washy washy correlations in some cases, and the information may be useful in aggregate for scientific understanding of disease, and possibly public health reasons, the utility for individuals is extremely low except in rare cases. (Though those rare cases of high individual utility do exist!)
Anyway, I appreciate that 23andMe is building this big database of literature associations, and a big database of human genotypes, but I do wonder why they get so much attention. It should be a little tiny corner for extreme hobbyists, not a general thing that everybody does. At least right now, IMHO.
posted by Llama-Lime at 3:40 PM on November 25, 2013 [12 favorites]
Also, one of the deceptive things about 23andMe is that they give people the impression that they're getting their genome "sequenced". However, usually sequencing would require that the test reports adjacent bases. Instead 23andMe is doing genotyping chips, which can only report on the existence of known variants, and therefore can't find any novel, never before seen sequence. That's why I kind of chafe at their use of the term "sequence". The technology 23andMe uses is 15-20 years old now, and consists of putting short, known DNA sequences in a grid on a small glass slide. They then extract DNA from your sample, fragment your DNA into small chunks, label it with a due, then hybridized it on the glass slide. Your DNA fragments preferentially locate to matching sequence, so to genotype your DNA they can just measure the amount of DNA on each grid spot after hybridization. For example, if it's known that there are alternate alleles of
ATTAGCCGCGTAACG
ATTAGCCTCGTAACG
Then both sequences have a spot on the chip. But if you have
ATAAGCCTCGTAACG
And that's never been reported before, 23andMe will miss it.
Also, the whole connection between common disease and genetic basis is completely not understood scientifically, and though we have very washy washy correlations in some cases, and the information may be useful in aggregate for scientific understanding of disease, and possibly public health reasons, the utility for individuals is extremely low except in rare cases. (Though those rare cases of high individual utility do exist!)
Anyway, I appreciate that 23andMe is building this big database of literature associations, and a big database of human genotypes, but I do wonder why they get so much attention. It should be a little tiny corner for extreme hobbyists, not a general thing that everybody does. At least right now, IMHO.
posted by Llama-Lime at 3:40 PM on November 25, 2013 [12 favorites]
Our genetic counselor at summer camp let us stay up late and tell ghost stories
And then we practiced on the Genetic Loom.
posted by The Whelk at 3:42 PM on November 25, 2013 [4 favorites]
And then we practiced on the Genetic Loom.
posted by The Whelk at 3:42 PM on November 25, 2013 [4 favorites]
Bizarrely, I was just at the doctor's earlier today, and we were all wondering whether our inheritable diseases results from 23andMe were actually usable. The doctor actually said, when he got to the the end of our 23andMe print-out, after having a fairly detailed discussion of our (largely negative) results, "Oh, huh, it says here it's not FDA approved," and he frowned a bit. This was a printout from a few weeks ago, incidentally, so they've been at least upfront about that. In the end, we decided to go ahead with the usual tests from a proper source, but that was an easy decision when insurance covers it. I must say, I've very much enjoyed the 23andMe data from a recreational point of view, and would never trust anything they reported without a second opinion. But I could easily imagine never bothering to test for stuff with doctors, until noticing a weird 23andMe report, which would cause me to look more thoroughly into it. So there is some value there. On the other hand, they do report some serious stuff -- genes that put you at much higher risk for Alzheimers, Parkinson's, and breast cancer, for instance (all of which have scary "are you sure you want to access this information?" buttons) -- that I admit to being wary of the dangers of false positives, even if you know enough to get a second opinion.
posted by chortly at 3:44 PM on November 25, 2013 [3 favorites]
posted by chortly at 3:44 PM on November 25, 2013 [3 favorites]
If you are in a population which might be susceptible to one of the disorders that 23andMe lists on their website, like sickle cell anemia or breast cancer, then your doctor (or a genetic counsellor!) can already order (targeted, scientifically validated) genetic test and discuss the impact of the results with you. What benefit does 23andMe provide by cutting out the middleman and marketing unscientific medical tests directly to the consumer?
Just as a data point, which I suppose could be used to argue either side: I got pregnant this year and wasn't offered any sort of genetic screening until the horse was quite out of the gate and I was already a few months pregnant. I even had a pre-conception appointment with a doctor who was aware we were going to start trying pretty much immediately, and that my partner was ethnically Jewish (I am not). I talked to my OB later about genetic screening and was told that it was extremely unlikely to be covered by insurance in my case, because I wasn't high-risk (although my partner was), so it'd be 100% out-of-pocket.
We had actually done 23andme a few months before--definitely for entertainment/nerdy interest purposes and not for medical reasons--and while I can say that we certainly didn't make our conception timing decision based on the "all clear" in terms of non-overlapping carrier status, I don't know that I would have considered a bad thing if we had. Given that 50% of births in the U.S. are unplanned, and only a small fraction of the remaining planned pregnancies are likely to have a woman who qualifies for insurance coverage of genetic testing, I tend to think that 23andme is really unlikely to be displacing the "proper" protocol of having a doctor order the tests.
posted by iminurmefi at 3:50 PM on November 25, 2013 [2 favorites]
Just as a data point, which I suppose could be used to argue either side: I got pregnant this year and wasn't offered any sort of genetic screening until the horse was quite out of the gate and I was already a few months pregnant. I even had a pre-conception appointment with a doctor who was aware we were going to start trying pretty much immediately, and that my partner was ethnically Jewish (I am not). I talked to my OB later about genetic screening and was told that it was extremely unlikely to be covered by insurance in my case, because I wasn't high-risk (although my partner was), so it'd be 100% out-of-pocket.
We had actually done 23andme a few months before--definitely for entertainment/nerdy interest purposes and not for medical reasons--and while I can say that we certainly didn't make our conception timing decision based on the "all clear" in terms of non-overlapping carrier status, I don't know that I would have considered a bad thing if we had. Given that 50% of births in the U.S. are unplanned, and only a small fraction of the remaining planned pregnancies are likely to have a woman who qualifies for insurance coverage of genetic testing, I tend to think that 23andme is really unlikely to be displacing the "proper" protocol of having a doctor order the tests.
posted by iminurmefi at 3:50 PM on November 25, 2013 [2 favorites]
The FDA is only asking them to stop advertising their service, so if you go and order the test directly, you're just fine.
Well, they can't market it, and since the complaints about the marketing are in regards to the website, they can't leave their website up as-is. So maybe they could get away with just having an order form up, but that's about it.
posted by smackfu at 3:51 PM on November 25, 2013
Well, they can't market it, and since the complaints about the marketing are in regards to the website, they can't leave their website up as-is. So maybe they could get away with just having an order form up, but that's about it.
posted by smackfu at 3:51 PM on November 25, 2013
The FDA is only asking them to stop advertising their service, so if you go and order the test directly, you're just fine.
"Therefore, 23andMe must immediately discontinue marketing the PGS until such time as it receives FDA marketing authorization for the device."
The news types are interpreting this as meaning the company can no longer sell the product. IANAL, so whether they're right or you're right is not for me to say.
posted by IndigoJones at 3:53 PM on November 25, 2013
"Therefore, 23andMe must immediately discontinue marketing the PGS until such time as it receives FDA marketing authorization for the device."
The news types are interpreting this as meaning the company can no longer sell the product. IANAL, so whether they're right or you're right is not for me to say.
posted by IndigoJones at 3:53 PM on November 25, 2013
I think the thing isn't so much that people don't know genetic counselors exist, it's that they (or at least me) have no idea how the hell you'd go about speaking to one.
I went to one in a genetic counseling clinic at a major, nationally-ranked medical school who tried to convince me that the fact that my father's sister had died of hemochromatosis which was only diagnosed posthumously, and that my father had died in his fifties after a mysterious long decline culminating in a fatal pancreatic hemorrhage, made it less likely that I had hemochromatosis than a member of the general population chosen at random.
I wasn't quite sure what to say to him, either.
posted by jamjam at 3:55 PM on November 25, 2013 [3 favorites]
I went to one in a genetic counseling clinic at a major, nationally-ranked medical school who tried to convince me that the fact that my father's sister had died of hemochromatosis which was only diagnosed posthumously, and that my father had died in his fifties after a mysterious long decline culminating in a fatal pancreatic hemorrhage, made it less likely that I had hemochromatosis than a member of the general population chosen at random.
I wasn't quite sure what to say to him, either.
posted by jamjam at 3:55 PM on November 25, 2013 [3 favorites]
"If there is all this genetic research going on, people are inevitably going to be curious which of those genes they have, even or especially for non medically–vital things. For instance, my tests showed I am pretty fast-twitch, muscle-wise. There was evidence to this previously: at swim meets I was routinely the first person off the blocks, fast reflexes, etc. It was cool to see it confirmed. I am not going to go to some genetic counsellor or doctor to get tests for fast-twichiness or caffeine metabolism — it isn't *that* important — but I am curious to know."
23 and me can't actually tell you this without you having to trust them on a lot of shit they're being very evasive with the FDA about. They do not actually sequence your genome, only using other tests to find single nucleotide variants that their tests have previously maybe shown correlations to things. You don't get back a sequence anyone can do their own independent analysis with, just the read out they give you that they are being pretty shady about verifying.
posted by Blasdelb at 3:55 PM on November 25, 2013 [3 favorites]
23 and me can't actually tell you this without you having to trust them on a lot of shit they're being very evasive with the FDA about. They do not actually sequence your genome, only using other tests to find single nucleotide variants that their tests have previously maybe shown correlations to things. You don't get back a sequence anyone can do their own independent analysis with, just the read out they give you that they are being pretty shady about verifying.
posted by Blasdelb at 3:55 PM on November 25, 2013 [3 favorites]
...because I wasn't high-risk (although my partner was)...
