Balaji Srinivasan, who may run the FDA for Trump, hates the FDA | Hacker News Reader

Balaji Srinivasan, who may run the FDA for Trump, hates the FDA

recode.net

65 pointsdevy117 comments

There was a book written near the death of Harvey Wiley, first head of the FDA, where he was already talking about regulatory capture by industry, https://en.wikipedia.org/wiki/Harvey_Washington_Wiley & https://www.google.com/search?q=the+history+of+a+crime+again...

> Wiley soon became a crusader and coalition builder in support of national food and drug regulation. His work, and that of Alice Lakey, spurred one million American women to write to the White House in support of the Pure Food and Drug Act ... Wiley was nicknamed the "Father of the Pure Food and Drugs Act" when it became law in 1906 ... On March 15, 1912, Dr. Wiley resigned his leadership of the Chemistry Bureau because, from nearly the beginning, he had been antagonized in the enforcement of the Pure Food And Drugs Act, and worse, he had seen the fundamental principles of that act either paralyzed or discredited ... After his resignation from government work in 1912, Wiley took over the laboratories of Good Housekeeping Magazine, where he continued his work on behalf of the consuming public.

The first head of the FDA resigned to work for a private laboratory that evaluated consumer products. Smartphone spectroscopy ("tricorder") with cloud analytics is being productized to enable mobile identification of food & drugs, http://mashable.com/2017/01/07/smartphone-with-scio-sensor-c...

Can new regulatory frameworks lower barriers to innovation (in drug discovery and feedback loops for consumer safety), beyond large incumbents and VC-backed challengers?

It should be abundantly clear by now that many of the incoming administration's cabinet/agency picks are aimed at dismantling the federal administrative structure, regardless of any scientific or practical reasons for how the structure got to be that way to begin with. Good times for snake oil salespersons, bad times for anyone else.

> Good times for snake oil salespersons, bad times for anyone else.

Odd thing to say as it has always been a good time for snake oil. FDA is riddled with exemptions.

Water, sugar, food color and I'm over here 'preventing' ailments naturally.

Reagan days are here again.

>It should be abundantly clear by now that many of the incoming administration's cabinet/agency picks are aimed at dismantling the federal administrative structure

I sure hope so.

I think the best thing the FDA could do is create a 'Conceal Carry' type of license which allows you to self prescribe certain types of pharmaceuticals (all would be great, but start with some). You basically take an 8-12 hour course and then pass a test, sign a piece of paper waiving the right to sue over non-compound, non-quality issues, and then get a license which allows you to purchase certain pharmaceuticals on your own. You could in theory reduce liability for doctors and pharmacists, opening up some money for something else like technology. Would boost sales for certain drugs and if done right, could result in a healthier more productive society.

I think this would be great. But I think the course should be per drug. For instance what you need to know to self prescribe MAOIs is very different from what you would need to know to self prescribe SSRIs.

It would lead to more people using drugs that are are highly effective but not prescribed for liability reasons like MAOIs.

This would be a recipe for disaster.

Anybody who deeply scrutinizes the FDA will find opportunities for reform. For instance, when the economists Daniel B. Klein and Alex Tabarrok applied their scrutiny, the result was http://www.fdareview.org/

What is surprising and exciting about the possibility of Balaji's appointment is not that he has identified opportunities for reform, it is that he may be appointed despite the fact that reform is his primary motivation!

He has no political motive, no personal motive, he does not "hate" the FDA (he is not the kind of person to be driven by hate)... he simply sees the FDA's mandate as an incredibly important one which it could do a better job of fulfilling. This should inspire optimism.

