Ketamine: Now By Prescription

Last week the FDA approved esketamine for treatment-resistant depression.

Let’s review how the pharmaceutical industry works: a company discovers and patents a potentially exciting new drug. They spend tens of millions of dollars proving safety and efficacy to the FDA. The FDA rewards them with a 10ish year monopoly on the drug, during which they can charge whatever ridiculous price they want. This isn’t a great system, but at least we get new medicines sometimes.

Occasionally people discover that an existing chemical treats an illness, without the chemical having been discovered and patented by a pharmaceutical company. In this case, whoever spends tens of millions of dollars proving it works to the FDA may not get a monopoly on the drug and the right to sell it for ridiculous prices. So nobody spends tens of millions of dollars proving it works to the FDA, and so it risks never getting approved.

The usual solution is for some pharma company to make some tiny irrelevant change to the existing chemical, and patent this new chemical as an “exciting discovery” they just made. Everyone goes along with the ruse, the company spends tens of millions of dollars pushing it through FDA trials, it gets approved, and they charge ridiculous prices for ten years. I wouldn’t quite call this “the system works”, but again, at least we get new medicines.

Twenty years ago, people noticed that ketamine treated depression. Alas, ketamine already existed – it’s an anaesthetic and a popular recreational drug – so pharma companies couldn’t patent it and fund FDA trials, so it couldn’t get approved by the FDA for depression. A few renegade doctors started setting up ketamine clinics, where they used the existing approval of ketamine for anaesthesia as an excuse to give it to depressed people. But because this indication was not FDA-approved, insurance companies didn’t have to cover it. This created a really embarrassing situation for the medical system: everyone secretly knows ketamine is one of the most effective antidepressants, but officially it’s not an antidepressant at all, and mainstream providers won’t give it to you.

The pharmaceutical industry has lobbyists in Heaven. Does this surprise you? Of course they do. A Power bribed here, a Principality flattered there, and eventually their petitions reach the ears of God Himself. This is the only possible explanation for stereochemistry, a quirk of nature where many organic chemicals come in “left-handed” and “right-handed” versions. The details don’t matter, beyond that if you have a chemical that you can’t patent, you can take the left-handed (or right-handed) version, and legally pretend that now it is a different chemical which you can patent. And so we got “esketamine”.

Am I saying that esketamine is just a sinister ploy by pharma to patent and make money off ketamine? Yup. In fact “esketamine” is just a cutesy way of writing the chemical name s-ketamine, which literally stands for “sinister ketamine” (sinister is the Latin word for “left-handed”; the modern use derives from the old superstition that left-handers were evil). The sinister ploy to patent sinister ketamine worked, and the latest news says it will cost between $590 to $885 per dose.

(regular old ketamine still costs about $10 per dose, less if you buy it from a heavily-tattooed man on your local street corner)

I’ve said it before: I don’t blame the pharma companies for this. Big Government, in its infinite wisdom, has decided that drugs should have to undergo tens of millions of dollars worth of FDA trials before they get approved. No government agencies or altruistic billionaires have stepped up to fund these trials themselves, so they won’t happen unless some pharma company does it. And pharma companies aren’t going to do it unless they can make their money back. And it’s not like they’re overcharging; their return to investment on R&D may already be less than zero. This is a crappy system – but again, it’s one that occasionally gets us new medicines. So it’s hard to complain.

But in this case, there are two additional issues that make it even worse than the usual serving of crappiness.

First, esketamine might not work.

Johnson & Johnson, the pharma company sponsoring its FDA application, did four official efficacy studies. You can find the summary starting on page 17 of this document. Two of the trials were technically negative, although analysts have noticed nontechnical ways they look encouraging. Two of the trials were technically positive, but one of them was a withdrawal trial that was not really designed to prove efficacy.

The FDA usually demands two positive studies before they approve a drug, and doesn’t usually count withdrawal trials. This time around, in a minor deviation from their usual rules, they decided to count the positive withdrawal trial as one of the two required positives, and approve esketamine. I suspect this was a political move based on how embarrassing it was to have everyone know ketamine was a good antidepressant, but not have it officially FDA-approved.