So anyone know what is the official Genetic Science stance on a Jew and a gentile having a kid? I know of so few mixed relationships between people who are planning to have kids that all I know about the situation is "less risk than two Jews," and this has recently gone from hypothetical to practical concern.
posted by griphus at 3:59 PM on November 25, 2013 [1 favorite]
So anyone know what is the official Genetic Science stance on a Jew and a gentile having a kid? I know of so few mixed relationships between people who are planning to have kids that all I know about the situation is "less risk than two Jews," and this has recently gone from hypothetical to practical concern.
posted by griphus at 3:59 PM on November 25, 2013 [1 favorite]
(Also my family medical history may as well be a photo of Trofim Lysenko shrugging.)
posted by griphus at 4:01 PM on November 25, 2013 [4 favorites]
posted by griphus at 4:01 PM on November 25, 2013 [4 favorites]
are you asking for recipes or is this a general inquiry on whether or not delicious tender babyflesh is kosher
posted by elizardbits at 4:03 PM on November 25, 2013 [2 favorites]
posted by elizardbits at 4:03 PM on November 25, 2013 [2 favorites]
As a biologist working with genomics, the 23 and me stuff would be entertaining but mostly untrustworthy for an actual medical thing. Like everyone else has said, it doesn't actually sequence your DNA, it just finds variations (whose sequences they are presumably not telling you) linked to X, Y, Z. Which is not something that you can really work with, especially if they're not giving you any of the actual statistics, papers, etc.
tl;dr yeah I can see how this is shady and deceptive while still being biologically "sound" in that they're not making data up -- it's just not terribly helpful data, even for a trained professional.
posted by aperturescientist at 4:04 PM on November 25, 2013 [3 favorites]
tl;dr yeah I can see how this is shady and deceptive while still being biologically "sound" in that they're not making data up -- it's just not terribly helpful data, even for a trained professional.
posted by aperturescientist at 4:04 PM on November 25, 2013 [3 favorites]
elizardbits: "And also that most people would not already have one on retainer, and that anyone who already HAD a genetic counselor would logically have no use for an outside service like 23andme."
That's the joke. </wolfcastle>
posted by danny the boy at 4:04 PM on November 25, 2013
That's the joke. </wolfcastle>
posted by danny the boy at 4:04 PM on November 25, 2013
Llama-Lime: "Also, one of the deceptive things about 23andMe is that they give people the impression that they're getting their genome "sequenced"."
Yeah but whose fault is that? They never use the term "sequencing" in any of their materials and before I signed up one of the first things they explained is the difference between sequencing (what they don't do) and genotyping (what they do).
posted by danny the boy at 4:07 PM on November 25, 2013 [1 favorite]
Yeah but whose fault is that? They never use the term "sequencing" in any of their materials and before I signed up one of the first things they explained is the difference between sequencing (what they don't do) and genotyping (what they do).
posted by danny the boy at 4:07 PM on November 25, 2013 [1 favorite]
A comment from the logged-in users-only FDA letter thread on 23andme.com:
"A little perspective, folks (I work in the medical device industry in Regulatory Affairs):
1. In order to market and sell a medical product in the US, there must be sufficient evidence of both safety & efficacy. The concern here is that 23andMe has not fully validated their processes to limit false positive and false negative results. The FDA letter lays out some examples of adverse outcomes from false readings, and they are worrisome (unneeded medical procedures or incorrect medications can do immense harm). 23andMe doesn't have to eliminate those false rates, but they will need to demonstrate what they are and (at a minimum) provide adequate warning and education around them to protect the consumer.
2. This isn't a cash grab. A 510(k) filing currently costs about $5000 plus a few person-weeks of work to complete. Since the letter indicates that 23andMe has submitted two filings and received feedback on each, the issues appear to be around the lack of safety/efficacy data and quality controls. Any money spent on that will go to the test firms, not to the Agency. So if anything, this will promote the business of test labs, not line the pockets of the FDA.
3. The outcome (cease & desist, seizure, etc.) will hinge largely upon what plans the company puts in place and how it responds to the Agency within 15 days. Unfortunately it would appear that they haven't handled things well so far. I've seen many of these letters, and this one suggests an unusual degree of poor behavior on the part of the company.
4. In no way does this represent an attempt to shut down genetic testing or services like 23andMe. It does demonstrate that the regulators are paying attention and expect an adequate level of safety & efficacy data. Given how long the file was in process (~15 months, compared to 3-6 for a lot of other devices), I would suggest the Agency has spent a lot of time trying to negotiate a resolution before resorting to the Warning Letter.
All that said, it would have been nice if 23andMe had provided ongoing guidance and warning to its customers, and I personally believe they should now immediately provide guidance as to what will happen with our data and samples that are currently in process."
The rest of the thread, in my view, is a war between sane people and "Thanks, Obama" nuts.
Some of my favorite comments in the thread stem prom persons who appear to have adopted the perspective that the best way to dissuade the FDA from penalizing 23andme for marketing a set of data which could, by ill-advised persons, be used to generate unwarranted medical treatment is to narrate how they saved thousands of dollars by skipping doctor-administered medical testing and instead used their test results to proceed immediately to various treatments.
posted by mwhybark at 4:10 PM on November 25, 2013 [22 favorites]
"A little perspective, folks (I work in the medical device industry in Regulatory Affairs):
1. In order to market and sell a medical product in the US, there must be sufficient evidence of both safety & efficacy. The concern here is that 23andMe has not fully validated their processes to limit false positive and false negative results. The FDA letter lays out some examples of adverse outcomes from false readings, and they are worrisome (unneeded medical procedures or incorrect medications can do immense harm). 23andMe doesn't have to eliminate those false rates, but they will need to demonstrate what they are and (at a minimum) provide adequate warning and education around them to protect the consumer.
2. This isn't a cash grab. A 510(k) filing currently costs about $5000 plus a few person-weeks of work to complete. Since the letter indicates that 23andMe has submitted two filings and received feedback on each, the issues appear to be around the lack of safety/efficacy data and quality controls. Any money spent on that will go to the test firms, not to the Agency. So if anything, this will promote the business of test labs, not line the pockets of the FDA.
3. The outcome (cease & desist, seizure, etc.) will hinge largely upon what plans the company puts in place and how it responds to the Agency within 15 days. Unfortunately it would appear that they haven't handled things well so far. I've seen many of these letters, and this one suggests an unusual degree of poor behavior on the part of the company.
4. In no way does this represent an attempt to shut down genetic testing or services like 23andMe. It does demonstrate that the regulators are paying attention and expect an adequate level of safety & efficacy data. Given how long the file was in process (~15 months, compared to 3-6 for a lot of other devices), I would suggest the Agency has spent a lot of time trying to negotiate a resolution before resorting to the Warning Letter.
All that said, it would have been nice if 23andMe had provided ongoing guidance and warning to its customers, and I personally believe they should now immediately provide guidance as to what will happen with our data and samples that are currently in process."
The rest of the thread, in my view, is a war between sane people and "Thanks, Obama" nuts.
Some of my favorite comments in the thread stem prom persons who appear to have adopted the perspective that the best way to dissuade the FDA from penalizing 23andme for marketing a set of data which could, by ill-advised persons, be used to generate unwarranted medical treatment is to narrate how they saved thousands of dollars by skipping doctor-administered medical testing and instead used their test results to proceed immediately to various treatments.
posted by mwhybark at 4:10 PM on November 25, 2013 [22 favorites]
"Genetic data on a massive scale is likely to be an extremely valuable commodity to pharmaceutical companies, hospitals, and even governments. This is where the real growth potential is."
To me, the more concerning part is that 23andme is really about taking your data and doing...something...with it in the future.
You pay them to analyze this stuff and then they eventually take all this data and sell it. Makes me a bit squicky.
posted by atomicstone at 4:11 PM on November 25, 2013 [8 favorites]
To me, the more concerning part is that 23andme is really about taking your data and doing...something...with it in the future.
You pay them to analyze this stuff and then they eventually take all this data and sell it. Makes me a bit squicky.
posted by atomicstone at 4:11 PM on November 25, 2013 [8 favorites]
That comment from the guy in regulatory affairs is spot on. Great inside view of the potential negotiations and flagging the longer process.
posted by Ironmouth at 4:14 PM on November 25, 2013
posted by Ironmouth at 4:14 PM on November 25, 2013
Systematic evaluation of personal genome services for Japanese individuals
CONFLICT OF INTEREST
Dr Atul Butte is a founder and advisor to Personalis, a company
providing clinical genome analysis and interpretation.
posted by VikingSword at 4:16 PM on November 25, 2013
CONFLICT OF INTEREST
Dr Atul Butte is a founder and advisor to Personalis, a company
providing clinical genome analysis and interpretation.
posted by VikingSword at 4:16 PM on November 25, 2013
To me, the more concerning part is that 23andme is really about taking your data and doing...something...with it in the future.
Yeah, I had them do whatever their mojo was to my spit way back when. And then they changed their privacy policy in December 2011 to allow 23andMe "to preserve and disclose any and all Personal Information to law enforcement agencies or others if required to do so by law or in the good faith belief that such preservation or disclosure is reasonably necessary to..." $REASONS.
I didn't want to get into an argument with them about what their good faith beliefs about sharing my info might entail. I did briefly enjoy the Ashkenazi pile-on that I got to be a part of. I do not find this recent series of events surprising.
posted by jessamyn at 4:17 PM on November 25, 2013 [2 favorites]
Yeah, I had them do whatever their mojo was to my spit way back when. And then they changed their privacy policy in December 2011 to allow 23andMe "to preserve and disclose any and all Personal Information to law enforcement agencies or others if required to do so by law or in the good faith belief that such preservation or disclosure is reasonably necessary to..." $REASONS.
I didn't want to get into an argument with them about what their good faith beliefs about sharing my info might entail. I did briefly enjoy the Ashkenazi pile-on that I got to be a part of. I do not find this recent series of events surprising.
posted by jessamyn at 4:17 PM on November 25, 2013 [2 favorites]
Blasdelb: "Everything else aside, genetic councilors are actually a real thing, whose services are very much worth considering for a lot of people - especially those in populations at risk for genetic disorders."
