Hacker News Reader

Comment by guylhem | Hacker News Reader

As usual, the government is trying to meddle with companies. Read the letter, but make no mistake - the "kind" tone, especially when reminding how they did they best to get in touch, have meetings, help insuring compliance etc. is just a decoy. The truth is with "must immediately discontinue marketing the PGS".

The gov wants to decide what's best for the people. Should the people decide differently, using their wallets for example, this anomaly will be quashed.

This trend is especially strong in the medical field - gov approval required everywhere, then people wonder why medical things are so damn expansive.

I use 23andme and I'm happy with the information provided. I know it's not reliable - it's not a lab test anyone will use to base important decisions on, since it is not a full sequence of genes.

Yet, by being commercially available and easy to use, it is paving the way for commercial offers of full genome sequencing, which I damn well intend to use when they reach the $500 threshold.

However, gov actions like this one may very well make that impossible, making sure the only full genome sequencing offers there will be will be "FDA cleared" at a huge markup.

Suggestions to "medical" like companies - get out the gov eye. Move your business to Asia, the caribbean or wherever the gov will not get in your way like this. I want to keep using (and recommending) your products!

View on HN

> The gov wants to decide what's best for the people. Should the people decide differently, using their wallets for example, this anomaly will be quashed.

Wallets are incredibly poor judges of medical efficacy. The FDA rules are in place for very good reasons, namely to prevent snake-oil salesmen and other unscrupulous merchandisers from foisting their -- potentially harmful -- products upon an unsuspecting public.

I would much rather have FDA clearance for medical products and devices than the alternative that you seem to be proposing.

I don't think anyone is going to argue about the FDA's role in ensuring drugs are safe, but most countries don't require positive proof of effectiveness and, frankly, have better health outcomes than the US. Doctors are quite capable of reviewing the evidence to see if drugs are effective, and the FDA doesn't in practice do anything to prevent the sail of non-poisonous snake oil.

A number of western countries take a strategy that tries to make questionable-but-safe drugs legal-but-rarely-used. For one, they ban direct-to-consumer advertising; the USA and New Zealand are the only two developed countries where it's legal to advertise to consumers. And for another, state health-care systems will typically not pay for drugs that lack evidence of efficacy, and most people are used to getting their healthcare through the state health system. That leaves it technically legal for consumers to buy, out-of-pocket, a drug that can't be advertised to them. In practice a handful of people do so, but not too many.

It looks like they do sell this in Europe. Although I'm not sure if that's because it's approved at all, or just because Europe considers it some kind of novelty product rather than a medical device; the FDA and EU categorization of drugs/devices/etc. is a bit different.

Doctors encompass the entire range of people (from morons to geniuses). As a group, they aren't some magnificent godlike all knowing figure.

"Programmers are quite capable of reviewing the evidence to see if code is effective" — yet, we end up with thousands of people using MongoDB based off fad and hype and losing data by the day.

(Wait, did I just advocate for a government-level code approval board? In the right hands it would probably help if it prevented more Mongos and Hadoops from entering the world.)

Losing data when you're a professional who should have known what you're doing is one thing.

Losing your life when you're a layman in the medical field is another.

Doctors make mistakes, but there isn't anything preventing that in any event. In theory a doctor can prescribe any drug for any ailment (though there are lots of checks and consequences), and prescribing drugs for things other than their FDA approved use is very common[1]. You have to rely on doctor judgment at some point and efficacy trials don't substitute for competent doctors. And sometimes the whole process goes awry because a drug is approved for one problem, later becomes the only treatment for some people with a second problem, then the FDA withdraws approval because there are now better treatments for the first problem, leaving those with the second problem in the lurch.[2]

[1]http://en.wikipedia.org/wiki/Off-label_use [2]http://nielsenhayden.com/makinglight/archives/007140.html There's some important context in the comments.

I think the US's for profit healthcare system makes the FDA very important. Here in the US, we have pharmaceutical salesman who come in and schmooze the staff to then get to the doctors. Pharmaceutical rep is a job known to be very amenable to attractive women with little skills outside of flirting.