I like that he's had a successful startup in the field, and is generally smart as hell: https://balajis.com/about/

Balaji S. Srinivasan (BSS) is a computer scientist, investor, entrepreneur, and academic. Dr. Srinivasan holds a BS, MS, and PhD in Electrical Engineering and an MS in Chemical Engineering from Stanford and taught data mining, statistics, and computational biology in the Department of Statistics at Stanford University. He was an NDSEG, NSF, and VIGRE fellow, was named to the MIT TR35 and Founders Fund F50 lists, and has been published in the New England Journal of Medicine, Nature Biotechnology, and Nature Reviews Genetics. He was the CTO and co-founder of Counsyl, a genomics startup that began in a Stanford dorm room and now tests ~4% of all US births. Counsyl won the Wall Street Journal’s Innovation Award for Medicine, was named one of Scientific American’s Top 10 World Changing Ideas, raised more than $65M in funding from David Drummond, Aydin Senkut, and Founders Fund, and has become one of the largest clinical genome centers in the world. After ten years in genomics and five years at Counsyl, he stepped back from an executive role in November 2012 to work on other areas of technology. He now invests in and advises startups, teaches a popular MOOC at startup.stanford.edu, runs the Stanford Bitcoin Group with Professor Vijay Pande of Folding@Home, and writes at balajis.com.

The "Yelp for drugs" pull quote is missing the second part: "with MD star ratings." i.e. only medical doctors would be able to review/rate. Which IMO is a reasonable and interesting idea worth exploring.

Imagine the little perverse incentives that companies offer to leave a Yelp rating (rate us and get some free breadsticks! or whatever), but with pharmaceuticals (leave us a five star rating and get a new car, or a new medical device, or whatever). That sounds really, really bad, and it would require a lot of regulation and oversight to make sure ratings weren't being manipulated. I suspect those two things will be in short supply in the incoming administration.

You think FDA employees are immune from this type of thing? How do we go about finding the angels needed to staff this agency?

No, but those elaborate FDA approval procedures are designed to reduce the risk of corrupting human influence much like any other scientific protocol that aspires to reproducibility.

Doctors, like lawyers, accountants, pharmacists and so on have professional associations with existing and quite strict ethics regulations about this sort of thing.

That hasn't stopped "buying expensive stuff for doctors to sway them" from being a thing, even now!

http://www.wsj.com/articles/SB100014241278873246951045784145...

I didn't say the existing rules were perfect, I'm just noting they exist. You can also say that a "Yelp for drugs" wouldn't change the existing incentives.

FWIW, the doctors I know take ethics very seriously; they take great pains not to be seen as accepting gifts, they treat pharma company claims with a great deal of reasonable skepticism, they always prefer non-drug and OTC treatment to prescriptions, and generics to branded drugs. A few bad apples do not spoil the tree.

Then why do we regularly see stories about doctors engaging in epically fraudulent billing practices, taking illegal kickbacks from drug manufacturers, prescribing unneeded drugs and so on? Have you heard of the opiate crisis? Because if so you cannot but be aware of how much of it is driven by prescription rather than black market drug sales. Not to put too fine a point on it, the drug market in the US is a total shitshow, and taking your comment at face value requires ignoring a mountain of evidence.

Please don't insult our intelligence with your argument from ignorance. If you actually believe what you wrote above, then you are massively under-informed and should pay more attention to the news.

You seem to be a very angry person.

We see stories about that because we are a country of 330 million people and there will be some subset of any [group of people] who do bad things in such a large population. The vast majority of doctors are highly ethical.

As for opiates, I'm afraid I don't quite fit into your strawman; I think doctors should be banned from prescribing them on an outpatient basis to anyone but patients with a terminal prognosis. Chronic pain is awful, but opiate addiction is still worse. Since the action (and addictive mechanism) of opiates causes the pain to worsen over time, it's not an appropriate treatment.

The main reason for overprescription of opiates is not some evil doctor conspiracy, it's because doctors, like everyone, have a very hard time telling people "no." And there's no group of people more skilled at emotional manipulation than junkies.

You wouldn't be saying that as a chronic pain sufferer. They are the only thing that makes my life livable. With them, I'm a competent member of society. Without... well, I'd be dead, thus negating my entire social and economic investments by society. Seems pretty stupid.

Overprescription of opiates is also a cause for "junkies", not just a consequence of their skills in emotional manipulation.

It's a feedback loop. I suspect the ur-cause of the current epidemic is higher rates of depression related to an aging population and chronic un- and underemployment.

[...] only medical doctors would be able to review/rate. Which IMO is a reasonable and interesting idea worth exploring.