Absolutely. I mean, the order of operations for me was: oh here's a thing that I just found out is in my family, I'm going to see my doctor to do some real actual testing. Got results, decided not to make an appointment with a counselor because it really only matters when I start having kids, and then only sort of.
Then I got more curious and I signed up with 23andme, for entertainment. If it said anything really interesting, I would have followed up.
My disbelief/amusement is just that huge disconnect that results from them hugging that fence between "medical device" and "entertainment" so tightly. Like, you're gonna tell me I'm a gonner for sure, but you know, go talk to someone else about it cus we're not responsible. And the absurdity of the implication that sure, we all already have someone lined up to talk to about this kind of stuff.
posted by danny the boy at 4:18 PM on November 25, 2013 [1 favorite]
Absolutely. I mean, the order of operations for me was: oh here's a thing that I just found out is in my family, I'm going to see my doctor to do some real actual testing. Got results, decided not to make an appointment with a counselor because it really only matters when I start having kids, and then only sort of.
Then I got more curious and I signed up with 23andme, for entertainment. If it said anything really interesting, I would have followed up.
My disbelief/amusement is just that huge disconnect that results from them hugging that fence between "medical device" and "entertainment" so tightly. Like, you're gonna tell me I'm a gonner for sure, but you know, go talk to someone else about it cus we're not responsible. And the absurdity of the implication that sure, we all already have someone lined up to talk to about this kind of stuff.
posted by danny the boy at 4:18 PM on November 25, 2013 [1 favorite]
griphus: So anyone know what is the official Genetic Science stance on a Jew and a gentile having a kid?
Both my OB and primary care physician, who tend to be very up-to-date on the latest child-conceiving and child-bearing recommendations from the national groups, said that pre-conception genetic screening is only recommended based on the risk factors of the woman who will be bearing the child. So, if I (female) had been Jewish and my partner (male) gentile, that would have kicked me into the group that is recommended to get certain pre-conception genetic screening--and only if I was found to be a carrier would my partner then be tested. Since I wasn't Jewish, his ancestry was considered non-relevant.
Of course, I am sure there are doctors out there who would have had a more interventionist stance and would have written down whatever was needed on my chart to justify testing in terms of insurance coverage; on balance, I'm glad both of my docs are the sort who don't over-test "just in case" but it was unexpected. I was pretty sure I was going to get the recommendation to be tested before we started trying because of my partner's background and that was not at all the case, his risk factors were considered basically irrelevant.
posted by iminurmefi at 4:19 PM on November 25, 2013 [1 favorite]
Both my OB and primary care physician, who tend to be very up-to-date on the latest child-conceiving and child-bearing recommendations from the national groups, said that pre-conception genetic screening is only recommended based on the risk factors of the woman who will be bearing the child. So, if I (female) had been Jewish and my partner (male) gentile, that would have kicked me into the group that is recommended to get certain pre-conception genetic screening--and only if I was found to be a carrier would my partner then be tested. Since I wasn't Jewish, his ancestry was considered non-relevant.
Of course, I am sure there are doctors out there who would have had a more interventionist stance and would have written down whatever was needed on my chart to justify testing in terms of insurance coverage; on balance, I'm glad both of my docs are the sort who don't over-test "just in case" but it was unexpected. I was pretty sure I was going to get the recommendation to be tested before we started trying because of my partner's background and that was not at all the case, his risk factors were considered basically irrelevant.
posted by iminurmefi at 4:19 PM on November 25, 2013 [1 favorite]
I did 23andme mostly for the ancestry and relative-search aspects of it, so it will be mildly annoying if it goes away, since it is quite useful for that (especially since I got both parents to do it as well.) But if you are going to market medical info, then dissing the FDA is not the game to play. Pretty stupid of them.
posted by tavella at 4:20 PM on November 25, 2013
posted by tavella at 4:20 PM on November 25, 2013
I'd make a joke about how know we know what the Googlebarges are for except I guess Sergey and Anne have split up.
posted by mwhybark at 4:21 PM on November 25, 2013
posted by mwhybark at 4:21 PM on November 25, 2013
Because you have access to health care and go to a doctor on an regular basis who does regular screenings (like my gynecologist, who screens for medical history markers for breast and ovarian cancer). If someone can't going to a doctor on a regular basis in the first place, selling unvalidated medical tests to them seems like it would be worse than just not screening them at all. They have no basis for the accuracy of the results, so now they're making a health care decisions based on junk data.
Is that typical? I'm not asking to be fighty, but I don't think my doctor has done this for me. Unless she has and I just don't remember it because it was all routine and results were negative so I've forgotten all about it.
posted by Joh at 4:22 PM on November 25, 2013
Is that typical? I'm not asking to be fighty, but I don't think my doctor has done this for me. Unless she has and I just don't remember it because it was all routine and results were negative so I've forgotten all about it.
posted by Joh at 4:22 PM on November 25, 2013
Just ordered this for my family as a Chanukah present, and we're doing it for fun as well. The medical stuff is interesting, but I'm pretty sure you're meant to go to a doctor for an actual diagnosis.
Sometimes the FDA is a total buzzkill.
posted by Ruthless Bunny at 4:22 PM on November 25, 2013 [1 favorite]
Sometimes the FDA is a total buzzkill.
posted by Ruthless Bunny at 4:22 PM on November 25, 2013 [1 favorite]
I'm curious, can 23andme tell me if I have Native American heritage?
posted by hellphish at 4:24 PM on November 25, 2013 [1 favorite]
posted by hellphish at 4:24 PM on November 25, 2013 [1 favorite]
You don't get back a sequence anyone can do their own independent analysis with, just the read out they give you that they are being pretty shady about verifying.You definitely DO get back a sequence of SNPs you can do an independent analysis with. It may take looking at all the menus, but it's not hard to download a million SNPs and run them through Promethease. snpedia.com has links on each SNP page back to your raw 23andme data.
I'm generally contrary to economic libertarians in favor of government regulation, but as a 23andme user, I feel I'm educated enough to know how to use this data. Yes, largely for entertainment--which includes self-education--but while I wouldn't make major medical decisions on this basis, I feel as if, once one understands the limitations of both the accuracy and current knowledge--this adds to my self-knowledge.
So anyone know what is the official Genetic Science stance on a Jew and a gentile having a kid? I know of so few mixed relationships between people who are planning to have kids that all I know about the situation is "less risk than two Jews," and this has recently gone from hypothetical to practical concern.As an offspring of two Jews who knows quite a few other such, I am little befuddled that this perception of risk is out there. There are a few potentially dangerous (but known) variants out there, notably Tay-Sachs, but otherwise not a problem.
posted by Schmucko at 4:26 PM on November 25, 2013 [6 favorites]
My fun is in my Neanderthalniss: I'm in the 97th percentile of cavemanliness! That's right: I'm more subhuman than you.
I'll have the duck a l'orange please.
posted by mwhybark at 4:27 PM on November 25, 2013 [3 favorites]
I'll have the duck a l'orange please.
posted by mwhybark at 4:27 PM on November 25, 2013 [3 favorites]
I'm curious, can 23andme tell me if I have Native American heritage?
yes.
posted by mwhybark at 4:28 PM on November 25, 2013 [1 favorite]
yes.
posted by mwhybark at 4:28 PM on November 25, 2013 [1 favorite]
My parents used Family Tree DNA for geneological/ancestry type info (totally fascinating!). FTDNA doesn't do medical advice/outlines, at all. You can transfer your 23 and me results to them to get the advantage of a their big database of people if you'd like or are concerned about the future of 23andme.
posted by julen at 4:33 PM on November 25, 2013 [3 favorites]
posted by julen at 4:33 PM on November 25, 2013 [3 favorites]
The US labor and health care systems are still set up to require people not to know what sorts of things they might be at risk for (so as to prevent an obligation to disclose and thereby make oneself uninsurable / unemployable). The protections from GINA are totally insufficient under these circumstances. That, plus the general risk from (1) people taking 23andme's results as a diagnosis and harming themselves either directly or indirectly as a result (2) the massive risk of privacy violation has always made the site seem like a walking lawsuit to me. I've been amazed, the few times I've looked there, that their lawyers deem their boilerplate EULA language adequate.
posted by gerryblog at 4:33 PM on November 25, 2013
posted by gerryblog at 4:33 PM on November 25, 2013
I did 23andMe when it went down to $100, purely as a "ooh, neat" thing, and so I could get back the raw data dump and run it through other analysis services/software.
My half-sister (through my father) also got hers done, and it was neat for the data to say "Hey, you're related to this person" without us telling them that, although it says "through grandmother" and I don't know why. Maybe it would be more accurate if we had more family members' data.
posted by mrbill at 4:34 PM on November 25, 2013 [3 favorites]
My half-sister (through my father) also got hers done, and it was neat for the data to say "Hey, you're related to this person" without us telling them that, although it says "through grandmother" and I don't know why. Maybe it would be more accurate if we had more family members' data.
posted by mrbill at 4:34 PM on November 25, 2013 [3 favorites]
Yes, hellphish. In fact, probably the most fun bit of data I got from 23andme was that that my two stretches of Native American DNA were shared with six people on the site (at least of those who cross-shared me with me), five of which could trace their ancestry back to John Tabor and his wife Hagar Stovall, in two cases by the other son from me, making it extremely likely that Hagar was Benjamin Stovall's out of wedlock daughter with a fellow indentured servant who happened to be Powhatan Indian.
I do love genetic detective work.
posted by tavella at 4:35 PM on November 25, 2013 [6 favorites]
I do love genetic detective work.
posted by tavella at 4:35 PM on November 25, 2013 [6 favorites]
I wonder if this is one of those situations where the regulations are sort of for the lowest common denominator. Sure, us internet savvy, jacks of all trades know that this test is not a medical device. But does the grandma who hears about it in Reader's Digest? I bet not. No matter how many times you beat some people over the head with "THIS IS NOT A MEDICAL DEVICE" warnings, they will only believe what they want to.
