Epstein: FDA Deprives "Informed Patients" Choice in Care, But So Does the Market

I'm wondering why the family couldn't simply contract away their right to file suit, effectively granting the company civil immunity. If the government isn't getting in the way, isn't that all that's left?

Or do they think letting the girl die untreated is better PR for them than the girl dying when they're treating her?

Wow! Start here: "Keeping drugs off the market deprives all informed patients the opportunity to correct FDA errors. Letting new drugs on the market leaves individual patients the option to decline their use."

When columnists or congressmen start carping about keeping lifesaving drugs off the market, they should be branded as "enemies of logic." They're simply begging the question. The FDA is the only entity empowered to decide whether or not a drug is life-saving. Not manufacturers, not lobbyists, not physicians, and especially not parents.

It is safe to say that there are no patients informed enough to make a decision about using untested and unapproved drugs. The sole exception might be physicians actively involved in the research who decide to use the agent on themselves.

The issue boils down to naive idealism (there's a miracle out there if we can find it) versus unlimited liability (this company will be sued out of existence if it takes risks with human lives.

Let's temper the naive idealism.

I have some points. 1st, it is unclear from the article that Neotropix is worried about a lawsuit. They may be worried about jeopardizing their $20M VC funding if the treatment does nothing for the little girl.

2nd: Coming from a blog about science, it is rather odd that you use an anecdote to respond to an argument about population-wide tradeoffs. "Would the population as a whole be better off if the FDA did not exist and all drugs could be bought without a prescription? Well, there is this one girl who may or may not be worse off, so obviously not." Sorry, but that doesn't cut it.

The alternative medicine example you use is stronger. But, your own argument avoids the question. Yes, many people waste money and some hurt themselves. Is the population as a whole better off? I would guess not, but I don't have any evidence of that.

Now, leaving aside strictly material considerations, how much does a desire for personal autonomy modify the argument. Personally, I would accept a small decrease in measured population-wide health for a large increase in personal autonomy, but not a very large decrease in population-wide health.

Howard,

When columnists or congressmen start carping about keeping lifesaving drugs off the market, they should be branded as "enemies of logic." They're simply begging the question. The FDA is the only entity empowered to decide whether or not a drug is life-saving. Not manufacturers, not lobbyists, not physicians, and especially not parents.

Nobody is "empowered" to be right about a factual issue. There is a large difference between saying that the FDA is likely to make the best assessment of the tradeoff between risks and benefits and saying that it is logically impossible for the FDA to be wrong; the former is defensible, the later is absurd.

It seems that people have a liberty interest in being able to decide for themselves what is "safe and effective". Take Vioxx for example. Many patients found relief from their debilitating pain when taking Vioxx and it was the only medicine that worked for them. Many peopke wanted to take Vioxx and they understood the cardiovascular risks associated with it.

Why a person should not be able to put potentially beneficial chemicals in their body when the manufactureer conducts safety studies and publishes the data so as not to mislead, is beyond my comprehension.

As long as manufacturers are not misleading the public and are conducting appropriate safety tests, why should the FDA make the decision on "safety"? Liberty should outweigh the government's interest in keeping people safe from themselves.

And in the case of Penelope, even if the family waives its right to sue and indemnifies the company, they will still need to report the outcome to the FDA. Using the drug in a patient for which it may never be recommended or even contraindicated, would jeopardize the development of the drug.

Here's a proposal. FDA review of new drugs should be optional. But if FDA approval as "safe and effective" is sought and granted, the result would be that manufacturers would be insulated from product liability for design defects (as opposed to manufacturig defects or failure to warn defects). This would act as a carrot to encourage manufacturers to seek approval. Of course manufacturers would need to disclose in big bold print that they have not received FDA approval if they haven't and they would not be allowed to advertise it as safe or effective.

The result is that people would be able to get the drugs they think are beneficial.

I've been looking for years, but I still haven't found a market full of rational, completely informed individuals!