But if ketamine is such a good antidepressant, how come it couldn’t pass the normal bar for approval? Like, people keep saying that ketamine is a real antidepressant, that works perfectly, and changes everything, unlike those bad old SSRIs which are basically just placebo. But esketamine’s results are at least as bad as any SSRI’s. If you look at Table 9 in the FDA report, ketamine did notably worse than most of the other antidepressants the FDA has approved recently – including vortioxetine, an SSRI-like medication.

One possibility is that ketamine was studied for treatment-resistant depression, so it was only given to the toughest cases. But Table 9 shows olanzapine + fluoxetine doing significantly better than esketamine even for treatment-resistant depression.

Another possibility is that clinical trials are just really tough on antidepressants for some reason. I’ve mentioned this before in the context of SSRIs. Patients love them. Doctors love them. Clinical trials say they barely have any effect. Well, now patients love ketamine. Doctors love ketamine. And now there’s a clinical trial showing barely any effect. This isn’t really a solution to esketamine’s misery, but at least it has company.

Another possibility is that everyone made a huge mistake in using left-handed ketamine, and it’s right-handed ketamine that holds the magic. Most previous research was done on a racemic mixture (an equal mix of left-handed and right-handed molecules), and at least one study suggests it was the right-handed ketamine that was driving the results. Pharma decided to pursue left-handed ketamine because it was known to have a stronger effect on NMDA receptors, but – surprise! – ketamine probably doesn’t work through NMDA after all. So there’s a chance that this is just the wrong kind of ketamine – though usually I expect big pharma to be smarter than that, and I would be surprised if this turned out to be it. I don’t know if anybody has a right-handed ketamine patent yet.

And another possibility is that it’s the wrong route of administration. Almost all previous studies on ketamine have examined it given IV. The FDA approved esketamine as a nasal spray – which is a lot more convenient for patients, but again, not a lot of studies showing it works. At least some studies seem to show that it doesn’t. Again, usually I expect big pharma not to screw up the delivery method, but who knows?

Second in our litany of disappointments, esketamine is going to be maximally inconvenient to get.

The big problem with regular ketamine, other than not being FDA-approved, was that you had to get it IV. That meant going to a ketamine clinic that had nurses and anesthesiologists for IV access, then sitting there for a couple of hours hallucinating while they infused it into you. This was a huge drawback compared to eg Prozac, where you can just bring home a pill bottle and take one pill per day in the comfort of your own bathroom. It’s also expensive – clinics, nurses, and anesthesiologists don’t come cheap.

The great appeal of a ketamine nasal spray was that it was going to prevent all that. Sure, it might not work. Sure, it would be overpriced. But at least it would be convenient!

The FDA, in its approval for esketamine, specified that it could only be delivered at specialty clinics by doctors who are specially trained in ketamine administration, that patients will have to sit at the clinic for at least two hours, and realistically there will have to be a bunch of nurses on site. My boss has already said our (nice, well-funded) clinic isn’t going to be able to jump through the necessary hoops; most other outpatient psychiatric clinics will probably say the same.

This removes most of the advantages of having it be intranasal, so why are they doing this? They give two reasons. First, they want to make sure no patient can ever bring ketamine home, because they might get addicted to it. Okay, I agree addiction is bad. But patients bring prescriptions of OxyContin and Xanax home every day. Come on, FDA. We already have a system for drugs you’re worried someone will get addicted to, it’s called the Controlled Substances Act. Ketamine is less addictive than lots of chemicals that are less stringently regulated than it is. This just seems stupid and mean-spirited.

The other reason the drugs have to be given in a specially monitored clinic is because ketamine can have side effects, including hallucinations and dissociative sensations. I agree these are bad, and I urge patients only to take hallucinogens/dissociatives in an appropriate setting, such as a rave. Like, yeah, ketamine can be seriously creepy, but now patients are going to have to drive to some overpriced ketamine clinic a couple of times a week and sit there for two hours per dose just because you think they’re too frail to handle a dissociative drug at home?