I base this on general experience, and the following telephone conversation I overheard once in the grocery store. A woman says "do we need cereal? No? OK. Wait, you have a cholesterol test tomorrow, I'll get you some Cheerios."
These are the idiots who are ruining it for the rest of us.
posted by gjc at 4:41 PM on November 25, 2013 [4 favorites]
I base this on general experience, and the following telephone conversation I overheard once in the grocery store. A woman says "do we need cereal? No? OK. Wait, you have a cholesterol test tomorrow, I'll get you some Cheerios."
These are the idiots who are ruining it for the rest of us.
posted by gjc at 4:41 PM on November 25, 2013 [4 favorites]
Why would anyone trust their genomic data to an unregulated commercial enterprise?
And people worry about what they reveal on Facebook!
posted by spitbull at 4:45 PM on November 25, 2013 [11 favorites]
And people worry about what they reveal on Facebook!
posted by spitbull at 4:45 PM on November 25, 2013 [11 favorites]
I was really only interested in the genealogical info and man, it was just the most boring, predictable tang ever. Mat line- most common in Europe. Pat Line- most common in Northern Europe. Possible relatives? Anyone who might have invaded the British isles in the last thousand years.
The only interesting in in my Mom's family tree hunting was finding a case of straight up, don't change the family name incestous marriage in the early 19th Century.
Isolated Scottish fishing villages are scary y'all.
posted by The Whelk at 4:52 PM on November 25, 2013 [2 favorites]
The only interesting in in my Mom's family tree hunting was finding a case of straight up, don't change the family name incestous marriage in the early 19th Century.
Isolated Scottish fishing villages are scary y'all.
posted by The Whelk at 4:52 PM on November 25, 2013 [2 favorites]
Both my OB and primary care physician, who tend to be very up-to-date on the latest child-conceiving and child-bearing recommendations from the national groups, said that pre-conception genetic screening is only recommended based on the risk factors of the woman who will be bearing the child. So, if I (female) had been Jewish and my partner (male) gentile, that would have kicked me into the group that is recommended to get certain pre-conception genetic screening--and only if I was found to be a carrier would my partner then be tested. Since I wasn't Jewish, his ancestry was considered non-relevant.
Well, sheeeeeeet.
I am an Ashkenazi Jew (45.7% according to 23andme) who was not recommended for any screenings at my pre-conception check-up. And I never thought to ask. Of course, I pass for pretty not-Jewish. Maybe my doctor didn't think I was a Jew? Story of my life.
But according to 23andme neither my partner--47.1% Ashkenazi Jewish--nor I carry any of the Ashkenazi Jewish Genetic Diseases so I don't feel terribly worried. Ironically.
posted by PhoBWanKenobi at 5:02 PM on November 25, 2013 [1 favorite]
Well, sheeeeeeet.
I am an Ashkenazi Jew (45.7% according to 23andme) who was not recommended for any screenings at my pre-conception check-up. And I never thought to ask. Of course, I pass for pretty not-Jewish. Maybe my doctor didn't think I was a Jew? Story of my life.
But according to 23andme neither my partner--47.1% Ashkenazi Jewish--nor I carry any of the Ashkenazi Jewish Genetic Diseases so I don't feel terribly worried. Ironically.
posted by PhoBWanKenobi at 5:02 PM on November 25, 2013 [1 favorite]
I'm curious, can 23andme tell me if I have Native American heritage?
My ex got this done last spring, and was fully expecting his results to have Native American in them, but nope -- only Northern/Western Europe. His dad was crushed. Family legends blah blah blah.
So yeah, I think it can tell you if you have Native American heritage.
Personally, I don't get the draw of these "heritage" tests, they creep me out.
posted by rue72 at 5:16 PM on November 25, 2013
My ex got this done last spring, and was fully expecting his results to have Native American in them, but nope -- only Northern/Western Europe. His dad was crushed. Family legends blah blah blah.
So yeah, I think it can tell you if you have Native American heritage.
Personally, I don't get the draw of these "heritage" tests, they creep me out.
posted by rue72 at 5:16 PM on November 25, 2013
Why is it always 1/16th Cherokee that everyone thinks they might be, and no other amount, anyway?
posted by Artw at 5:20 PM on November 25, 2013
posted by Artw at 5:20 PM on November 25, 2013
rue72, the other amusing thing thing was that like half the people who matched with me on those bits of Native American DNA, and I believe in all cases they were the only stretches they had, had Cherokee princess type family stories, but a) they were all different and b) just about the only possible ancestral confederation was Powhatan, not Cherokee.
posted by tavella at 5:24 PM on November 25, 2013 [1 favorite]
posted by tavella at 5:24 PM on November 25, 2013 [1 favorite]
It's a good question, Artw. Maybe they were seen as more tragically romantic? When I was verifying family genealogies against real data, it was always the Scot-Irish branches that were the most ridiculous, which all sorts of random noblemen glued on top. Maybe the same lost-cause type romanticism.
posted by tavella at 5:27 PM on November 25, 2013 [1 favorite]
posted by tavella at 5:27 PM on November 25, 2013 [1 favorite]
I would think that the FDA would be pushing 23andme because how else will the NSA learn everything they want to know about us?
posted by tafetta, darling! at 5:27 PM on November 25, 2013 [2 favorites]
posted by tafetta, darling! at 5:27 PM on November 25, 2013 [2 favorites]
Why is it always 1/16th Cherokee that everyone thinks they might be, and no other amount, anyway?
Cause it's kind of memetic unit and explains why you don't look. Heroine at all? I don't know, my family came to the US in the 20s and basically never left the NYC metro area. I know not your western ways.
posted by The Whelk at 5:28 PM on November 25, 2013
Cause it's kind of memetic unit and explains why you don't look. Heroine at all? I don't know, my family came to the US in the 20s and basically never left the NYC metro area. I know not your western ways.
posted by The Whelk at 5:28 PM on November 25, 2013
Tay-Sachs disease is devastating and the chance of Ashkenazi Jews being carriers is 1 in 27 compared to the general population risk of 1 in 300. But other populations also have higher risks: French- Canadians, Cajuns, Amish, and Irish. So..... Be aware.
posted by bq at 5:30 PM on November 25, 2013 [2 favorites]
posted by bq at 5:30 PM on November 25, 2013 [2 favorites]
i still have no idea what the "unassigned" is so i just tell people it is xenomorph
i am 6.8% xenomorph
posted by elizardbits at 5:31 PM on November 25, 2013 [15 favorites]
i am 6.8% xenomorph
posted by elizardbits at 5:31 PM on November 25, 2013 [15 favorites]
which seems awfully low all things considered
posted by elizardbits at 5:33 PM on November 25, 2013 [5 favorites]
posted by elizardbits at 5:33 PM on November 25, 2013 [5 favorites]
That explains SO MUCH
posted by The Whelk at 5:33 PM on November 25, 2013 [4 favorites]
posted by The Whelk at 5:33 PM on November 25, 2013 [4 favorites]
The only reason I'd take the test would be to see how the family legend* about our 1/nth share of Muskogee blood shakes out.
*Said legend boils down to "our nth-great grandpappy needed a wife in a hurry for legal advantage and accordingly went where there was a sizable selection of women desperate enough to marry a complete stranger." I'm not sure why you'd make up a legend like that.
posted by Iridic at 5:34 PM on November 25, 2013 [1 favorite]
*Said legend boils down to "our nth-great grandpappy needed a wife in a hurry for legal advantage and accordingly went where there was a sizable selection of women desperate enough to marry a complete stranger." I'm not sure why you'd make up a legend like that.
posted by Iridic at 5:34 PM on November 25, 2013 [1 favorite]
elizardbits: "i am 6.8% xenomorph"
Indeed?
posted by jquinby at 5:34 PM on November 25, 2013 [5 favorites]
Indeed?
posted by jquinby at 5:34 PM on November 25, 2013 [5 favorites]
omg
posted by elizardbits at 5:34 PM on November 25, 2013 [3 favorites]
posted by elizardbits at 5:34 PM on November 25, 2013 [3 favorites]
The only reason I'd take the test would be to see how the family legend* about our 1/nth share of Muskogee blood shakes out.
I read this too quickly and read that as Muggle and now I want that test.
posted by The Whelk at 5:35 PM on November 25, 2013 [3 favorites]
I read this too quickly and read that as Muggle and now I want that test.
posted by The Whelk at 5:35 PM on November 25, 2013 [3 favorites]
If they find that you're high-risk for some disease, are you now obligated to report this to your insurance company? If later on your insurance company finds out, does that count as a pre-existing condition? How does this work?
posted by reformedjerk at 5:36 PM on November 25, 2013
posted by reformedjerk at 5:36 PM on November 25, 2013
I would lie, lie forever, lie always to the insurance company, lie under oath, lie in threat of fire and death and dragons.
I am no one's role model and this is probably bad advice.
posted by elizardbits at 5:37 PM on November 25, 2013 [16 favorites]
I am no one's role model and this is probably bad advice.
posted by elizardbits at 5:37 PM on November 25, 2013 [16 favorites]
I mean really I don't consider it "lying" to withhold potentially factually incorrect information from a large corporation that has everything BUT my personal wellbeing in mind any more than I would consider it "lying" if I tell a mugger that all the money I have on me is the money in my wallet.
posted by elizardbits at 5:41 PM on November 25, 2013 [3 favorites]
posted by elizardbits at 5:41 PM on November 25, 2013 [3 favorites]
My dad was hoping to find out why his Norwegian side of the family has such dark skin and dark curly hair. I think he suspected a Kingsblood Royal situation (that's a Sinclair Lewis deep cut), but no dice.
posted by Area Man at 5:41 PM on November 25, 2013 [2 favorites]
posted by Area Man at 5:41 PM on November 25, 2013 [2 favorites]
For honesty's sake, I'd fax a digest of my results to Blue Cross.
posted by Iridic at 5:45 PM on November 25, 2013 [5 favorites]
posted by Iridic at 5:45 PM on November 25, 2013 [5 favorites]
You know, in some cases the genomic data about contemporary indigenous communities with which people are discovering exciting tiny fragments of genetic kinship, was taken without informed consent?