Here's a dreary fact about life: getting power is not the same as keeping it. We have a widely-held myth that if somebody invents Mousetrap 2.0, the world will beat a path to their door; however true that may be, the appeal of the myth prevents us from seeing that the technology necessary to build Mousetrap 3.0 doesn't naturally follow. Inventions often occur by accident, or when the components come together like pieces in an unforeseen jigsaw (look up James Burke's Connections for plenty of examples). Giving the world Mousetrap 2.0 is no guarantee that the next set of jigsaw pieces will come together on your workbench rather than anybody else's. You have to invest in the process of change (a lesson Edison understood full well).

Investment leads to organization, and organization means that the players in the market are no longer human beings but emergent collectives whose properties are by no means necessarily the same as those of their components. (Cars move forward, but the traffic jam flows backward.) People can be kind, but that doesn't stop governments and corporations from becoming psychopaths.

It is not in the interest of the beings born from a free market to keep the market free. (One classic trick: use advertising to keep the consumer stupid.) Cases in point:

1. Cigarettes.

2. CFCs.

3. Greenhouse gases.

"Alternative medicine" might fall into the same scheme, too.

Blake, you're my hero as always.

He hits the problem dead on. I previously wrote about Epstein's article on the old blog, the problem here is that of the myth of the omniscient consumer.

DCB is putting forward the classic, and highly unrealistic, scenario of the world of informed consumers which never has existed and never will. The real reason people will end up choosing one drug over another is based upon which commercial appeals to them the most, whether or not it has a cute bee CGI effect, or an old person rising out of a wheelchair and dancing etc. The argument that Vioxx is the only thing that would fix someone's pain is absurd. The Cox-2's were the most over-rated and over-prescribed drugs probably in history. They were never shown to be more effective in pain relief than Naproxin, Ibuprofen, or Aspirin, the patient pool that needed a selective inhibitor was only about 1-2% of those who had to chronically take NSAIDS, and if anything, rather than being cardioprotective or decreasing side effects they increased side effects and cardiac complications while enjoying only a 50% reduction in GI complications.

To suggest that even informed consumers make rational choices based on knowledge of published side-effect profiles is absurd (a defense of Cox-2 inhibitors is a sure sign). Even physicians make terrible choices about which drugs to prescribe because even they fail to remain perfectly informed about the choices that exist. Case in point, Nexium. No physician who's done his reading would prescribe that drug. All it is is omeprazole for 8x the cost! It's the biggest scam ever. But physicians routinely prescribe it because they simply can't keep up with all the information on pharmacology. So when a patient shows up asking for the "purple pill" they saw on TV, the physician will happily change them from their omeprazole to Nexium without realizing all they've done is increase their patient's prescription drug cost 8x.

There is a need for a paternalistic body to say that when better alternatives exist, that the market should be limited to the safer drugs. The idea of the "omniscient consumer" is a tired libertarian myth. There is simply no way the majority of consumers will ever use real information to make rational choices about drug use, and there is simply no way the market will detect things like 1-2% increases in cardiac mortality over 6 months of drug use. We need agencies that carefully monitor drug safety, that approve drugs for efficacy, and make sure what reaches the market isn't just useful, but certainly not worse or more dangerous than current alternatives.

After all, we tried a world without the FDA. It was great. the pharmacologic arsenal consisted of several thousand patent medications that made big claims, sold real well and did nothing, except occasionally poison hundreds of people from things like ethylene glycol contamination (the incident that precipitated the formation of the modern FDA).

hey were never shown to be more effective in pain relief than Naproxin, Ibuprofen, or Aspirin, the patient pool that needed a selective inhibitor was only about 1-2% of those who had to chronically take NSAIDS, and if anything, rather than being cardioprotective or decreasing side effects they increased side effects and cardiac complications while enjoying only a 50% reduction in GI complications.

A 50% reduction in GI complications can be pretty significant if GI complications are preventing effective pain relief via more traditional NSAIDS. A medication is prescribed to treat a patient not a population. A drug that is only superior to the alternatives for a small subset of patients still will be clinically useful. The main fallacy that the Cox-2's benefit from is that prescription drugs are "more powerful" than OTC alternatives, when the distinction has much more to do with how safely the patients can dose themselves.