This is still an incredibly bad idea, for the entire reason we have medical science. One physician, no matter how principled and rigorous, won't have enough data. This "data" comes by way of raw number of patients, information collected about each of those patients (general and relevant to a specific drug study), how that data was collected (i.e. good experimental methodology).

Worse, even if good data were magically available, random physicians won't have the background in experimental science and statistical methods to analyze that data meaningfully. Even trained research scientists often have problems with this stuff. Add to that the already extant risks of conflict of interest from physicians receiving kickbacks from big pharma... sheesh.

So yeah, Yelp Anecdata FTW! Welcome to the tyranny of the snake-oil salesmen!

Do we really want pharmaceutical medicine, which already has major systemic problems, to degenerate to the level of "audiophile" arguments on web forums?

Oh because people who have MDs are incorruptible? Be realistic, lots of doctors take illegal kickbacks from pharmaceutical companies. As a Euro I can't believe that pharma companies are still allowed to advertise direct to consumers on television. They claim first amendment protection, but advertising is not the press and commercial speech should not be subject to the same legal protections as that of individuals.

Are the people at the FDA incorruptible?

No. Do FDA trials involve elaborate administrative procedures designed to overcome that problem? Yes, and most of the time they work OK. Go on, tell me how easy it is to rig an FDA trial.

I don't know precisely how easy it is to "rig" an FDA trial, and I don't know precisely how easy it would be to "buy" enough MD reviews to get a good "Yelp" rating for an unsafe treatment. I suspect both are quite possible given enough money, but I don't know which would be easier.

A responsible MD wouldn't rate a drug without basing their rating on a review of the research. Seems redundant to me.

Pharma corrupting MDs has been routine for quite some time. The restriction doesn't do anything but focus the target of manipulation to exact the population Pharma has spent the most effort learning how to manipulate.

Are you saying that because only doctors would be able to give ratings, those ratings would be more credible? I agree that __should__ be the case, but it wouldn't/isn't (http://jnci.oxfordjournals.org/content/99/15/1148.full). I cannot find the source, but I once read that there was a very small minority of doctors in the country subscribing the most brand-name drugs simply because of perks offered by drug companies.

Just because doctors have to deal with moral hazards (and sometimes fail) doesn't mean that non-doctors are more reliable sources for information about pharmaceuticals. This should be some form of cognitive bias (maybe it is and I just don't know the name for it): someone thinks something is a black and white issue, and has settled on "white" (ex.: doctors always give the most unbiased information about pharmaceuticals). Then, that person learns of a gray area in what was previously thought to be black and white (ex.: pharmaceutical companies aggressively market to doctors). Then, the cognitive bias: rather than revising their opinion about the binary, black and white nature of their belief, they keep believing it's all binary, but the evidence of the gray area causes them to flip their opinion from "white" to "black" (ex.: because pharmaceutical companies market to doctors, doctors are now considered altogether unreliable). When the reality is that even a potentially compromised doctor is most likely (not always, just most likely, but we have no way of identifying the exceptions) going to be a better source of pharmaceutical information than the collective wisdom of the rest of humanity.

Why would you replace research into its effects with a star rating from MDs based on what? Their gut feeling? Them testing it in the wild with no controls? These are human lives we're gambling with now.

Which is ridiculous, Pharmacists are often far more knowledgeable about medication, and even then crowdsourcing reviews for medicine is a crapshoot.

Try it, test it, see if it works.

OK, you be the first guinea pig. Or you can nominate one of your family members if you'd prefer.

Has it never occurred to you that we have already tried this approach in the past before the FDA was set up? OK Yelp didn't exist then but but the medical community's endorsement has long been used as a marketing tool - 'more doctors smoke camels.'

Why do you think the FDA was set up in the first place? Because some communists decided they needed a way to keep sick people in worse health for longer, or because the general public was sick of the uncertainties that inhere in patent medicines and the like?

You could do that, of course. But you should absolutely not depend on it.