Basically the Cherokee princess tale is the inverse of the one-drop rule.
posted by spitbull at 5:47 PM on November 25, 2013 [3 favorites]
Basically the Cherokee princess tale is the inverse of the one-drop rule.
posted by spitbull at 5:47 PM on November 25, 2013 [3 favorites]
Also, elizardbits, you don't consider it a problem to trust your genetic profile to another corporation that not only has no reason to advance your interests, but little regulatory oversight (of the sort hospitals and insurance companies face) to safeguard those interests?
What if a consortium of insurance companies simply bought 23andme with all its accumulated data, for example? I don't know the HIPAA status of 23andme's data. But isn't the whole point of FDA regulation to make sure your interests are being considered here? They hardly have a perfect track record of preventing abuse by the medical/Pharma industries, but they're all we've got.
posted by spitbull at 5:56 PM on November 25, 2013 [1 favorite]
What if a consortium of insurance companies simply bought 23andme with all its accumulated data, for example? I don't know the HIPAA status of 23andme's data. But isn't the whole point of FDA regulation to make sure your interests are being considered here? They hardly have a perfect track record of preventing abuse by the medical/Pharma industries, but they're all we've got.
posted by spitbull at 5:56 PM on November 25, 2013 [1 favorite]
my backup plan is somali warlording
posted by elizardbits at 5:59 PM on November 25, 2013 [5 favorites]
posted by elizardbits at 5:59 PM on November 25, 2013 [5 favorites]
Old meme. You want to grab a Syrian oil well and set up a kingdom.
posted by The Whelk at 6:11 PM on November 25, 2013 [1 favorite]
posted by The Whelk at 6:11 PM on November 25, 2013 [1 favorite]
Isolated Scottish fishing villages are scary y'all.
But they brought us you, Mr. The Whelk. We owe them a debt of gratitude.
My wife, who is Ibero-Cuban via the Canary Islands and Basque highlands, very surprisingly came up 1% Native American. This is pretty strange, given that all the Native Americans in Cuba were gone in a century. I think it probably means I need a better understanding of Cuban migration patterns.
posted by mwhybark at 6:12 PM on November 25, 2013 [1 favorite]
But they brought us you, Mr. The Whelk. We owe them a debt of gratitude.
My wife, who is Ibero-Cuban via the Canary Islands and Basque highlands, very surprisingly came up 1% Native American. This is pretty strange, given that all the Native Americans in Cuba were gone in a century. I think it probably means I need a better understanding of Cuban migration patterns.
posted by mwhybark at 6:12 PM on November 25, 2013 [1 favorite]
This is pretty strange, given that all the Native Americans in Cuba were gone in a century. I think it probably means I need a better understanding of Cuban migration patterns.
It also means there are some pretty significant error bars around those values, particularly once you're getting down to single-digit percentages. What makes Native American DNA different from any other DNA? A higher probability of specific sequences in certain areas. That same sequence could arise from random mutation in an individual with no Native American ancestry.
posted by Jimbob at 6:15 PM on November 25, 2013
It also means there are some pretty significant error bars around those values, particularly once you're getting down to single-digit percentages. What makes Native American DNA different from any other DNA? A higher probability of specific sequences in certain areas. That same sequence could arise from random mutation in an individual with no Native American ancestry.
posted by Jimbob at 6:15 PM on November 25, 2013
I got it for $99 DNA day special and registered it under an alias with a throwaway email account.
Are they going to shut down the website with the existing information based on our spittle?
posted by discopolo at 6:20 PM on November 25, 2013
Are they going to shut down the website with the existing information based on our spittle?
posted by discopolo at 6:20 PM on November 25, 2013
Well, they spelled "Ann" without an "e" on the FDA letter. This just isn't her year:-(
I'm sorry to see 23andMe go. I had fun with it. I hope the site stays up for people who already purchased it and have their data gathered.
posted by discopolo at 6:27 PM on November 25, 2013
I'm sorry to see 23andMe go. I had fun with it. I hope the site stays up for people who already purchased it and have their data gathered.
posted by discopolo at 6:27 PM on November 25, 2013
I first started with 23andMe for the ancestry and relative finder. And the health traits section only confirmed what I already new about my family's health history (Celiac Disease! Skin Cancer! A few drug interactions! Can't drink coffee at all due to very slow caffeine metabolism! Allergies! Slow muscle twitch - aka never got picked for any team at school due to being slow. etc.).
For me it has been about entertainment and a way to support my half-sister who is looking for paternal side relatives as her dad was adopted. And it has been a success on both accounts.
If the FDA is going to require 23andMe to shut down, I have already downloaded my raw data and run it through a few other services like Promethease.
Entertainment aside, I do have the right to my data as it is my body. I want to be able to pay for my own tests and get my own data and take charge of what I can control.
posted by msjen at 7:03 PM on November 25, 2013 [4 favorites]
For me it has been about entertainment and a way to support my half-sister who is looking for paternal side relatives as her dad was adopted. And it has been a success on both accounts.
If the FDA is going to require 23andMe to shut down, I have already downloaded my raw data and run it through a few other services like Promethease.
Entertainment aside, I do have the right to my data as it is my body. I want to be able to pay for my own tests and get my own data and take charge of what I can control.
posted by msjen at 7:03 PM on November 25, 2013 [4 favorites]
If I send in a sample from my daughter who has trisomy 21, will their testing rig catch fire?
posted by plinth at 7:33 PM on November 25, 2013 [2 favorites]
posted by plinth at 7:33 PM on November 25, 2013 [2 favorites]
This is all you need to know: if you look like your father, that's heredity; if you look like the milkman (remember those), that's environment.
posted by Napoleonic Terrier at 8:07 PM on November 25, 2013 [2 favorites]
posted by Napoleonic Terrier at 8:07 PM on November 25, 2013 [2 favorites]
23andMe is fundamentally a bet about the nature of this chart of the price of sequencing a human genome. When it was founded in 2006, the slope of progress didn't look so great. Things really kicked in motion at the start of 2008, but we've entered another plateau around the $5k mark. 23andMe can certainly use microarray to put their infrastructure to the test and iron out kinks like the one neil mentions above, but exciting thing coming down the pipeline is whole genome sequencing.
So they really need that graph to continue looking like super Moore's law growth. Cheap data means huge profits for people selling data processing. And right now, costs are not falling very quickly. I don't know if it's because of fundamental limitations of physics, a lagging consequence of the financial crisis of 2008, a change in government policy in a country previously making huge investments in sequencing, or what. Given that their price cut last year came alongside an announcement of another round of VC funding, it's unclear to me how much longer they can stay tread water before sequencing is affordable enough.
posted by pwnguin at 8:09 PM on November 25, 2013
So they really need that graph to continue looking like super Moore's law growth. Cheap data means huge profits for people selling data processing. And right now, costs are not falling very quickly. I don't know if it's because of fundamental limitations of physics, a lagging consequence of the financial crisis of 2008, a change in government policy in a country previously making huge investments in sequencing, or what. Given that their price cut last year came alongside an announcement of another round of VC funding, it's unclear to me how much longer they can stay tread water before sequencing is affordable enough.
posted by pwnguin at 8:09 PM on November 25, 2013
I'm not sure how much faith I'd put in the comprehensiveness of the Ashkenazic genetic screening from 23andme. If you're in a region with a large Jewish population, you may want to see if the Victor Center does screening in your area; it's free (they check if your insurance will cover it, but otherwise they do). I got screened by them, and they appeared to be both thorough and in close touch with the current literature on genetic diseases (e.g. they only include tests which have good evidence behind them). They also provide genetic counseling both before and after testing.
In my own case, I'm 100% Ashkenazic while my wife is of British and Swedish heritage; the biggest possible concern we had was cystic fibrosis, which is comparatively frequent in Scandinavians as well as Jews.
For what it's worth, the Victor Center's recommendations on who should get testing are:
"Who should be screened:
- All Jews of child-bearing age.
- If even only one of your grandparents is of Ashkenazi descent.
- If you are heterosexual or gay and considering getting pregnant through donor egg/sperm.
- If you are part of an interfaith couple, the Jewish partner should be screened first. If he/she is found to be a carrier, the non-Jewish partner should also get screened."
posted by janewman at 8:19 PM on November 25, 2013 [5 favorites]
In my own case, I'm 100% Ashkenazic while my wife is of British and Swedish heritage; the biggest possible concern we had was cystic fibrosis, which is comparatively frequent in Scandinavians as well as Jews.
For what it's worth, the Victor Center's recommendations on who should get testing are:
"Who should be screened:
- All Jews of child-bearing age.
- If even only one of your grandparents is of Ashkenazi descent.
- If you are heterosexual or gay and considering getting pregnant through donor egg/sperm.
- If you are part of an interfaith couple, the Jewish partner should be screened first. If he/she is found to be a carrier, the non-Jewish partner should also get screened."
posted by janewman at 8:19 PM on November 25, 2013 [5 favorites]
My dad recently discovered that his father (my paternal grandfather) was of a completely different ethnic heritage than what my father had always been told. He stumbled across this while doing traditional genealogical research (census records, birth certificates, graveyard ledgers, etc) and sent his DNA in to whatever ancestry.com uses for testing, which confirmed his newly discovered lineage as like 49% of his genetic makeup. So while I've always been skeptical of 23andme, it was cool to have some confirmation that yup, the leap my dad took to find that census record was reasonable. On the other hand, I know that my genomics knowledge is limited at best, but if I send in my DNA, I want the raw data back. I want to process that sucker myself. I'm gonna BLAST those SNPs good.
posted by maryr at 8:35 PM on November 25, 2013
posted by maryr at 8:35 PM on November 25, 2013
23+Me had free testing for African- Americans. This is a group that often wants and needs genetic information ( sickle cell, diabetes risk etc.) while not generally being able to afford it.