Your definition of "for profit" is somewhat misleading, as many Canadian doctors, nurses, pharmacists, etc. are the residual claimants ('profiteers') of almost any course of testing or treatment; and I suspect that this is true of many countries with less strict drug regulations.

Can you please give example of such a country? I can't think of any country with working healthcare where you are allowed to claim anything medical (or heaven forbid perform) before clearing it with the authorities.

No, every country I can think of requires new medicines to be cleared with the authorities. The difference is what that clearance entails.

> Wallets are incredibly poor judges of medical efficacy.

So are bureaucrats who are subject to the influence of lobbyists. Snake-oil salesmen and peddlers of alternative medicine do sell their stuff under the current system, while true innovation is crushed, as is evident in this case.

23&me isn't particularly innovative...They just took existing genotyping and made it available to everyone in pretty graphs

Care to share a link to a competitor?

Any hospital offering primary care has a lab at their disposal that can do this. Genotyping is nothing new...but as for competitors in 23&me's area (aka public access genotyping for cheap an that I can remember off the top of my head): deCODEme, lumigenix, Dna Genotek and MapMyGenome

Navigenics, with Navigenics Health Compass, was a competitor to 23andMe with their PGS offering, but Navigenics appears to have discontinued the competing offering (Navigenics was received initial contact from the FDA at about their product at the same time 23andMe did, they also went through similar rounds with the State of California for licensing to operate their before that.)

There's a number of other offerings that have been made in the same general space, as well.

>Wallets are incredibly poor judges of medical efficacy. The FDA rules are in place for very good reasons, namely to prevent snake-oil salesmen and other unscrupulous merchandisers from foisting their -- potentially harmful -- products upon an unsuspecting public.

There is snake oil for sale in every store, and no one has ever been hurt from their test.

> The gov wants to decide what's best for the people. Should the people decide differently, using their wallets for example, this anomaly will be quashed.

There are huge information asymmetries involved when it comes to medical drugs, and also huge barriers to experimenting with different products. "Voting with your wallet" is appropriate for determining which cell phone should be popular, not which medical drugs or medical tests.

We as a society have determined that the appropriate risk versus innovation trade-off when it comes to medicines and drugs is less risk at the potential cost of less innovation. The FDA exists to safeguard that policy.

It's funny that you mention Asia, because the trend in Asian countries is to follow the American/European model. Asians have had enough of thinking "is this rat meat?" every time they bite into a lamb kebob, and are regulating their food and drug industries. And by "asians" I mean the Chinese, because the Japanese and Koreans have adopted the Western regulatory model long ago.

The Chinese have the death penalty for food adulteration with toxic or non-food products[1]. It's funny seeing people complain about heavy handed regulation in comparison.

[1] https://en.wikipedia.org/wiki/Capital_offences_in_the_People...

> There are huge information asymmetries involved when it comes to medical drugs...

This is what the FDA should be doing. Correcting information asymmetries. With labeling requirements, unbiased PSAs, drug pamphlets and literature. Not telling people what they can and can't put in their bodies, or what businesses can and can't sell on the open market.

In fact, I'd say most government regulatory functions should be set up this way. Clearly inform and educate your citizens, don't opaquely dictate the behavior of your subjects.

Well, but this is exactly what they are doing here - the letter claims that they required corrections in the marketing claims (which would be correcting information asymmetries / labeling requirements), this wasn't done properly for a long time, so they are now enforcing it.

For any reasonable requirement, if people/companies don't comply then it has to escalate to [threats of] harsher actions in order to force the reasonable thing to actually happen.

This isn't a drug.

I know you're responding to a more general point, but I don't think any of your post applies to this situation. This is about marketing materials, and more importantly the factual information therein.

> I know it's not reliable - it's not a lab test anyone will use to base important decisions on, since it is not a full sequence of genes.

If they made that more clear, then they wouldn't have had to send the letter. It's why you see "This is not intended to diagnose, treat, or cure any disease" on all the non-FDA-approved supplements/energy drinks. My money is on them paying the fine and scaling back the marketing.