And drugs don't work like restaurants, of course

I met someone born with no arms due to exposure to thalidomide [2] as a fetus several years ago. Governments in Europe had approved its use as a morning-sickness pill without adequately reviewing its side effects, and thousands of children paid the price. The FDA, at the time, did not approve thalidomide for use.

Eliminating the FDA is a far cry from evaluating its performance, the latter (I think) many people agree is necessary. The former will only result in public experimentation and, consequently, suffering.

[2]: https://en.wikipedia.org/wiki/Thalidomide

Far more people die because of drugs the FDA does not approve, when compared to deaths avoided by it, despite anecdotal evidence of the contrary.

We hold people and groups accountable for the things that they do, not the things that they don't do (except in cases where inaction is a form of action).

It's wrong to rob someone, but it's not wrong to be afraid of stopping a robber as a bystander. It's not a good thing that people die because of the FDA's long approval process, but it's not blameworthy in the same way that death and suffering caused by hastily approved medicines (like thalidomide) is.

This may change once the barrier for market entry is lower. What would stop suppliers from flooding the market with dangerous drugs? I am sure there are areas where the FDA can be improved but we should be careful.

But he gave a great talk a few years ago:

Y Combinator: Balaji Srinivasan at Startup School 2013

https://www.youtube.com/watch?v=cOubCHLXT6A

https://twitter.com/TonyRomm/status/821105716069023744

> Scoop: Balaji Srinivasan met w/Trump, but the VC is not up for FDA slot, source close to transition tells me.

There should be a Yelp for restaurants too, then we wouldn't need food safety regulations.

What's to stop "innovative drug companies" from buying fake reviews to boost their numbers? It already happens in places like TrustPilot and others.

It will frown on fake reviews, but there will be an option to pay to remove bad reviews posted by the widows of the deceased stating that their loved ones died from complications of a foodborne illness at that establishment. The widows were more "adventurous" and thus survived.

When do I get my board seat?

Right now, not much. The internet and academia both lack a good solution to information manipulation.

A robust decentralized reputation system would go a very long way in solving a huge number of problems. I don't think anyone is that close though. I feel like we are just one major breakthrough away, but that breakthrough is likely to take a form that nobody is expecting.

The way to solve it is to have different branches with checks and balances. The FDA is one such branch that should, in theory, not care about anything other than consumer safety, the legislative branch limits it's powers so it doesn't go too far and stop innovation, and the researchers try and beat the best of FDA's powers with evidence that their drug is safe.

If something is broken there, it's because of leaky incentives (ie, people not really caring about the consumer on the FDA because of a revolving door with industry) or because it's imbalanced (the FDA has too much power and it can shut down drugs with legitimate evidence of safety, for example, or not enough power, or drug companies don't have enough money anymore to convince the FDA, etc)

The leakiest incentive is that the FDA gets in a huge amount of trouble if it lets through a drug that has unforseen negative side effects, but faces essentially zero consequences for holding up the approval of a worthwhile drug indefinitely, and requiring billions of dollars to be spent on its development.

> but faces essentially zero consequences for holding up the approval of a worthwhile drug indefinitely, and requiring billions of dollars to be spent on its development.

The FDA does not hold up worthwhile drugs indefinitely. It does take time and money to scientifically confirm the safety and efficacy of drugs. The whole point is to make sure drugs are safe and effective -- the FDA does a pretty good job of that.

Sure the pharma industry would have you believe it's just a bunch of pointless roadblocks. I would like for you to point out a regulation or requirement of the FDA that you think should be eliminated.

The problem is that the pharma industry wouldn't have you believe it's a bunch of pointless roadblocks. The pharmaceutical industry, as currently constituted, generally thinks the FDA's system is great, because it means that not only are smaller, novel, and foreign competitors easily closed out of the market, but that they have long-term sole approval for their drugs. The large drug companies find this an extremely comfortable relationship - it's stable, predictable, and they are well-practiced at it.