Mr. Roquette has wanted to find out his father's actual ancestry. Now it's back to services that run $150+. We really can't handle that.
He's just curious, but curiosity is a valid reason to get tested in my opinion.
I would love to be tested just to verify or debunk family legends.
Same goes for stuff my ex had to say about HIS ancestry. I wish I could get at least one kid to go along with a test.
Now it's yet more out of reach.
Then there's that whole Neandrathal thing....
posted by Katjusa Roquette at 8:38 PM on November 25, 2013
Mr. Roquette has wanted to find out his father's actual ancestry. Now it's back to services that run $150+. We really can't handle that.
He's just curious, but curiosity is a valid reason to get tested in my opinion.
I would love to be tested just to verify or debunk family legends.
Same goes for stuff my ex had to say about HIS ancestry. I wish I could get at least one kid to go along with a test.
Now it's yet more out of reach.
Then there's that whole Neandrathal thing....
posted by Katjusa Roquette at 8:38 PM on November 25, 2013
I went to one in a genetic counseling clinic at a major, nationally-ranked medical school who tried to convince me that the fact that my father's sister had died of hemochromatosis which was only diagnosed posthumously, and that my father had died in his fifties after a mysterious long decline culminating in a fatal pancreatic hemorrhage, made it less likely that I had hemochromatosis than a member of the general population chosen at random.
I wasn't quite sure what to say to him, either.
That could be reasonable, jamjam. I don't know much about hemochromatosis or you, but if it were sex-linked and your paternal aunt and father had it and you were male, you'd have a good chance of not having the disease, because you are guarenteed not to inherit your father's X chromosome. (If you're male, you are XY. The only parent who can provide a Y chromosome is your father. So your X chromosome definitely came from your mother.)
So I don't know everything that you told the counselor and I don't know everything the counselor knew about hemochromatosis, but there could be something there that really was conclusive - or rather that inheriting the disease from your father would have required some super rare event that was less likely than the chance of having a father who was a carrier in the general population.
(This is sort of the opposite case, but... There's a quiz problem that popped up in my Intro Genetics class in undergrads, one of those things you get wrong but, goddammit, it was admittedly a learning experience. If you have two Carriers (Aa and Bb) of a gene and the homogeneous recessive (ab) is lethal, what are the odds of a child being a carrier? The answer in a non-lethal case is 1/2 (there are four options - AB, Ab, aB, & ab - Ab and aB are carriers) but in a lethal case the homogeneous recessive offspring (ab) is never born - therefore the only viable offspring are AB, Ab, and bA - and the chance of the child being a carrier is 2/3. I always thought that problem was a mean trick but I do remember it.)
posted by maryr at 8:48 PM on November 25, 2013 [1 favorite]
I wasn't quite sure what to say to him, either.
That could be reasonable, jamjam. I don't know much about hemochromatosis or you, but if it were sex-linked and your paternal aunt and father had it and you were male, you'd have a good chance of not having the disease, because you are guarenteed not to inherit your father's X chromosome. (If you're male, you are XY. The only parent who can provide a Y chromosome is your father. So your X chromosome definitely came from your mother.)
So I don't know everything that you told the counselor and I don't know everything the counselor knew about hemochromatosis, but there could be something there that really was conclusive - or rather that inheriting the disease from your father would have required some super rare event that was less likely than the chance of having a father who was a carrier in the general population.
(This is sort of the opposite case, but... There's a quiz problem that popped up in my Intro Genetics class in undergrads, one of those things you get wrong but, goddammit, it was admittedly a learning experience. If you have two Carriers (Aa and Bb) of a gene and the homogeneous recessive (ab) is lethal, what are the odds of a child being a carrier? The answer in a non-lethal case is 1/2 (there are four options - AB, Ab, aB, & ab - Ab and aB are carriers) but in a lethal case the homogeneous recessive offspring (ab) is never born - therefore the only viable offspring are AB, Ab, and bA - and the chance of the child being a carrier is 2/3. I always thought that problem was a mean trick but I do remember it.)
posted by maryr at 8:48 PM on November 25, 2013 [1 favorite]
I've refused to do anything like 23andMe because there are things like breast cancer and Alzheimer's in my family history that LALALALALALALALA I CAN'T HEAR YOOOUUUUU.
Obviously I know I have some level of risk, but to find out my risk is increased? Then what? Panic for fifty years? Wait for my boobs to fall off? I just couldn't find out I had an elevated risk, even if false positive, for something and not just freak out about it all the time until I actually got the disease, which would at that point be a bit anti climactic.
posted by sonika at 9:03 PM on November 25, 2013
Obviously I know I have some level of risk, but to find out my risk is increased? Then what? Panic for fifty years? Wait for my boobs to fall off? I just couldn't find out I had an elevated risk, even if false positive, for something and not just freak out about it all the time until I actually got the disease, which would at that point be a bit anti climactic.
posted by sonika at 9:03 PM on November 25, 2013
(Also, for my family background? There's no mystery. There's also no ancestor of mine that ever saw the sun after 3PM in the winter.)
posted by sonika at 9:05 PM on November 25, 2013
posted by sonika at 9:05 PM on November 25, 2013
Obviously I know I have some level of risk, but to find out my risk is increased?
I found out I had a higher than average risk for a condition that runs in my family. I was kind of relieved to know, to be honest, so I could be realistic about thinking of it being in my life. I pay special attention to my doctor's visits and when things come back normal, I'm kind of pleased.
It doesn't feel like a weight on my shoulders. It feels like I can and am handling it.
Though I know lots of ppl who just don't want to know. I wish I were like that but I'm just not. Not knowing makes me anxious. Knowing makes me feel stronger and more in control or at least, more able to plan for the worst case scenario, which pleases me.
posted by discopolo at 10:13 PM on November 25, 2013 [4 favorites]
I found out I had a higher than average risk for a condition that runs in my family. I was kind of relieved to know, to be honest, so I could be realistic about thinking of it being in my life. I pay special attention to my doctor's visits and when things come back normal, I'm kind of pleased.
It doesn't feel like a weight on my shoulders. It feels like I can and am handling it.
Though I know lots of ppl who just don't want to know. I wish I were like that but I'm just not. Not knowing makes me anxious. Knowing makes me feel stronger and more in control or at least, more able to plan for the worst case scenario, which pleases me.
posted by discopolo at 10:13 PM on November 25, 2013 [4 favorites]
I did QA at a big, serious, surgical device company for a few years, with a cube next to the records room. There was an uncomfortable week spent watching three big, serious VP's make copies and sprint them down to the nearby meeting room where the FDA agents were. They would then sprint back with the next request.
Don't mess with the FDA.
posted by underflow at 11:04 PM on November 25, 2013 [5 favorites]
Don't mess with the FDA.
posted by underflow at 11:04 PM on November 25, 2013 [5 favorites]
I'm not a subscriber to 23andme but I wish I was though. These sorts of DNA tests are quite useless for people of Asian decent as most historical data are European.
posted by postertext at 2:01 AM on November 26, 2013
posted by postertext at 2:01 AM on November 26, 2013
23+Me had free testing for African- Americans. This is a group that often wants and needs genetic information ( sickle cell, diabetes risk etc.) while not generally being able to afford it.
Hmmmm, and I thought 23andme did not claim to be offering medical tests. I guess only for poor people?
Many people can't afford many forms of health care. That should not force them into unregulated options or broad, non-specific testing, or trading their most personal data ("free!" In this case is hardly so, since if you're not the customer, you're the product).
The Sickle Cell Foundation offers free legitimate testing, too.
posted by spitbull at 2:25 AM on November 26, 2013 [7 favorites]
Hmmmm, and I thought 23andme did not claim to be offering medical tests. I guess only for poor people?
Many people can't afford many forms of health care. That should not force them into unregulated options or broad, non-specific testing, or trading their most personal data ("free!" In this case is hardly so, since if you're not the customer, you're the product).
The Sickle Cell Foundation offers free legitimate testing, too.
posted by spitbull at 2:25 AM on November 26, 2013 [7 favorites]
Their latest television ad campaign doesn't sell it as a fun and informative way to discover your ancestry. In fact, ancestry isn't mentioned at all. The new ads are all about the "hundreds of things" you can learn about your health and how you can then make lifestyle changes.
Sure there are disclaimers like "On average, users receive at least one or two results that may help with proactively managing health." and "For informational purposes only. Not a substitute for professional medical advice." but ... they are very much selling themselves as a company that tests for medical issues these days and not so much discovering hidden ancestry and family tree building.
posted by Orb at 7:28 AM on November 26, 2013 [1 favorite]
Sure there are disclaimers like "On average, users receive at least one or two results that may help with proactively managing health." and "For informational purposes only. Not a substitute for professional medical advice." but ... they are very much selling themselves as a company that tests for medical issues these days and not so much discovering hidden ancestry and family tree building.
posted by Orb at 7:28 AM on November 26, 2013 [1 favorite]
Does this mean that we won't see any promotions for the company on MetaTalk anymore?
posted by euphorb at 7:54 AM on November 26, 2013
posted by euphorb at 7:54 AM on November 26, 2013
Individual users can use MetaTalk for whatever they want, within reason. There was never any official promotion of 23andMe. So no, it doesn't necessarily mean that.
posted by jessamyn at 8:05 AM on November 26, 2013
posted by jessamyn at 8:05 AM on November 26, 2013
They are doing 990 000 SNP markers, which is state-of-the-art worldwide, short of sequencing a person's entire genome. The question is what one does with the risk factors they report. They give confidence levels, and also plenty of warnings before showing the results. An MD would use the same test, and then charge you a big sum for reading the results and the online caveats to you. The FDAs hangup concerns what constitutes "practicing medicine" and what is a "medical device." I think the issue should be about a person's inherent right to information about themselves.