The purpose behind these FDA rules are to make sure things aren't marketing as "cures" when they aren't. People's health is serious business; there's no room for failure. If 1.0 doesn't work (or makes things worse), then the people that need it might not live to see 1.1.

Deleted comment

That's true, but it's not a slam-dunk case for rushing ahead without sufficient testing. There's a reason that "first, do no harm"[1] is considered one of the fundamental principles of medical ethics. Medical interventions can have disastrous side effects.[2]

[1] https://en.wikipedia.org/wiki/Primum_non_nocere [2] https://en.wikipedia.org/wiki/Thalidomide#Birth_defects_cris...

It's not a slam-dunk case, just keep in mind it's very easy to measure the harm done by things like Thalidomide, but not so easy to measure the amount of people dead or maimed by not getting treatment due to the glacial pace of regulatory approval. "First, do no harm" could just as easily apply to not dragging your feet in giving someone potentially life-saving treatment.

I'm curious, are there any other examples than Thalidomide? It's a true horror story, but there should be a limit to how much precaution one disaster justifies in the future. Many other drugs with ultimately fatal but initially undiscovered side effects have made it through FDA approval, too.

http://en.wikipedia.org/wiki/Diethylstilbestrol causes some drastic problems for the daughters (and arguably also the sons) of women who took it while pregnant.

This example is rather famous in the legal world because a) the drug was off-patent for many years before the problems were identified, and so there were many different manufacturers, and b) the problems were often not apparent until decades after exposure had occurred - two factors which make it very difficult to fairly ascribe liability to those who suffered avoidable injuries. This led to some innovative thing in damages/liability allocation which is now an essential part of understanding tort law.

I mention this because I've had some conversations with libertarian legal thinkers where they bring up this same argument of Thalidomide vs. opportunity cost - and its true that the barriers to approval were so low back in the late 50s that it's very unlikely such a drug could ever get on the market these days. Also, the effects were so drastic that that they were quickly noticed, and the number of birth defects ascribable to Thalidomide are in the low 10,000s. By contrast, millions of people are living with the impact of DES, and the magnitude of the affected class is also a factor in the legal significance of the damages formula. So any lawyer who's talking about balancing innovation and regulation in the drug market without addressing DES is being a bit disingenuous, and you should weight your judgment accordingly.

That's fascinating. The Wikipedia links cover the landmark nature of how liability was distributed, but I'm as curious about how liability was established. I don't know much about tort law, but is it essentially automatic that a manufacturer would be liable for a product that caused such damage, no matter how difficult it would have been for them to have been aware of that potential?

FYI I like tort law but am not a lawyer, just a law school dropout. My view is that liability depends on the degree to which the manufacturer was in a position to be aware of the risks, and also on the relative ability of the consumer to discern such risks.

To take the latter point first, drug consumers are generally in a poor position to evaluate the risks of a given medication. If I set up a wheel manufacturing business, and sell square wheels, you are in a poor position to complain about the resulting bumpy ride [1] because the superiority of round wheels is so glaringly obvious. Unless you hold a medical or pharmaceutical degree, though, your ability to evaluate a drug's safety and efficacy is probably very limited. I'm smart enough to read and get the gist of a biochemistry paper to have an idea of what a drug does, but nowhere near knowledgeable enough to review or spot all but the most egregious errors. So while I am the sort of person who reads all the fine print on drug labels and sometimes digs into the academic research, in the end I'm taking a great deal on faith whenever I take medicine. by contrast, the drug manufacturer has not just one biochemical expert, but whole teams of them. This certainly puts them in the best position to foresee and even look for potential problems.

Now back to the first point: drug discovery is not the exact science we'd like it to be due to the gaps in our knowledge. Accordingly, drug companies must be given some leeway, compared to other manufacturers who enjoy a greater degree of foreseeability. suppose I set up a ladder company, but the ladders I manufacture employ such thin metal rungs that they cannot bear any weight above 100 pounds. They look good enough, but soon consumers begin suffering accidents when using them. I could not claim the accidents were unforeseeable since the tensile properties of metals are so well known and the tests are so easy to perform, that I should have realized the risk when designing my product, never mind before shipping it! It would be unfair to hold drug manufacturers to that degree of liability, but on the other hand a) experimental protocols exist precisely to address this problem and are widely understood within the drug industry, and b) since DES was on sale from 1941 onwards and class action lawsuits were brought against manufacturers in the late 1960s, more than enough time had elapsed for industry participants to have apprised themselves of the risk and limited the risk to consumers to a greater extent than they did.