I think that it's not a problem to have an FDA which reviews and approves drugs, and even one which determines which drugs are qualified for Medicaid/Medicare payments. But what I would like to see is an "opt-out" system, where patients could sign a waiver declaring that they are aware that the drugs they would like have not been approved by the FDA and that they may be hazardous, to access otherwise-unapproved drugs. I think there are genuine costs to the current system that are too commonly dismissed: http://www.jpands.org/vol15no4/kazman.pdf

I'm not saying "abolish the FDA", I'm just saying "allow people to bypass its judgement if they really feel that they have to".

What wonder drugs are on the market in other developed countries that US consumers are dying to get hold of? If this is a big problem it should be easy to name several.

I didn't say they were on the market in other developed countries, since the drug approval bodies in those countries tend to face similar incentives. My point is that the process which requires drugs, even many existing generics, to go through extremely extensive testing for each possible application, is a problem because it means that only very large companies have the financial wherewhithal to get drugs through approval, and there is a strong incentive to focus on drugs which will be big sellers in a large market, rather than drugs for smaller markets.

The trouble is that one does not see what was prevented from existing.

Then I'd say the primary problem is to reduce the cost of an access to clinical trials or improved flexibility in the trial requirements proportional to the size and plight of the patient population. The 'similar incentives' you mention are wanting to avoid public-health crises of the sort we have experienced many times in the past.

The trouble is that one does not see what was prevented from existing.

Opportunity cost is a basic fact of economics. It seems to me that people put a higher premium on reliability than novelty in this sphere; that is to say, the cost to society of people dying from terminal disease is less than that of people dying unexpectedly after treatment for less deadly conditions. Also, your statement assumes those cures will never be discoverable.

Fake reviews are part of the innovation.

Instead of the FAA we should have a Yelp for airlines. After a few plane crashes the market can easily decide who's the safest!

How about Yelp for politicians? No need for elections.

I know you jest but it's being tried http://www.politicalbank.com

That looks a little more like GoFundMe for politics, than Yelp.

Yelp for everything, then we won't need anything!

Looking forward to a Yelp for Yelp Clones, where I can rate yelp-for-X sites on how trustworthy the ratings are.

How about a Meta-Yelp for Meta-Yelps that don't Meta-Yelp themselves?

Ok people, time to think a little outside the box now - are you ready!?? How about NOLP! It's very zeitgeisty, but innovative.

I think the rating FDA's efficacy requires a strong understanding of Bayes' theorem. You could look at the organization as a test itself. Like many tests you've got false and true positives, false and true negatives, sensitivity and specificity. https://en.wikipedia.org/wiki/Sensitivity_and_specificity

I find this this topic easy to misunderstand, personally, so I'm wary of making statements as to whether the FDA should be more lenient, or exactly how they should improve.

Wow, this might be Trump's worst pick yet. Dude took $120M in VC money and just flushed it down the toilet.

I suppose we should write off the leaders of all the other failed entrepreneurs in Silicon Valley as well? A place where 9 out of 10 startups fails.

I would say Srinivasan is particularly egregious. Regardless of how you feel about bitcoin, and even among many bitcoin enthusiasts, the economics of the 21 bitcoin computer were clearly absurd from the outset, yet Srinivasan still managed to fleece million of dollars from investors using buzzwords and guile. I think he is fundamentally untrustworthy and a terrible choice for the head of a government agency.

That's not grounds for writing someone off, but it's not always a positive thing either. I would fail miserably at running a marathon, but I don't think that gives me any insight into running marathons. Going someplace where everyone fails and (surprise!) failing might also just as easily be a sign of poor judgement as the failing itself, depending on the circumstances. Tons of rich kids who can afford to fail come here and attempt it just because for them, failure has minimal consequences, and they figure they'll either win the startup lottery or give up in a few years, and worst-case they will have spent a few post-college years having fun working with their buddies and delaying getting a "real" job (in their view). Other people might get much more out of their failed startup experience, but there's just a really wide range of who does it for what reasons, and what they get out of the experience.

> we should write off the leaders of all the other failed entrepreneurs in Silicon Valley as well?

For running government agencies? Yes. Such risk-taking does not belong there.

Why not?

A government agency can't declare bankruptcy.

For a public leadership role I would prefer someone who has a solid track record of success. It's one thing you burn VC money but running the FDA requires a different mindset.