I wanted to finally be able to prescribe ketamine to my patients who needed it. Instead, I’m going to have to recommend they find a ketamine clinic near them (some of my patients live hours from civilization), drive to it several times a week (some of my patients don’t have cars) and pay through the nose, all so that some guy with a postgraduate degree in Watching People Dissociate can do crossword puzzles while they sit and feel kind of weird in a waiting room. And then those same patients will go home and use Ecstasy. Thanks a lot, FDA.

And the cherry on the crap sundae is that this sets a precedent. If the FDA approves psilocybin for depression (and it’s currently in Phase 2 trials, so watch this space!) you can bet you’re going to have to go to a special psilocybin clinic if you want to get it. Psychedelic medicine is potentially the future of psychiatry, and there’s every indication that it will be as inconvenient and red-tape-filled a future as possible. If you thought it was tough getting your Adderall prescription refilled every month, just wait.

So far, I am continuing to recommend that my patients who want ketamine seek intravenous racemic ketamine at an existing ketamine clinic, since this has a stronger evidence base. Once insurance starts covering esketamine, I may change my mind if money becomes an issue. But I’m annoyed that it’s come to this.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

164 Responses to Ketamine: Now By Prescription

  1. albatross11 says:

    Scott:

    If you could set the FDA/prescription drug regs up however you wanted, how would you change it?

    Are Ketamine/Psylobilin approved outside the US anywhere?

    • Scott Alexander says:

      That’s a tough question. Any answer I can give is very preliminary, because I haven’t researched this in enough detail to know whether a seemingly-good idea might make things much worse. But it would look something like:

      1. Auto-approve any medication that has been approved in another developed country (or underdeveloped country if they can prove they follow good regulatory processes).

      2. Legal to import generics from any other developed country (or underdeveloped country if they can prove they follow good manufacturing processes).

      3. Make it much easier for generic manufacturers to compete in making generic medications.

      4. Gather a board of distinguished scientists who can analyze existing research, and sometimes say that the existing research is good enough to support approval of an existing chemical (eg ketamine, melatonin, etc). Anything approved in this way is off-patent and any generic manufacturer can make it with minimal fuss.

      5. Companies may no longer run their own trials. They may pay the FDA to run their trials for them, with their advice and involvement, and report back. All trials to be done to modern standards, involve adversarial collaboration, etc.

      6. Sometimes the FDA will give its own trial-running arm some money to run trials on a promising new medication that for some reason doesn’t gather commercial interest. Anything funded in this way will be patent-free and publicly available to any generics manufacturer who wants it.

      7. Weird nootropics people no longer get cease-and-desist letters. Instead they get “how nice of you to agree to act as lab rats, the US government thanks you for your service, let us know if you find anything interesting” letters.

      8. Have a very long talk with the DEA about whether they really need to ban all psychedelics. If they don’t see things my way, grant FDA approval to curare, delivery method “blowdart”, sole on-label indication “poisoning DEA officials”, until they change their mind.

      Ketamine is approved almost everywhere for anaesthesia. I think the US is first to approve it for depression, though I’m not sure other countries’ systems are all necessarily set up in a way where this distinction makes sense.

      • Ghillie Dhu says:

        8. Have a very long talk with the DEA about whether they really need to ban all psychedelics. If they don’t see things my way, grant FDA approval to curare, delivery method “blowdart”, sole on-label indication “poisoning DEA officials”, until they change their mind.

        Overall, there’s a subset of Federal agencies that are just a dog’s breakfast; reorganizing them along more coherent boundaries would fix a lot of issues.

        -Bureau of Alcohol, Tobacco, Firearms, & Explosives: Weapons & drugs (legal, recreational)
        -Drug Enforcement Agency: Drugs (illegal)
        -Food & Drug Administration: Drugs (legal, medicinal) & food
        -US Department of Agriculture: Food

        A reorg’ing of these four into three whose areas of responsibility better cleave reality at the joints would be a big improvement in general:

        -Bureau of Firearms & Explosives: Weapons [plus, this would be the only one of the three that actually needs armed agents]
        -Drug Administration: all drugs
        -Food Safety Agency: just food

        • Aftagley says:

          -Bureau of Firearms & Explosives: Weapons [plus, this would be the only one of the three that actually needs armed agents]
          -Drug Administration: all drugs
          -Food Safety Agency: just food

          This is a terrible idea.