If you read Judge Mosk's opinion, in footnote 24 [2] he summarizes a 7-point test for establishing (rather than allocating) liability. The answer you seek is there:

  The suggested requirements are as follows:
  1. There existed an insufficient, industry-wide standard of safety as to the manufacture of the product.
  2. Plaintiff is not at fault for the absence of evidence identifying the causative agent but, rather, this absence of proof is due to defendant's conduct.
  3. A generically similar defective product was manufactured by all the defendants.
  4. Plaintiff's injury was caused by this defect.
  5. Defendants owed a duty to the class of which plaintiff was a member.
  6. There is clear and convincing evidence that plaintiff's injury was caused by a product made by one of the defendants. For example, the joined defendants accounted for a high percentage of such defective products on the market at the time of plaintiff's injury.
  7. All defendants were tortfeasors.

This is a very brief abstraction of a key theory in a law review article written by a student at Fordham University, referred to in the opinion as 'the Fordham comment.' This unusually influential article [3] basically sets out a whole theory of 'enterprise liability' and damages allocation addressing why the manufacturers should share liability in the context of the DES in much greater detail, although it's rather painful to read for people who are not legal nerds as it approaches some sort of footnote event horizon. Whether the Sindell court's findings allow the legal academy to be held liable for typographically-induced brain trauma remains unresolved.

1. http://dogknobit.com/wp-content/uploads/2013/04/square-wheel...

2. http://online.ceb.com/calcases/C3/26C3d588.htm#SA000885

3. http://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=231...

Doesn't exactly answer your question, but here's a list of formerly approved drugs that were withdrawn due to risks to the patient (22 in the last decade).

http://en.wikipedia.org/wiki/List_of_withdrawn_drugs

That's kind of my point. Does that mean we should ratchet up FDA's precautionary attitude dramatically further, to the point where that number is perhaps 2 instead of 22? I don't think so. There's a sensible tradeoff to be made, but tuning exactly where that tradeoff should be isn't the kind of thing you settle with dramatic examples of things gone wrong.

It's difficult to measure how many lives were saved or made substantially more livable because the FDA was not more cautious than it has been. Surely that number is not zero, and perhaps it's greater than the net harm done by drugs whose effects might have been caught by more prolonged testing. I don't think we know.

There's another aspect to tradeoffs where Lumiracoxib was introduced to compete with Celecoxib. Celecoxib has a known acknowledged issue where it increases the risk of heart attack although its relatively easy on the liver. Unfortunately Lumiracoxib occasionally destroys livers. So superficially you'd think having both on the market would be a great idea, because people with a personal or family history of liver disease would probably be better off with Celecoxib but people with a personal or family history of heart attack would be better off with Lumiracoxib. However, worldwide in general all the .gov disagree with me, and as the new guy on the block Lumiracoxib more or less got the boot. It never even made it to market in the USA as far as I know. Probably has something to do with relative incidence of fatality.

BTW these are not made up names although they look like it. Google them for a good time if you're bored. I'm more interested in the chemistry issues than the politics or biochemisty, so I'd be more comfortable (authoritative?) talking about the location of the methyl groups than liver toxicity, but my first paragraph has no known intentional errors (although it probably has some unintentional errors LOL).

I guess the point of this ramble is if a company exists to provide highly personalized customized services, then using a metric of "joe average" is probably not overly meaningful. The response of joe average to a disease percentage is maybe not relevant to the response of someone with a high percentage to the same result.

We have a pretty good medical system designed for joe average, and its intersection with radically personalized services is unsurprisingly going to be an epic disaster.

> are there any other examples than Thalidomide?