No physician who's done his reading would prescribe that drug. All it is is omeprazole for 8x the cost! It's the biggest scam ever. But physicians routinely prescribe it because they simply can't keep up with all the information on pharmacology. So when a patient shows up asking for the "purple pill" they saw on TV, the physician will happily change them from their omeprazole to Nexium without realizing all they've done is increase their patient's prescription drug cost 8x

This isn't an accurate representation of how prescribing is done in the contemporary context. I doubt there are many physicians who wouldn't know that nexium is in the same class as omeprazol and recommend the generic, especially given that the copays will typically be different. Even if the physician forgets, there is still the insurance company's formulary list which will determine if the more expensive drug would be covered when there aren't any counterindications for the less expensive generic in the same drug class and most software which supports e-prescribing include checks for less expensive alternatives. I think the primary way they sell these is with the free samples - it's even cheaper up front to get the patient started on the free samples of the new me-too drug than a generic in the same class, so the drug companies give out free samples and hope that the patient and physician don't bother to revisit what drug was prescribed when the samples run out.

@Delurking--you're right on #2. I haven't developed this theme yet, but am going to soon--the idea that the reporting in the WSJ is sometimes in direct oppositions to the pap on the opinion page. Today's Journal, for instance, discusses how CAFE standards were a failure, while in the Personal Journal section, the reporting explains that fuel efficiency has remained the same because manufacturers are using gains in techmology to create more powerful engines. I'll flesh out this theme later. Or maybe today. I'm tired from the redeye.

@David, I think the system that you describe is basically what we have with altie meds, right? "Claims made by [insert scam product here] not tested by the FDA..." on packages? There's a lot of misleading behavior there. There's a lot of misleading behavior even in drugs regulated by the FDA. I used to date a girl who worked for one of the big ones in market research. It was her job to figure out how to convince doctors to use a drug that was incredibly expensive, but barely effective. The way they did it was to suggest that not trying the drug was unethical, even if it had a very, very small chance of working.

Why a person should not be able to put potentially beneficial chemicals in their body when the manufactureer conducts safety studies and publishes the data so as not to mislead, is beyond my comprehension.

The short answer, David, is because acquiring information is very costly to consumers in the case of products like prescription medications.

And Chris, your comparison to the market for supplements/herbs/etc. is spot on. I was in a discussion of this issue over at The Volokh Conspiracy last week and tried to get the same point across--the market for alternative meds, being almost totally unregulated, is a very good illustration of what the market for drugs would look like if the "abolish the FDA" crowd like Epstein and Steven Landsburg got their way.

...tells the story of a 4-year-old girl who cannot get an experimental cancer treatment.

But think of the children!

Anyway, Epstein suggests that perfectly informed individuals will make perfect choices in this best of all possible worlds. I guess that's an argument. But what's even more interesting is that in yesterday's Journal, Saying No to Penelope tells the story of a 4-year-old girl who cannot get an experimental cancer treatment. Guess who's the boogeyman? The FDA? No! It's safety, but it's also the market. That most perfect of perfect perfections is denying a little girl her treatment. There ought to be a law!

I can't read Epstein's article because I don't have a WSJ subscription, but I'm skeptical that Epistein claimed that consumer decisions are perfect and based on the quote I strongly suspect that the actual claim was that if consumers are given the option to take drugs that aren't FDA approved the aggregate effect of those choices will be positive. This also is the heart of delurking's second point - the ancedote you cite doesn't speak against Epstein because he's talking about the net impact of a policy change in a population. Imperfect outcomes themselves are not grounds for regulation - any system of regulations will produce worse than theoretically possible outcomes too, so it doesn't prove anything one way or another, and the relative impacts of the approaches need to be assessed. Beating on a strawman of "perfect markets" does nothing to speak to this point.

There should be a blockquote in there from "Epstein" to "law!". My bad.