          You’d have the organization that oversees drug trials be the same people who are responsible for preventing and interdicting boat/truckloads of cocaine being smuggled over the border? You’d send out DEA agents against the cartels unarmed on the hope that since the cartels aren’t smuggling weapons, they wouldn’t happen to be armed themselves?

          I get that we’re about increasing efficiency and rationality here, but that doesn’t always mean that we need to assume that the existing structures only exist because of inane bureaucratic nonsense.

          • Ghillie Dhu says:

            …preventing and interdicting boat/truckloads of cocaine being smuggled over the border

            There are already armed agencies tasked with protecting the water & land borders: the Coast Guard and Customs & Border Patrol, respectively. No point having those functions collocated with determination of which drugs belong on which schedule of the CSA.

          • Aftagley says:

            Right, but none of those organizations are the experts on the illegal movement of drugs. Speaking a member of one of the agencies you listed, we get somewhere above 80% of our drug targeting information from DEA Agents. Without them and their intel, we’d be forced to go back to the bad old days of breaking down every car or truck that tries to drive north from Mexico and sending our Cutters out to to go cut holes in the water for two months.

            Any credible implementation of a counterdrug policy is going to require some collection of (armed) serious-minded investigative/intelligence inclined individuals who focus specifically on drugs. Currently, those people reside at the DEA and the DEA is built around supporting the specific needs of those people.

            Edit: That being said, sure – take away the DEA’s ability to control the CSA. That’s a political, not investigatory power. Just don’t throw the baby out with the bathwater.

          • Ghillie Dhu says:

            Full disclosure: I’m approaching the issue from the perspective that there shouldn’t even be a counterdrug policy.

            I see breaking up the agency whose very existence is conditional on its continuance as a means to eliminate the institutional incentives to maintain prohibition.

          • Aftagley says:

            ok, we’re probably not going to agree then.

      • Bugmaster says:

        From an attacker’s point of view, this policy is pretty generous.

        Points 1 and especially 2 allow big pharma companies to subvert the approval process in other, less developed countries; this is going to be much easier than doing so in our own country, potentially as easy as driving a compact car full of money to a legislator’s house.

        Point 3 is, admittedly, somewhat worrying (to an attacker), but I’m not entirely sure what it means, so maybe it’s not too bad.

        Point 4 is similar to points 1 and 2: what counts as a “distinguished scientist” ? The answer is, “here’s a mid-size sedan full of money, now go and publish some puff pieces for us on Dr. Notashill”.

        Point 5 is, admittedly, a large hurdle. However, tasking one agency with running all the trials is going to bog them down essentially forever; the only possible outcomes is to either just get the agency to rubber-stamp almost anything (kind of like the Patent Office), or deregulation. Obviously the first outcome is preferable — assuming the attacker cares at all, since they can now import all their trials from Belarus, anyway.

        I’m not sure how point 6 would work, but if a medication doesn’t gather commercial interest, then it’s probably pointless anyway.

        Point 7 is somewhat irrelevant, since any data you’d gather that way would be hopelessly confounded, so it’s kind of tangential to the entire process.

        Point 8 is emotionally satisfying but unlikely to work. Those DEA agents have spent the last few years developing an immunity to curare.

        • Matthias says:

          All those attacks rely on money. We know that you can already use money to work the current system.

          • Bugmaster says:

            Yes, of course; but fabricated trials (as well as fabricated doctors) are bound to be much cheaper in Belarus or Whateverstan than in America.

        • John Schilling says:

          #1 seems like it would have led to the USA approving thalidomide, if #1 had been the rule at the time.