A whole lot of people have died from contaminated blood transfusions: http://articles.latimes.com/1994-08-31/news/mn-33183_1_aids-...

That's another case of a medical intervention ending up doing more harm than good.

"I'm curious, are there any other examples than Thalidomide?

I would guess you're asking to google for list of withdrawn drugs which lead me to

http://en.wikipedia.org/wiki/List_of_withdrawn_drugs

Another interesting side effect is most mass shooters were on anti-depressants, although that's probably not the kind of direct fatal side effect you're looking for. Its a challenging google game to track down the toxicology reports for each mass shooter weeks after the smoke clears and identify which anti-depressant they were on at the time. Sometimes its a different kind of med, sometimes they're not on any med at all, although thats kind of rare.

The cost to over-rapid approval is that unborn children are mutilated by medicine taken by the pregnant mother, or that very many people are harmed by side-effects from expensive meds that are no more efficient (sometimes less so) than existing meds.

>very many people are harmed by side-effects from expensive meds that are no more efficient (sometimes less so) than existing meds

That one is easy: don't fast track drugs that are only marginally better than their peers, if at all, because by definition they aren't as important.

I don't have as clear an answer for pregnancy, other than exempting pregnant users from the fast track approval, but I'm not sure how a slower process actually protects pregnancies in the first place.

Its more complex than you would think, this is a good tome on the subject (albeit a bit dry) http://www.amazon.com/Bad-Pharma-Companies-Mislead-Patients/...

A lot of the modern pharmaceutical marketing regulation, and the regulations that makes approving new drugs a slow process, came in the wake of Thalidomide, which I think still provides a fairly strong case for why we should thoroughly test new medicines, and shouldn't rush them to market.

People die from the opposite approach as well from drugs that shouldn't have been on the market and aren't safe or effective. People dying is not a reason to just allow anything. Regulation is good when it comes to forcing companies to prove a drugs effectiveness by actually doing the science.

That's a good point. I don't know as much as I should about the clinical trial process or the ability of terminally-ill patients with nothing to lose to seek experimental drugs. In a better world, it'd be as simple as patient and doctor both giving an informed and well-reasoned consent, knowing the risks involved but choosing to take that risk anyway.

This is more complex than the simplistic picture you paint.

I'm moderately on the side of government not interfering where possible. But medicine and medical devices is one area in which such interference is quite likely a positive influence, on the whole.

Fact is, we don't want a return to the era of patent medicines. For every user of 23andme who is well-informed enough to know that the information is not reliable, experience shows that there will be another who is not. And these people will base life- and health-affecting decisions on the results provided by unregulated products. Then maybe they'll get sick or die.

That's why there are regulatory standards for medical devices and services. These standards don't prevent the sale of 23andme's services, but require that they are either a) regulated and proven to work, or b) explicitly and clearly described as not being so.

Yes, there are rampant inefficiencies within this system. But I'd wager that it's probably better than the alternative.

For every user of 23andme who is well-informed enough to know that the information is not reliable, experience shows that there will be another who is not.

The funny thing here, is that barring lab errors, which I think are highly unlikely unless they are being incredibly shoddy, which would take effort, the information that 23andMe provides is very reliable. Now it is debatable as to whether the individual's interpretation is (for example in the Alcohol Dehydrogenase gene, many, many, many caucasian users don't bother to read that the test is only informative for asians, but it's definitely there, in bold letters, etc.)

Could you explain a scenario where basing life and health decisions on 23andMe's product could lead to sickness and death? Maybe it's "because I'm sophisticated", but this biochemist can't really see that happening.

The funny thing here, is that barring lab errors, which I think are highly unlikely unless they are being incredibly shoddy, which would take effort, the information that 23andMe provides is very reliable

That's the point of certification - is the process reliable? Are the warnings clear? I won't cast any aspersions on the product (because it's pretty awesome), but I do see these as risks and requirements.

Could you explain a scenario where basing life and health decisions on 23andMe's product could lead to sickness and death?

"I've got all these symptoms of diabetes. But 23andme said I had a low risk! It's probably nothing. Guess I'll eat another pie."