There should be a blockquote in there from "Epstein" to "law!". My bad.

A 50% reduction in GI complications can be pretty significant if GI complications are preventing effective pain relief via more traditional NSAIDS. A medication is prescribed to treat a patient not a population. A drug that is only superior to the alternatives for a small subset of patients still will be clinically useful. The main fallacy that the Cox-2's benefit from is that prescription drugs are "more powerful" than OTC alternatives, when the distinction has much more to do with how safely the patients can dose themselves.

One thing that troubles me about modern medicine is that even though we talk about how the pharmacogenomics is the future of medical treatment, there is still a tendency to treat patients as averages rather than as a sample of a variable population. This is particularly at issue with prescribing and reimbursement guidelines that sometimes mandate the use of the most cost-effective (on the average) drug, without considering the possibility that another drug might be better for a particular patient.

A somewhat mysterious aspect of Vioxx is that I've talked to quite a few people who have told me some variant of "Vioxx is a miracle drug! I've taken all of the other NSAIDS and none of them helped much, but I took Vioxx for a short while and my pain just went away." And I've spoken to orthopedists who have told me that they've heard the same thing from some patients. Of course, analgesics tent to have substantial placebo effects, but this seems a bit extreme. On the other hand, if you look at the clinical studies it looks like Vioxx has effectiveness more or less on the same level as other NSAIDS, with a little less stomach upset (and as we now know, more cardiac risk). I've never taken Vioxx myself, but I took Bextra at one point, and as far as I could tell it worked for me about as well as ibuprofen, pretty much as you'd expect from the clinical trials. However, I haven't heard such enthusiastic anecdotal reports about Bextra, although that might reflect the greater popularity of Vioxx. I think I'd be willing to try a short course of Vioxx, cardiac risk and all. I understand that some people at Merck have personal stashes squirreled away.

Matt, you are missing the point I was trying to make about the Cox-2 inhibitors. These drugs were being prescribed to everybody, at great cost, despite no proof of increased efficacy for pain relief above ibuprofen (if anything naproxen and aspirin are far more effective) based on a complication that crops up in about ~2% of users of Cox nonspecific NSAIDS. If these drugs were being distributed rationally, doctors would have prescribed the standard OTC cheap stuff, if the patients developed stomach pain you'd switch them to the selective Cox-2. But this never happened. There is a bias towards new and prescription meaning more powerful or more effective. People were taking these drugs without a preliminary empirical trial of the other NSAIDS, doctors were prescribing them like candy, and everybody though they were more effective than the NSAIDS because if it's prescription, it must be stronger. Yeah right.

Same with Nexium. I encounter people taking this crap all the time. It's one of the biggest sellers on the market (4.6 billion out of 22billion total market share). I would make the statement that 100% of prescriptions of Nexium are unnecessary, and wasteful since it is the exact same thing as Prilosec. Astra Zeneca got the process of isolating esoomeprazole from the mixed enantiomer patented by some act of God, now it's one of their biggest sellers despite being essentially the exact same thing as omeprazole! The trials that justified its use in patients compared 20mg of omeprazole to 20mg of Nexium and showed the efficacy "jumped" from about 86% in 3 months to 89% for relief of GERD. But all they really did was compare a 10mg dose of esoomeprazole to a 20mg dose! It's a joke, a valid comparison would have been 40mg of omeprazole(prilosec) to 20mg of esoomprazole (Nexium).

So, if people aren't really being prescribed this drug, and it's "no problem", how would explain the ~4.5-5 billion this sham drug rakes in every year? That's about 4 billion dollars of completely wasted money, down the drain, irrationally spent, completely stupid, and borderline criminal, a quarter of the PPI market. Every year.

Consumers are irrational, physicians can't keep up with the marketing and as a result there is this big stupid waste of money and not just in these examples, but all the time across multiple classes of drugs. There is no evidence consumers make informed choices, even the highly informed physicians are make poor choices. How is it that the average consumer or even a highly-informed consumer is going to figure out something like the Nexium trick? How many people even know what an enantiomer is?