          And Scott’s proposed Rule #8 seems like it could be profitably extended to cover anyone who uses “But Thalidomide!” when discussing drug development policy. The thalidomide tragedy was sixty years ago. How about pointing to some of the other tragically dangerous drugs those crazy European regulators have approved in the last half century? Or else admit that thalidomide was so freakishly rare as to be one of the three-sigma hard cases that make bad law, or that other first-world countries have managed to tailor their regulations to provide adequate safety without being as absurdly expensive or obnoxiously obstructionist as our own.

          • myers2357 says:

            +1, Thalidomide’s Legacy has doomed far more people than Thalidomide ever did.

          • Squirrel of Doom says:

            +2 on Thalidomide!

            One very important thing to understand about security regulations like these is that they’re tradeoffs between the people you kill by (A) allowing bad drugs erroneously and (B) banning effective drugs erroneously.

            If you never do any type A errors, it means you’re way doing way too many type B errors! Those errors are unfortunately silent.

          • sclmlw says:

            Thalidomide was also sold as a racemic mixture. It turns out one of the enantiomers is great at preventing morning sickness, while the other is great at preventing angiogenesis (new blood vessel growth).

            We now know how to synthesize and separate the stereoisomers of thalidomide. Strangely, nobody has proposed bringing back thalidomide for treating morning sickness. I wonder why.

            Some drugs do act the same regardless of stereochemistry, but not all do. If esketamine doesn’t work as well as expected given ketamine’s reputation, the strongest probability is the structural difference. Think of it this way: imagine trying to shake hands with someone who had no right arm. It might sort of fit. You can kind of make it work. But it’s not really the same interaction as shaking hands with someone who has no left arm.

          • The Nybbler says:

            Strangely, nobody has proposed bringing back thalidomide for treating morning sickness. I wonder why.

            Just as well. It turns out enantiomerically pure thalidomide racemizes in vivo.

        • Cliff says:

          just get the agency to rubber-stamp almost anything (kind of like the Patent Office)

          LOL no- nothing like the patent office. Where would you even come up with the idea that the patent office rubber-stamps everything?

          Your other points seem to ignore that a) the other countries have to prove they follow good regulatory processes and b) the corrupt foreign practices act works

      • Furslid says:

        I’d also think we need to reform prescription for other conditions. It seems that the problem here was ketamine for depression needed the same testing procedures as *totally untested compound* for depression. There should be a much more streamlined procedure for approval for new uses for existing drugs.

        I’m not sure how it should be streamlined, and it should probably take into account the potential for abuse.

        • Aapje says:

          Obvious streamlining is to require fewer or no trials that show safety.

          • Garrett says:

            It would be very nice if a scientist somewhere could put together a non-patent approval packet based solely on eg. Cochrane studies or whatever. Comb through all of the existing post-approval research material, synthesize it and go “hey – it works sometimes and isn’t significantly more dangerous than everything else we do”. Put all of that together in a Really Big Binder, turn it in to the FDA and have that be considered for approval.

          • Murphy says:

            Goodhart’s law applies unfortunately.

            Currently Cochrane reviews are good because they’re a metric.

            But make them a billion dollar target and manipulation of them becomes a target.

            And somebody still has to actually do the big safety trials.

      • Lucid Horizon says:

        #1 seems like it would have led to the USA approving thalidomide, if #1 had been the rule at the time. Perhaps rather than just approving of the cut of another country’s regulatory jib, this hypothetical FDA should have to review individual studies done by their foreign counterparts for questionable methodology and unconsidered angles?

        • Furslid says:

          Yes. It would have lead to the US approving Thalidomide. Thalidomide in Europe was a tragedy. If it was approved in the US there would have been more suffering. However, this is not the only way that a regulatory regime can fail.

          When someone doesn’t get needed medication because it isn’t approved, they are suffering because of the regulatory system as much as a person who gets a dangerous drug. When extra billions are spent on getting drugs approved, they aren’t spent on something else. When drug costs go through the roof, people can’t afford them or have to give up other good things. These also increase societal suffering.

          Any sane decision making procedure will have type one and type two errors. Sometimes it will reject safe drugs. Sometimes it will approve dangerous drugs. If the only acceptable policy is one that never approves dangerous drugs, then a lot of safe and effective drugs will have to be rejected.