If 23andme has shoddy lab practices, or unclear explanation of the limitation of their testing, then that's a risk. The process of certification involves managing that risk.

>Could you explain a scenario where basing life and health decisions on 23andMe's product could lead to sickness and death?

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD001877....

(this study is about mammograms of healthy women, another example of a health screening tool)

We found that breast cancer mortality was an unreliable outcome that was biased in favour of screening, mainly because of differential misclassification of cause of death. The trials with adequate randomisation did not find an effect of screening on cancer mortality, including breast cancer, after 10 years (RR 1.02, 95% CI 0.95 to 1.10) or on all-cause mortality after 13 years (RR 0.99, 95% CI 0.95 to 1.03).

Numbers of lumpectomies and mastectomies were significantly larger in the screened groups (RR 1.31, 95% CI 1.22 to 1.42) for the two adequately randomised trials that measured this outcome; the use of radiotherapy was similarly increased.

So women who get more screening get more treatment but didn't live longer (unnecessary treatment)

Study continues:

This means that for every 2000 women invited for screening throughout 10 years, one will have her life prolonged and 10 healthy women, who would not have been diagnosed if there had not been screening, will be treated unnecessarily. Furthermore, more than 200 women will experience important psychological distress for many months because of false positive findings. It is thus not clear whether screening does more good than harm. To help ensure that the women are fully informed of both benefits and harms before they decide whether or not to attend screening, we have written an evidence-based leaflet for lay people that is available in several languages on www.cochrane.dk.

yes, but getting a mammogram has an additional layer of regulation in between the end-user of 23andMe. A licensed radiologist should look the person askew when they insist on getting extra mammograms because of 23andMe, and order a follow-up BRCA test before proceeding.

Wow, no. You aren't seeing it.

A mammogram is a screening test. 23andMe is another type of screening test. You aren't getting extra mammograms because of 23andMe, they are the same thing.

Mammogram can screen you for early detection of a serious disease. 23andMe can do the same (they advertise - this is what the FDA has a problem with).

More screening for all diseases doesn't mean better outcome, and can be harmful, depending on the test, the people screened, the disease, etc., etc. and can lead to more disease (needless treatment). Throwing more screening at the wall for every disease doesn't do more good all the time. Especially when we learn that some diseases (prostate cancer) cause no symptoms most of the time. Treating these asymptomatic prostate cancers is very harmful, because they are harmless unless they progress. The treatment is risky and has serious side effects (impotence!)

Every time we have a screening test, we have to look at the science. What are we screening for? What is the good? What is the bad? Who is helped? Who is harmed? Do the positives outweigh the negatives?

Another outdated screening tool is the breast self exam - it was found (in scientific studies) it didn't do any good and caused harm (http://en.wikipedia.org/wiki/Breast_self-examination) Lots of over treatment without a reduce in cancer rates. Also, just learning about it increases women's worry and anxiety about breast cancer. This is a harmful screening tool. This was studied and abandoned.

----

BRCA test has nothing to do with this example.

Mammograms are currently very common in women a certain age >40 or >50. Most women get them every few years as part of their health maintenance to check for the presence of cancer, tumors or lumps. Right now there is research coming out that is questioning that practice, but the practice hasn't caught up to the research and most doctors say "get your mammogram" without informing patients of the risks, they treat it like it is "all good." That was the point of this research to ask "are we doing the right thing here by screening all these healthy women with mammograms? Let's study that."

I will personally opt out of the mammogram hype when I reach the mammogram age. I'm also going to inform my mother of the current research on the matter, I know she's already had a few mammograms herself.

> The funny thing here, is that barring lab errors, which I think are highly unlikely unless they are being incredibly shoddy, which would take effort, the information that 23andMe provides is very reliable.

Certainly, there is science to support certain claims of the class made in the reports 23andMe provides being made with some degree of reliability based on the kind of inputs 23andMe takes in. Whether what 23andMe actually does from end-to-end makes the actual reports they provide reliable in real world condition is...well, exactly the question they have seem to have decided not to expend the effort to answer even after initiating the process of getting clearance from the FDA to market their diagnostic product.