I don't have a subscription to the WSJ, but it seems to me there is an argument to be made from Epstein's quote (I don't know if this is his position though).

The FDA has a tricky, hard job. It can theoretically make two types of mistake: approve an unsafe drug, and reject a safe drug. Any decision it makes is liable to be wrong, but one type of mistake is far more visible than the other - namely unsafe drugs that make the cut. Reducing these mistakes (a "First, do no harm" approach) likely raises the other. And where to draw the line is an individual concern - he obviously advocates for a riskier approach. This is a point that isn't often made and you don't need to be an advocate of complete deregulation to think that the FDA's current process is too strict in one direction.

Now, I think the dangers of his approach are many - and I don't know how much doctors and patients can be trusted to judge a new drug. In fact, I can see lots of informational asymmetries, principal/agent concerns, and other stuff that scream "market imperfection" when it comes to new drugs.

And the "causes far more harm than it prevents" hardly seems right. I don't think there is any evidence at all about that.

To me, this sounds like taking a reasonable argument and taking it to a far-too-extreme position in the hopes of shifting the outcome closer to your own preferred point. I won't defend it, but it seems to be effective.

Mark,

I think we actually agree on the substance of the Cox-2 issue, although we put different emphasis on the significance of the subset of cases where GI problems would be an issue.

On the issue of nexium, I didn't claim that it wasn't overprescribed, just that it probably has more to do with free samples than a lack of physician knowedge of the PPIs out there. Also, the issue with nexium has relatively little to do with safety or efficacy regulations and more to do with the IP rules. It definitely will be as safe and effective as the unpurified mixture, and hence should be marketable, so the question is how much of a IP-based subsidy should be provided for me-too style drugs. I personally think that it should be shorter than drugs that exploit a novel mechanism of action, but not eliminated entirely since sometimes even minor changes can result in clinically useful improvements. I'd contemplate linking it to some form of demonstrated improvement vs the current standard, but not all of the potential improvements may show up in the comparisons made in the normal approval process.

They have about 20% of the total PPI market and 4.5 million in sales. It's not free samples. It's sales.

The example given by the "Saying No to Penelope" is actually a perfect example for why Epstein is dead wrong.

The drug that they are testing is a live virus. It has been just barely phase I tested. For folks who aren't familiar with what phase I tests are, it has just been tested for safety, not efficacy. That said, live viruses require a fair amount more safety testing than many other types of treatments, so they're still not done with phase I testing (and they have to proceed slowly, in case they find out it's not safe). So they still don't even know if the virus is safe. *If* it turns out to be safe, then they'll test to see if it works. How many cancer treatments get past phase I testing only to fail at phase II? Lots. Most of them. (I'm too lazy to look up hard numbers at the moment, but perhaps Mark can help out with this one). So, so far, it's a treatment that we don't know if it is safe, and we don't know if it will work. And someone wants to put this in their kid? Good thing the FDA is protecting the child from their dumbass parents.

Ok, finally, this company is expending a lot of resources to make and purify a very small amount of virus in a way that it can be injected into patients. They are not in *production* mode. The people purifying this stuff are researchers, not manufacturers. So it would be rather like having someone walk into my lab, and ask me for 10,000 gallons of bacteria for industrial use. We simply don't have the facilities to make that much. I can grow 3 or 4 gallons at a time, if I go all out. And it's simply not worth the company's time or money to build a larger facility *before* they demonstrate that the treatment will be safe or effective.

Let's imagine that scenario. Parents sign paperwork saying they won't sue. Company builds a large production facility, and starts making drug available to desperate parents. Even if people continued to sign up for drug trials, what happens when they demonstrate that the live virus is not only not safe, but it doesn't work? Now the company, who is about to go bankrupt, owns $50 million worth of production equipment that will never see use. Investors decide that putting money into biotech is a bad idea, because dumbass scientists waste all their money on building production facilities for cures that don't work.

Most Americans don't realize how grateful they should be for the amazing work done at the FDA.