Startups =\= Government

Well, when you put it like that, yes, all of them.

There's a difference between giving an honest try and failing, and coming up with THIS:

https://www.amazon.com/21-INC-21BC1-Bitcoin-Computer/dp/B014...

(Yes, that IS a Raspberry Pi that they're selling for $150)

Your comment is so misleading. That is not a Pi for $150. That is a Pi as a convenient interface to a 50 Gigahash ASIC, and blockchain store.

So is yours. That "50ghz ASIC" isn't even going to generate enough BTC to even pay for the power it uses, much less actually make money. So it literally might as well not be there.

Well, the GH/s is currently around 2500000000 GH/s so contributing 50 GH/s you would have a 1 in 50 million chance of mining a bitcoin every 10 minutes. In a little over 951 years you are expected to mine 1 bitcoin.

Something tells me bitcoin isn't going to be a thing in 951 years.

I suppose it's a small blessing that he didn't pick Martin Shkreli. Or Elizabeth Holmes.

Perhaps one of them can be surgeon-general. I'm actually sort of surprised la Holmes hasn't shown up in the lobby of Trump Tower yet.

Dude also started a > $1 billion dollar company out of his dorm room: https://counsyl.com and taught one of the most popular MOOCs (Startup Engineering on coursera).

It also is a bit premature to write off 21.co as dead

His "replace the FDA with a Yelp for drugs" is obviously an extreme position, but it's the right direction to explore in. There are many consumers who can make intelligent tradeoffs about the risks and benefits of not-yet-approved drugs, and they should be able to do so.

A counter-example: Thalidomide. Refused approval by the FDA multiple times (the drug company kept resubmitting it), there were only 17 children born with Thalidomide-related birth deformities in the USA.

Promoted as a wonder drug for morning sickness, there were 10 thousand children born with birth deformities worldwide.

https://web.archive.org/web/20090512235601/http://www.fda.go...

Perhaps memories of this are shorter in the US precisely because the FDA got this right and there are fewer visible daily reminders why we need drug regulation.

The FDA does excellent work on safety from a first-do-no-harm perspective. From that perspective, everything is great.

Still, more than 100 million people die every year from diseases. Even with the best available health care, few live beyond 90. So the very best outcome imaginable has not been achieved.

The risk with the current FDA is that by minimizing the risk of being harmed by new drugs, we've locked in the current life expectancy of ~75.

> we've locked in the current life expectancy of ~75.

This is just false. The life expectancy has been rising through modern history, except a stagnation in 1993 and one this year. You can't point to 1 data point in 50+ years and say we're locked in. Look at the record with the FDA (data from world Bank)

https://www.google.com/search?q=us+life+expectancy

Granted UK and Japan are doing better. They also have healthier lifestyles in general and strong regulatory agencies.

> Even with the best available health care, few live beyond 90. So the very best outcome imaginable has not been achieved.

What if, on the other hand, we have achieved the very best outcome that is safely attainable? We certainly don't want the needle of achievement to start moving downward.

Exactly. When I was growing up in teh Ireland in the 1970s I knew of several people around my (small) town that had the most horrific physical deformities as a result of their mothers' thalidomide use. Perhaps it was a little more visible in Ireland because abortion is illegal there and so pregnancies that would have been terminated in other countries were carried to term there, regardless of the known risks or consequences. But any time someone tells me we don't need something like the FDA my mind goes straight to a picture of a guy with his hands growing out of his shoulders that freaked me out when I was a kid.

Anyone, and I don't care how distinguished they are, who talks about drug approval policy without examining cases like thalidomide or DES is a liar. There is no way anyone with the legal/economic insight to have a rational conversation on alternative regulatory structures is not aware of those cases.

My mom almost took it when she was pregnant with my sister. Pretty scary.

I think I'm a pretty average consumer and I can say without a doubt that I would not trust my ability to make a well reasoned choice with regard to unapproved drugs. I'm not a pharmacist.

"The other reviewers saying you can't drink while taking Tylenol have no idea what they are talking about. I do it all the time and it's totally safe."