Ah, the "people will vote with their wallets" argument against medical regulation.

I'm with you. I'd much prefer a bunch of people die first so the rest of us can "vote with our wallets" instead of having a government agency step in and try to fix things before anyone gets hurt. I mean, my start-up is going to sell baby food laced with arsenic because I once heard that arsenic makes babies smarter and I think I once ingested some accidentally and I didn't die. If it really does end up killing babies, well, the market will eventually correct itself. And no harm, no foul, man. Except the dead babies. But at least I wouldn't have some nebbish bureaucrat knocking at my door trying to tell this American how to live his life.

What an absurd argument. If you sell baby food laced with arsenic, and babies die, you've committed murder. In a true free market, you wouldn't have a liability shield and would face severe consequences for putting out dangerous products. Just because it's a free market doesn't mean you get to do whatever you want with impunity. And also, the FDA is not some magic fairy dust that will shield you from bad products. Mountain Dew still has BVO in it, for example.

> In a true free market, you wouldn't have a liability shield and would face severe consequences for putting out dangerous products.

Liability for defective (whether through malice, negligence, or in some cases on a strict liability basis) products is government intervention in the marketplace, and is specifically designed to address one of the ways that reality does not align with the presuppositions under which an unregulated "free market" is ideal, to wit, that people generally do not, in fact, act with perfect knowledge of the outcomes of economic decisions when they engage in them.

It is no different than many other government interventions in the marketplace in its basis.

> And also, the FDA is not some magic fairy dust that will shield you from bad products. Mountain Dew still has BVO in it, for example.

That does not mean that its approval procedures (and procedures for reviewing products already on the market) do not have utility in providing clearer information on products within its regulatory domain and helping consumers avoid bad products.

You understand that the goal is to prevent the horrible thing from happening, right? Not to let the horrible thing happen and then try to fix something that's unfixable?

Just to pare back the hyperbole -- there's really no practical way for someone to die from a 23andMe kit. We should be able to agree on that.

There have been a bunch of concerns in this thread that are valid -- e.g., people might have to deal with bad news, or may see false positives, etc., that could cause them some inconvenience (though I would argue that 23andMe doesn't tell you that you have cancer, just maybe whether or not you're likely to).

Regardless, the idea that I could die from a 23andMe test is about as far-fetched as one can get.

> I use 23andme and I'm happy with the information provided. I know it's not reliable - it's not a lab test anyone will use to base important decisions on, since it is not a full sequence of genes.

You have unwittingly affirmed the very argument you are attempting to refute. Their inability to make accurate functional predictions from genomic data has much less to do with the technology (sequence vs. array) than it does with the fact that, isolated instances aside, nobody knows how to do this well. I am a 23andMe user myself and, knowing a bit about statistical genomics, I was surprised how credulously the results of one or two GWAS papers are presented as an almost-diagnosis on their website. GWAS are plagued by all sorts of confounding issues and the statistical evidence in favor of their conclusions is often extremely weak. This is not something that the general public appreciates, so I can see why FDA might get involved.

Hey! You should check out Elixir Sulfanimide! It prevents infections, but it's available over the counter. So easy! And it's from a Massengill, an established pharma company, so you can trust it!

So actually you can't buy elixir sulfanimide anymore. Because it was poisonous. It resulted in more than 100 deaths in 1937. This was before the FD&C Act, requiring safety tests. The founder of Massengill refused to take responsibility for the results: "We have been supplying a legitimate professional demand and not once could have foreseen the unlooked-for results. I do not feel that there was any responsibility on our part." It should be noted that there was a sulfanimide craze on at the time, as the first commercial antibiotic, so there was much money to be made.

One woman wrote to President Roosevelt, about her dead daughter: "The first time I ever had occasion to call in a doctor for [Joan] and she was given Elixir of Sulfanilamide. All that is left to us is the caring for her little grave. Even the memory of her is mixed with sorrow for we can see her little body tossing to and fro and hear that little voice screaming with pain and it seems as though it would drive me insane. ... It is my plea that you will take steps to prevent such sales of drugs that will take little lives and leave such suffering behind and such a bleak outlook on the future as I have tonight."