          These rules would have allowed a tragedy. All feasible rules have tragedies. We need to do actual cost-benefit analysis rather than point to the flashiest example.

          • Murphy says:

            There’s other issues though as well.

            The approval process isn’t just some symbolic ritual. It generates a dataset from all the safety and efficacy trials it insists upon.

            How do you value the worlds doctors having access to high-confidence, high-quality set of data on safety, efficacy and side effects, drug interactions etc of the drugs they prescribe?

            If that data just didn’t exist (and make no mistake, if nobody is forced to generate it, it mostly simply doesn’t exist as we’ve learned from various cases where drugs were fast-tracked with a pinkie-swear that they’d do the trials later… then didn’t) you’re stuck with doctors gut feelings and sporadic reports… which tend to miss anything that isn’t killing an extreme number of their patients by rare/flashy means.

            The data-fairy doesn’t turn up and fix everything.

            So, now the worlds doctors need to make all their choices without access to a high quality data. Forever. If they’re lucky they get the occasional small trial run by a university based on some academics hunch.

            Remember to count the corpses from that as well. Worldwide.

            Currently if someone really really believes they’re free to travel to some country with fuck-all regulation and take any random concoction that the local doctors insist will cure their cancer. Those clinics exist…

            But I have yet to hear of one of those thing’s that wasn’t just a scam designed to prey on the vulnerable when they’re at their weakest and most desperate.

            Typically it doesn’t work and generates effectively no useful data for future use either.

            Plus throw in some extra loss of trust in the medical system.

          • Furslid says:

            I didn’t say scrap the approval process. I implied that our process is skewed towards rejecting valuable drugs and pointed out that one flashy example doesn’t win the argument.

            As for generating data. I’m not sure that the approval studies are the best or only way to get it. Especially repeating the studies done in other countries.

        • ana53294 says:

          You could also add the provision “and has been widely used and not recalled for at least 5 years”, which would exclude thalidomide and other mistakes.

          • Garrett says:

            This would be worthwhile. A contrast to thalidomide: metformin. I was watching ER reruns and heard the characters talking about the life-changing drug but it was really expensive ($300/month). It turns out it was approved in the US in 1995. Now it’s so cheap that my pharmacy-in-a-grocery store supplies it for free.

            It turns out it was first introduced as a medication in France in 1957. So any comparison to thalidomide should also include a comparison to metformin.

          • Douglas Knight says:

            Metformin isn’t such a great example. It wasn’t used continuously in France. It was the clinical trials that the FDA demanded that caused Europe to revisit the drug! (Also, maybe contemporary UK trials.) So the initial high price was coupled to paying for the trials, ie, the system working correctly. I don’t know where I heard that the story was complicated, but I’m getting most of the details from here.
            (Maybe UK trials in the 60s should have convinced people to use metformin, but they didn’t, not outside the UK and maybe not inside, either. Spreading information is difficult. Marketing provides a valuable service.)

      • HeelBearCub says:

        1. Auto-approve any medication that has been approved in another developed country (or underdeveloped country if they can prove they follow good regulatory processes).

        5. Companies may no longer run their own trials. They may pay the FDA to run their trials for them, with their advice and involvement, and report back. All trials to be done to modern standards, involve adversarial collaboration, etc.

        Now, maybe Scott thinks that point 5 applies to point 1, and that good regulatory process is necessarily government run trials. But I doubt it, as I am not aware of any country that does this.

        This is some sort of logical fallacy, I would have to think. It seems there is an underlying assumption that if something is broken, each part must be broken. It seems related to the basic “good/evil” kind of thinking that results in fat being first evil (and carbohydrates good) and then carbohydrates evil (and therefore fats good).

        I also find it interesting that the same person who leads a charge against the replicability crisis also seems to think that drugs should be able to be shown to be safe and efficacious cheaply.