Also: Mendelspod
posted by semmi at 8:41 AM on November 26, 2013 [2 favorites]
Also: Mendelspod
posted by semmi at 8:41 AM on November 26, 2013 [2 favorites]
The question is what one does with the risk factors they report. They give confidence levels, and also plenty of warnings before showing the results. An MD would use the same test, and then charge you a big sum for reading the results and the online caveats to you. The FDAs hangup concerns what constitutes "practicing medicine" and what is a "medical device."
The FDA is correct that this is a medical device, just like glucose meters that are marketed directly to consumers and allow consumers to collect and interpret information about themselves. Their concern appears to be that 23andMe is offering risk factors and confidence levels that they can't substantiate scientifically. The right to genetic data doesn't have anything to do with this, and I think it clouds the issue.
posted by muddgirl at 8:52 AM on November 26, 2013 [6 favorites]
The FDA is correct that this is a medical device, just like glucose meters that are marketed directly to consumers and allow consumers to collect and interpret information about themselves. Their concern appears to be that 23andMe is offering risk factors and confidence levels that they can't substantiate scientifically. The right to genetic data doesn't have anything to do with this, and I think it clouds the issue.
posted by muddgirl at 8:52 AM on November 26, 2013 [6 favorites]
As a happy 23&me customer, I will say this...
For $99, you get a lot of good general information about your ancestry and your genetic health markers.
It's not a magical diagnostic machine, if it was, they could charge whatever they wanted.
What it does do better than your doctor, is look at a big picture based on your DNA.
better than a fortune cookie, not quite a Star Trek tricorder.
In addition, there is access to communities of people who are also interested in exploring and hacking their health.
Sharing the information with my doctor didn't rip any doors wide open and solve any mysteries, but it did spark a series of conversations that were helpful and encouraging.
posted by bobdow at 10:03 AM on November 26, 2013 [2 favorites]
For $99, you get a lot of good general information about your ancestry and your genetic health markers.
It's not a magical diagnostic machine, if it was, they could charge whatever they wanted.
What it does do better than your doctor, is look at a big picture based on your DNA.
better than a fortune cookie, not quite a Star Trek tricorder.
In addition, there is access to communities of people who are also interested in exploring and hacking their health.
Sharing the information with my doctor didn't rip any doors wide open and solve any mysteries, but it did spark a series of conversations that were helpful and encouraging.
posted by bobdow at 10:03 AM on November 26, 2013 [2 favorites]
But isn't the whole point of FDA regulation to make sure your interests are being considered here? They hardly have a perfect track record of preventing abuse by the medical/Pharma industries, but they're all we've got.
My interests are not being considered if a product I want to buy, that has no comparable price point at other companies, is being taken off the market.
posted by corb at 10:48 AM on November 26, 2013
My interests are not being considered if a product I want to buy, that has no comparable price point at other companies, is being taken off the market.
posted by corb at 10:48 AM on November 26, 2013
This seems like a pointless venue for a debate as to whether or not the government has a duty to protect citizens from sham products.
posted by muddgirl at 11:00 AM on November 26, 2013 [1 favorite]
posted by muddgirl at 11:00 AM on November 26, 2013 [1 favorite]
My interests are not being considered if a product I want to buy, that has no comparable price point at other companies, is being taken off the market.
Please do not start this derail here. You can start your own thread about arguments for/against government regulation generally but it's approaching self-parody for you to start that derail here.
posted by jessamyn at 11:27 AM on November 26, 2013 [11 favorites]
Please do not start this derail here. You can start your own thread about arguments for/against government regulation generally but it's approaching self-parody for you to start that derail here.
posted by jessamyn at 11:27 AM on November 26, 2013 [11 favorites]
I'm sorry if I was unclear: my point was that this doesn't protect anyone from anything, there are no documented cases of abuse or harm, and the only effect that the FDA seems to have had in "Regulating" genetic testing companies is to price them out of most people's reach.
I think even if you do believe in government regulation, this could and should well be considered a not-optimal state of affairs. Not everyone has a few grand to throw down at a not-23andme-test.
posted by corb at 11:40 AM on November 26, 2013
I think even if you do believe in government regulation, this could and should well be considered a not-optimal state of affairs. Not everyone has a few grand to throw down at a not-23andme-test.
posted by corb at 11:40 AM on November 26, 2013
Of your three points, none hold any water. Analytical validations protect people from bad laboratory practices, e.g. somebody dropping a sample on the floor and then not knowing whether or not it can be properly analyzed afterwards. Or even knowing what the error rate is at all! There's a documented case of harm above. It's ridiculous to claim that following FDA regulations would price this out of anybody's reach. At 23andMe's size and volume, I wouldn't be surprised at all if they kept to the same price, and even with inconceivably poor implementation on 23andMe's side I couldn't see it adding more than $50 per sample, with $1 to $10 being a much more realistic range.
The non-optimal state of affairs is that 23andMe isn't even trying to follow the law, and giving personal genomics a black eye for it. I wish there was a more reputable company, with better practices, that had as much mindshare.
It's far more optimal that the FDA provide clear and concise ways to prove that a device does what it says, rather than relying on lawsuits and juries to decide if a device has been fraudulently advertised. I'd much rather deal with the FDA than the courts.
posted by Llama-Lime at 1:01 PM on November 26, 2013 [5 favorites]
The non-optimal state of affairs is that 23andMe isn't even trying to follow the law, and giving personal genomics a black eye for it. I wish there was a more reputable company, with better practices, that had as much mindshare.
It's far more optimal that the FDA provide clear and concise ways to prove that a device does what it says, rather than relying on lawsuits and juries to decide if a device has been fraudulently advertised. I'd much rather deal with the FDA than the courts.
posted by Llama-Lime at 1:01 PM on November 26, 2013 [5 favorites]
I have been hearing a LOT of what sounds to me like poor reporting on this in the media today, mostly reports misunderstanding the basis of the FDA warning letter (focusing on failure to provide documentation of accuracy as equivalent to not scientifially accurate, an understandable error) and missing that it's largely the company's failure to effectively work with the FDA that has created the situation.
I'm also seeing stern suggestions to download your data ASAP, which, well, that can't do any harm, but I did not see anything in the letter suggesting that the FDA was requiring them to pull the data.
Finally, and here's something I can't determine from the language of the letter, the FDA instructs 23andme to immediately cease "marketing." Does that mean no more advertising but membership sales may continue? Or does it mean no more sales of membership? Or does it mean no sales and no advertising? If it does mean no advertising, that is, no presentation of marketing information, one would assume that would have implications for the non-member UI of the website.
So far I haven't seen any long-look business analysis of how insurers like United Healthcare may be pressuring the company or the FDA, just that passing mention in the Bloomberg story. I assume that reflects the apparent and obvious catalyst for the letter, 23andMe's poor management of their regulatory communications.
Anyway, I don't have answers to these questins and concerns. It is disappointing to see the poor reporting.
posted by mwhybark at 1:35 PM on November 26, 2013 [1 favorite]
I'm also seeing stern suggestions to download your data ASAP, which, well, that can't do any harm, but I did not see anything in the letter suggesting that the FDA was requiring them to pull the data.
Finally, and here's something I can't determine from the language of the letter, the FDA instructs 23andme to immediately cease "marketing." Does that mean no more advertising but membership sales may continue? Or does it mean no more sales of membership? Or does it mean no sales and no advertising? If it does mean no advertising, that is, no presentation of marketing information, one would assume that would have implications for the non-member UI of the website.
So far I haven't seen any long-look business analysis of how insurers like United Healthcare may be pressuring the company or the FDA, just that passing mention in the Bloomberg story. I assume that reflects the apparent and obvious catalyst for the letter, 23andMe's poor management of their regulatory communications.
Anyway, I don't have answers to these questins and concerns. It is disappointing to see the poor reporting.
posted by mwhybark at 1:35 PM on November 26, 2013 [1 favorite]
Legal analysis: http://www.legalgenealogist.com/blog/2013/11/26/fooling-with-fda/
posted by mwhybark at 1:43 PM on November 26, 2013 [1 favorite]
posted by mwhybark at 1:43 PM on November 26, 2013 [1 favorite]
Coincidentally, I fired my falconer shortly after getting my 23andMe results, which confirmed some long-held suspicions as to my genetic heritage.
Clearly you're just a viral campaign for a Ladyhawke reboot.
posted by MiltonRandKalman at 3:25 PM on November 26, 2013 [1 favorite]
Clearly you're just a viral campaign for a Ladyhawke reboot.
posted by MiltonRandKalman at 3:25 PM on November 26, 2013 [1 favorite]
the only effect that the FDA seems to have had in "Regulating" genetic testing companies is to price them out of most people's reach
I work in a genomics lab and the expense of sequencing has nothing to do with the FDA. You very simply do not know what you are talking about, at all.
posted by Blazecock Pileon at 6:14 PM on November 26, 2013 [2 favorites]
I work in a genomics lab and the expense of sequencing has nothing to do with the FDA. You very simply do not know what you are talking about, at all.
posted by Blazecock Pileon at 6:14 PM on November 26, 2013 [2 favorites]
From 23andme: An Update Regarding The FDA’s Letter to 23andMe
posted by Blazecock Pileon at 6:35 PM on November 26, 2013 [1 favorite]
posted by Blazecock Pileon at 6:35 PM on November 26, 2013 [1 favorite]
Well, at least they're going to pretend to move towards compliance. But I doubt that it will be straightforward. Also I did not know that they had a "lab partner" that was actually running the stuff, they don't even do that in house?! Perhaps they haven't complied because they have no clue what's actually going on in the lab?
Somewhat astounding that they've raised that much capital for only doing the software.
posted by Llama-Lime at 6:52 PM on November 26, 2013 [1 favorite]
Somewhat astounding that they've raised that much capital for only doing the software.
posted by Llama-Lime at 6:52 PM on November 26, 2013 [1 favorite]
At least I can be sure that I'm 100% me. Always had that worry in the back of my mind...
posted by mrbill at 10:36 AM on November 27, 2013
posted by mrbill at 10:36 AM on November 27, 2013
Somewhat astounding that they've raised that much capital for only doing the software.