Source: http://en.wikipedia.org/wiki/Elixir_sulfanilamide

Massengill paid a minimum fine and continued on until it was acquired in 1971.

I don't mind that we have set up an agency to make sure that medical products and devices are safe to use and do what they say they are going to do. It's far more efficient than having to do it individually. And there's always, evidently, people who won't, because they trust their doctor or the brand. And, being medical products and devices, sometimes those people or their loved ones die.

I agree for the most part (I tend towards libertarian ideals if you've read any of my comments), but it's a difficult balance to maintain between allowing for medical innovation and preventing an objectivist distopia a la Bioshock. The government has a legitimate responsibility to ensure that medical technology is used and marketed responsibly, while at the same time giving companies some wiggle room to innovate.

Look how smallpox vaccine was discovered. Edward Jenner inoculated a young boy with cow pox, then attempted to infect him with smallpox a dozen times. Obviously we can't go around convincing young children to take infectious doses of deadly viruses. But we also cant send FDA thugs after every Silicon Valley startup pushing the boundaries of genomics research. So where do we draw the line?

There are lots of proposals to simplify FDA's approval process, for example, automatically approve it if it's already approved in a number of Western countries whose approval standards are high on average.

But in this specific case, seems FDA isn't being totally unreasonable. For example, from the letter

"Thus, months after you submitted your 510(k)s and more than 5 years after you began marketing, you still had not completed some of the studies and had not even started other studies necessary to support a marketing submission for the PGS"

seems like a good idea to require companies to do this

This sounds like a false dichotomy: "The government has a legitimate responsibility to ensure that medical technology is used and marketed responsibly"

So either the government ensures the above or it does not get accomplished? There are no third parties that could take this on? I find that hard to believe.

Non-government regulatory agencies aren't a cure all for complex products. For instance, check out the 2008 financial crisis. Sub-prime mortgage derivatives were rated AAA by private ratings agencies.

In fact, I would venture they are worse because they have to be paid somehow, and if they are paid by industry, you are ripe for regulatory capture. It can be difficult to get consumers to pay for it unless you include some kind of tax in the product they buy in which case you've arrived at a pseudo-government agency.

Possibly somewhere that takes into account the significant advances in laboratory science since the 18th century?

From what I hear from my co-workers in the quality department (I work in a medical device company) it's even harder to get approval to go through regulation in China than in the US. And you most certainly cannot just base your company from Asia and sell your stuff to the rest of the world (unless of course you go through regulation in the rest of the world).

You might be able to find countries where regulation is less strict or non-existent but I'd imagine that people's trust in medicine in said countries is less than in the regulated ones (as you most certainly would have rouge players abusing the lack of regulation).

edit: Added last sentence in 1st paragraph

it's not a lab test anyone will use to base important decisions on

Yet their adverting makes it sound like it is. Specifically talking about how it can be used find un-diagnosed disease and health problems.

I have some snakeoil I would like to sale you, $9.99 cures all ills

Oh, calm down Rand Paul.

Why be so snide, instead of actually writing an informative / useful reply? That would certainly take more effort, but this is Hacker News, there's supposed to be a higher standard of discourse here.

The original comment is nothing more than a generic libertarian rant, complete with the "gov" strawman, force-mapped to the matter at issue. It scarcely deserves more than the response to which you take objection.

What's the point of the down vote then, if not to enable an easy response (and influence) to something disagreeable without spurring entirely unnecessary and often negative comments?

Dismissing an ideology you don't agree with only highlights your own ignorance.

Copy and paste and argument from the internet if you can't form one on your own. Don't pretend "herp derp Rand Paul" is valid.

Reflexive yet vague anti-government sentiments deserve derision.

Wrong. Derision begets derision and tends to create an anti-intellectual feedback loop. There is no place for that here. Flag the people throwing insults around and move on.

You're right. I shouldn't have done that but I just couldn't believe that that was the top/first comment. It was one of my first posts and if I wanted to comment I should have taken the time to write an useful response.

Reply on news.ycombinator.com