I'm not quite sure what point you're trying to make. I was under the impression that normal doses of Tylenol taken infrequently after drinking alcohol is not measurably dangerous.

Taking acetaminophen after drinking is extremely dangerous. It's potentially the leading cause of liver damage in the United States.

The thing is that the damage only occurs in a specific instance, that is somewhat rare even if taken with alcohol. It occurs when the typical biochemical pathway for processing tylenol is saturated, which typically doesn't happen, but when it does it causes a lot of damage very quickly.

A lot of drugs are extremely dangerous when not taken as prescribed, but the danger exists only in edge cases. So you could have 99 people saying "this drug is fine" and the drug will kill that 100th person every single time because of a particular mutation in a biochemical pathway.

The Yelp method is a stupid way to regulate drugs because making correct risk assessments isn't something non-professionals can do. It's not like Amazon, where a buying a charger that mysteriously doesn't work with your phone is an inconvenience -- people are risking their health and their life.

The site would be a bad idea even the power to review was limited to MDs. MDs are not all equally qualified to review drugs. Even if they know a lot about their specialty, they don't understand everything they need to know to prescribe drugs safely. This is why we have pharmacists. Their job is to basically understand all of the ways drugs work to predict novel interactions. Dumping a bunch of mystery drugs on the market is going to destroy that safety net. Furthermore, it's insane to think you can properly measure the efficacy of a drug by having doctors write reviews about it. Double-blind studies are a necessity for a reason.

I don't disagree with your broader point that "Yelp for drugs" is insufficient, but I'd still like to see some sources for your specific claims about Tylenol. From what I can tell from quick Googling, liver damage is quite rare unless the dosages are well beyond the recommendation, or taken habitually, or combined with an already extreme and dangerous dose of alcohol.

One argument is that the FDA is too conservative. Take the example of a terminal cancer patient. There might be an experimental treatment that would offer them another month or three, or it could kill them immediately. If they are okay with that gamble, then why not let them? If it works out then we as a society have gained valuable information about a new treatment option. If it doesn't work out, we have also gained valuable information.

Or, for a less life-and-death example, the FDA shouldn't be overly concerned with sunscreen ingredients [0].

I don't think the FDA should go away, but I think there is room for improvement. Maybe we could have some harmonization with similar agencies in EU countries, automatically granting approval for drugs available there. Crap, maybe if we just did it for sunscreen it would be a win for consumers.

[0] http://cen.acs.org/articles/93/i20/Decade-FDA-Still-Wont-All...

You've actually made an argument for the FDA, not against it. Otherwise patients with cancer would have to decide between thousands of chemical concoctions which will either kill them, let them live, or do nothing, in which case they will still only live or die. Why would there only be one experimental drug on the market if any one is allowed to market to patients independent of efficacy or risk?

http://www.fda.gov/drugs/developmentapprovalprocess/howdrugs...

You think the average consumer is going to read (let alone understand) dense clinical trials and review data from NIH and elsewhere to make their own determination on the risks and benefits of a drug? LOL, give me a break.

Efficacy and patient safety matter. That's why the FDA exists.

There are many consumers who think they can make an intelligent decision when it comes to anything medical, not that many actually can. I'm personally willing to trust doctors and scientists more than random people reviews.

I don't know, some random anonymous reviewer on the internet said this was pretty good! I'll take it and see what happens!

Not-yet-approved implies that the research isn't finished. How are you going to 'intelligently' assess the risk?

By getting sick and dying I suppose, but you before me, so I won't make the same mistake. I honestly can't think of any other alternative.

A ratings website is not a replacement for controlled experiment. The FDA and the current Pharma industry have problems, but this is in no way an intelligent alternative.

There are several stages of research. For a sufficiently serious condition, it can make sense to take drugs that are only validated at a pre-clinical stage.

And there are many conclusions of finished research. A common conclusion is "cures problem A, but may cause problem B", and depending on A and B, those drugs can fail approval. But individuals may place different values on A and B.

There might be many who can, but those who can't and/or will fall for the even most stupids scam will be multiple more.