Doing the marketing, surely.
posted by jaduncan at 11:56 AM on November 27, 2013
Doing the marketing, surely.
posted by jaduncan at 11:56 AM on November 27, 2013
I sure envy those lucky Somalis, no government regulations to make their lives difficult! Maybe 23andme should move to Mogadishu, I hear the market for medical services there is very hungry for innovation!
posted by spitbull at 6:08 AM on November 28, 2013 [1 favorite]
posted by spitbull at 6:08 AM on November 28, 2013 [1 favorite]
Regarding raising the capital for this firm, I believe that it was founded by Sergey Brin's now ex-wife. The firm was founded in a good location where the confluence of talent, funds and connections for a firm like this would be optimal; even if they do not do things in-house.
posted by jadepearl at 10:03 AM on November 29, 2013
posted by jadepearl at 10:03 AM on November 29, 2013
Scientific American checks in.
23andMe Is Terrifying, But Not for the Reasons the FDA Thinks
posted by Obscure Reference at 1:51 PM on December 1, 2013 [2 favorites]
23andMe Is Terrifying, But Not for the Reasons the FDA Thinks
posted by Obscure Reference at 1:51 PM on December 1, 2013 [2 favorites]
Gah, that's a terrible piece. Aside from anything else, this section makes me roll my eyes:
posted by jaduncan at 2:16 AM on December 2, 2013
Back when Google was first launched, the founders insisted that the company would never sell you out to advertisers. The company admitted that it would share aggregate information about users’ behavior with anyone who ponied up enough money, but the company’s privacy policy promised that “[i]ndividually identifiable information about you is not willfully disclosed to any third party without first receiving your permission.” A decade and a half later, after countless minuscule frog-in-boiling-water changes, Google’s privacy policy is craftily worded, diluting the word “consent” so that it’s implicit in most cases.Google never shares the non-aggregate data with advertisers, purely because it's their secret sauce. 23andMe, no matter if they are doing data mining or not, will not share non-aggregate data for the same reason.
posted by jaduncan at 2:16 AM on December 2, 2013
23andMe hit with class action over “misleading” genetic ads
In a complaint filed last week, San Diego resident Lisa Casey claims she paid for 23andMe’s services, and received results by email, after learning of the home-based saliva test via TV and online ads. Now, days after the FDA issued a stinging letter that ordered the company to stop its marketing, Casey seeks to represent the “ten or hundreds of thousands” of others across America who paid for the product.
According to the complaint, which seeks at least $5 million under various California state laws, 23andMe makes false and misleading claims about the tests’ ability to provide relevant genetic information about breast cancer, diabetes, lactose intolerance and various other conditions.
posted by Blazecock Pileon at 5:32 PM on December 2, 2013
In a complaint filed last week, San Diego resident Lisa Casey claims she paid for 23andMe’s services, and received results by email, after learning of the home-based saliva test via TV and online ads. Now, days after the FDA issued a stinging letter that ordered the company to stop its marketing, Casey seeks to represent the “ten or hundreds of thousands” of others across America who paid for the product.
According to the complaint, which seeks at least $5 million under various California state laws, 23andMe makes false and misleading claims about the tests’ ability to provide relevant genetic information about breast cancer, diabetes, lactose intolerance and various other conditions.
posted by Blazecock Pileon at 5:32 PM on December 2, 2013
23andme genotypes are all wrong
A key question the FDA has asked, as it does for any diagnostic test, is whether the SNP calls are accurate. The answer is already out there. First, someone has performed a 23andme replicate experiment precisely to assess the error rate. In an experiment in 2010 with two replicates, 85 SNPs out of about 600,000 were different. Today, Illumina types around 1 million SNPs, so one would expect even more errors. Furthermore, a replicate analysis provides only a lower bound, since systematic errors will not be detected. Another way to examine the error rate is to look at genotypes of siblings. That was written about in this blog post which concluded there were 87 errors. 23andme currently uses the Illumina Omni Express for genotyping, and the Illumina spec sheet claims a similar error rate to those inferred in the blog posts mentioned above. The bottom line is that even though the error rate for any individual SNP call is very very low (<0.01% error), with a million SNPs being called there is (almost) certainly at least one error somewhere in the genotype. In fact, assuming a conservative error rate leading to an average of 100 errors per genotype, the probability that a 23andme genotype has no errors is less than 10^(-40).
The fact that 23andme genotypes are wrong (i.e. at least one error in some SNP) wouldn’t matter if one was only interested in a single SNP. With very high probability, it would be some other SNPs that are the wrong ones. But the way people use 23andme is not to look at a single SNP of interest, but rather to scan the results from all SNPs to find out whether there is some genetic variant with large (negative) effect. The good news is that there isn’t much information available for the majority of the 1 million SNPs being tested. But there are, nevertheless, lots of SNPs (thousands) to look at. Whereas a comprehensive exam at a doctor’s office might currently constitute a handful of tests– a dozen or a few dozen at most– a 23andme test assessing thousands of SNPs and hundreds of diseases/traits constitutes more diagnostic tests on an individual at one time than have previously been performed in a lifetime.
posted by Blazecock Pileon at 9:31 AM on December 5, 2013 [1 favorite]
A key question the FDA has asked, as it does for any diagnostic test, is whether the SNP calls are accurate. The answer is already out there. First, someone has performed a 23andme replicate experiment precisely to assess the error rate. In an experiment in 2010 with two replicates, 85 SNPs out of about 600,000 were different. Today, Illumina types around 1 million SNPs, so one would expect even more errors. Furthermore, a replicate analysis provides only a lower bound, since systematic errors will not be detected. Another way to examine the error rate is to look at genotypes of siblings. That was written about in this blog post which concluded there were 87 errors. 23andme currently uses the Illumina Omni Express for genotyping, and the Illumina spec sheet claims a similar error rate to those inferred in the blog posts mentioned above. The bottom line is that even though the error rate for any individual SNP call is very very low (<0.01% error), with a million SNPs being called there is (almost) certainly at least one error somewhere in the genotype. In fact, assuming a conservative error rate leading to an average of 100 errors per genotype, the probability that a 23andme genotype has no errors is less than 10^(-40).
The fact that 23andme genotypes are wrong (i.e. at least one error in some SNP) wouldn’t matter if one was only interested in a single SNP. With very high probability, it would be some other SNPs that are the wrong ones. But the way people use 23andme is not to look at a single SNP of interest, but rather to scan the results from all SNPs to find out whether there is some genetic variant with large (negative) effect. The good news is that there isn’t much information available for the majority of the 1 million SNPs being tested. But there are, nevertheless, lots of SNPs (thousands) to look at. Whereas a comprehensive exam at a doctor’s office might currently constitute a handful of tests– a dozen or a few dozen at most– a 23andme test assessing thousands of SNPs and hundreds of diseases/traits constitutes more diagnostic tests on an individual at one time than have previously been performed in a lifetime.
posted by Blazecock Pileon at 9:31 AM on December 5, 2013 [1 favorite]
To get an idea of why Blazecock's links are so important, even very small error rate in screening programmes can be dramatically harmful in ways that are not really intuitive without a really solid grasp of statistics. For example, say a new medical test has been shown to be effective in the early detection of an illness, and as an otherwise healthy patient you take it. The chance that the test correctly identifies the illness as positive is 99%, and the chance that the test correctly identifies someone without the illness as negative is 95%. The incidence of the illness in the general population is 0.0001. You test positive, what is the probability that you actually have the illness?
You would naturally think you're pretty boned but your real chances are actually about 1 in 500, because statistics. This is a really good guide for making sense of medical testing,
You would naturally think you're pretty boned but your real chances are actually about 1 in 500, because statistics. This is a really good guide for making sense of medical testing,
Making Sense of Testing [PDF]But really what we're talking about is a lot more like medical screening, which uses a copletely different kind of math that is a lot less favorable to getting useful answers,
Why scans and health tests for well people aren't always a good idea. Adverts and media reports say that people with no symptoms, nor reason to suspect they have a disease can find out what they will get in the future, “reverse the disease processes before symptoms appear”, or even discover how they will die. People are promised instant results, valuable insights and ‘peace of mind’. What many people are getting is a lot of confusion and anxiety, ongoing trips to the doctor and, sometimes, unnecessary medical procedures. The guide presents a few insights and highlights common misconceptions about having health tests and scans.
Making Sense of Screening [PDF]posted by Blasdelb at 11:02 AM on December 5, 2013
A guide to weighing up the benefits and harms of health screening programmes. Public expectations about screening don't match what screening programmes can deliver. By addressing misconceptions about how screening works, its limitations and the calculation of benefits and harms, scientists and clinicians hope to bridge the gap between the active debates of the scientific community and the concerns raised by the public.
I wonder if the interactions of these issues (false positives and multi-nomial effects thereof) contributed to 23andme stiffarming the FDA. I mean, I percieve the utility case for the 23andme results to be, get result, go "huh," show results to doc, get valid medical tests, create medical plan. But if the interactions of the higher error rate in the 23andme kit and the base error rate of more rigorous testing will predictably increase testing demand where the only testing flag is the 23andme test, isn't that a textbook case of "unneeded tests," a (possibly stupid) bugaboo of healthcare cost elevators? Hm.
posted by mwhybark at 1:45 PM on December 5, 2013
posted by mwhybark at 1:45 PM on December 5, 2013
As to what precisely ceasing "marketing" means, they had my sample in hand but I had not received results yet when this news broke. I received my results starting a couple days ago, so they're at least still doing what was already in process.
posted by jocelmeow at 6:27 PM on December 5, 2013
posted by jocelmeow at 6:27 PM on December 5, 2013
« Older Dogfood | For a Follow Up, a Pirate Plays Daft Punk on 8... Newer »
This thread has been archived and is closed to new comments
23andme made a very serious misdiagnosis, and the author played "amateur bioinformaticist" and figured out it was a bug
posted by neil pierce at 2:09 PM on November 25, 2013 [19 favorites]