Adding to your point, the thing that is often overlooked is that those people who are unable to exercise informed judgement are also a huge burden on society as a whole. The ailments resulting from their "stupidity" means:

- they're not as economically active

- they're dependent on healthcare, driving premiums upwards and placing undue burden on medicaid

- many have minors and other people who are unable to make informed decisions in their care, etc.

I'm troubled by the "us vs them" mentality that seems to pervade the superficially-reasonable position of "let the consumer decide". Nobody wants a nanny-state, but a free-for-all isn't good for anyone, either.

Adding to your point, the thing that is often overlooked is that those people who are unable to exercise informed judgement are also a huge burden on society as a whole.

Hmm, that, and the rest of your comment, seems to be suggesting that those people are expendable. That society is better of without them.

Nobody said they were expendable, just that their low intelligence* and the resulting poor choices cost the smarter people a fortune (because of our moral and legal obligation to take care of them despite the entirely predictable nature of their morbidity). If you're not willing to acknowledge that fact then you've abandoned reality for ideology and are engaged in politics instead of science.

* which term I am using as a catch-all for everything from cognitive ability to critical thought skills to quality of information sources, rather than as a proxy for IQ.

How on earth did you get that impression?

The point is that letting people make bad decisions is not "just" bad for them -- it's bad for society as a whole.

Example: See "anti-vax" movement.

Doubtful. Probably nobody, not even medical experts, can make a reasoned choice about all medicines they need or want to take in their lifetime. After all, nobody can be an expert on all those things.

EDIT: erasing my non constructive troll comment but leaving the rest.

We are talking about people who think vaccines cause autism and people who believe in homeopathy. Do you really think 'many consumers' can do the research and make an intelligent judgment about modern pharmaceuticals? I consider myself pretty smart and I can't tell you how drugs interact or what causes them to work.

Isn't he the one who spent millions of investor's money only to build a useless bitcoin computer, which is basically a raspberry pi with a mining chip which only mines a few satoshis an year?

I'm not looking forward to people buying experimental cancer drugs from their local homeopath...

I'm not an expert here, but the notion that dismantling the FDA would unleash medical innovation seems to ignore that there are already many countries around the world with much less stringent drug review than the United States. If that were a significant advantage, wouldn't these countries be showing measurable gains in life expectancy or some other metric?

I'm someone who would look forward to being able to buy drugs without a prescription. Doctors are expensive and in many cases useless. I have low testosterone. It was easy to diagnosis via a blood test. It's easy to treat via simple drugs, and it's easy to follow best practices(every 6 months get a hematocrit, and testosterone test) But I have to drive across town and spend $200 every 6 months for some guy to do the same thing.

But you're right there is the issue of it's hard for an individual to consent without being informed. So I think the manufacturers should be responsible for putting out a course that teaches and informs the end-user about the drug, and it's risks. (of course they would be legally responsible for any unreasonable omissions, or reasonable misunderstandings)

To be fair only failed countries have poor drug regulations. There is no advanced country (able to do quality medical research) without strict regulations.

They have much cheaper cost of health care.

American life expectancy is decreasing.

Oh no! People might start taking drugs not explicitly approved by the government after an extremely long, expensive and monopolizing approval process, like kratom or something crazy.

Can't have that.

Hey mods: can we reinstate the no politics rule, please? I know I could just ignore these threads, but for some idiotic reason I can't seem to help myself from peering into and contributing to these pointless debates where everyone essentially preaches to the choir and comes out feeling frustrated and angry at the other side's stupidity/ignorance/dishonesty/corruption.

One of my New Year's resolutions was to mercilessly unfollow anyone on Facebook whose posts were over 66% political -- no matter whether Republican or Democrat. Now I'm facebooking like it's 2005, and I gotta say, I'm liking it. It's so nice looking at happy people on holiday with their families and kids rather than mulling through all the political hatred and divisiveness. I never thought I'd say that, but such is the degree of poison afflicting political discourse that anything to block it out feels like desperately needed palliative.

You took a courageous decision to ban politics and to my mind it was bliss... more of the same please!

Reply on news.ycombinator.com