      • Aapje says:

        @Scott Alexander

        What I would add:

        Require much better post-approval evaluations, for example, in the form of post-treatment and/or yearly surveys. I think that it is rather absurd that so much effort is spent before approval to check safety and efficacy, but so little is done afterwards, even though then you may have immense numbers of users that can report (positive or negative) side effects, problems with the administration method, problems with sticking to the treatment, etc.

        Obviously the results of these surveys wouldn’t be well-controlled or such, but it seems to me that the large numbers would allow drawing useful conclusions and/or suggest useful avenues for further research.

        We are increasingly being asked to judge services that we use. For example, my garage wants to know how I liked the service, Uber wants to know how you liked the ride and Amazon wants to know how you liked the product. Yet I’ve never been asked to rate & review medicines. Yet I’d much rather rate medicines than pretty much everything else.

        • SamChevre says:

          I’d second this and add on:
          Anything that’s newly approved is limited in some way–maybe only certain specialists can prescribe it. Limits are steadily relaxed over ~5 years as it is used and side effects are reported and are acceptable. This avoids the “everyone uses new drug with unrealized side effects” like Vioxx.

      • Douglas Knight says:

        5. Companies may no longer run their own trials. They may pay the FDA to run their trials for them, with their advice and involvement, and report back. All trials to be done to modern standards, involve adversarial collaboration, etc.

        How much does that differ from the current situation? Don’t you know that the drug company and the FDA design the pre-registered analysis together? And that the FDA, not the drug company, performs the actual analysis? FWIW, in practice it’s always a CRO, not the drug company performing the trial. Are you worried about some kind of fraud in the actual administration of the trial? Much of the point of double-blinding is to make fraud difficult.

        • MilfordTrunion says:

          I was gonna say, this is the kind of thing that people say when they don’t actually know how clinical trials work. The FDA demands a full review of everything you want to do before you’re allowed to write the study protocol, let alone put actual drugs into actual people. And they will make suggestions about what you should do, and those suggestions will be put into the plan, otherwise you aren’t gonna do any clinical-trialing.

          “the FDA, not the drug company, performs the actual analysis?”

          not entirely the FDA — they more often read over the analysis done by the submitter and decide whether or not they approve — but you’re correct that it’s not Joe Pfizer who does the analysis, it’s a CRO, and in any case if the FDA thinks it’s bunk they’ll stamp a big F-for-effort on your NDA and roundfile it.

      • Rudbek says:

        How about…

        9. Issue ‘I’m a real grown up card’ to any adult willing to accept (financial and legal) responsibility for their own choices. Cardholders would be allowed to do crazy things like chose to take an unapproved medication recommended by their doctor or ride a motorcycle without a helmet.

        • ec429 says:

          A million times this.

          But of course if they allowed it, we’d be in the Eldraeverse in no time and then what would the slavers do?

        • Andrew Cady says:

          Not really possible since other people who didn’t sign the consent form (e.g. dependents, including minors) can still sue a snakeoil salesman for their wrongful death of their relative.

    • Matthias says:

      How about reverting to the older FDA regime were you just had to prove safety and not efficacy?

      Leave efficacy to private actors to decide up on. It’s what we already do to choose between the different antidepressants or pain killers etc, or when there’s an off-label use.

  2. John Schilling says:

    Ketamine is a drug people take for fun. Fun is an unacceptable reason for taking drugs (except for alcohol, and marijuana in blue states). Therefore, if someone wants society’s official approval to take a fun drug like ketamine, we need to make sure there are official proctors watching closely to make sure they don’t have any fun. Thus is virtue ensued.

    I don’t see any reason to believe it’s anything more or less than this. And, yes, it will be the same for psilocybin et al.

    • Scott Alexander says:

      To be fair, there’s been such a backlash to opioid addiction that everyone is being super super cautious now. I don’t think we need to posit any further motives beyond that.

      • John Schilling says:

        The preferred narrative surrounding opioids is that people don’t take them for fun; they take them for pain only to find that Evil Big Pharma and the Pill-Pushing Doctors have conspired to make the stuff maximally profitable addictive, whereupon they are condemned to miserable addiction unto death. That’s a different story than the one about party drugs like ketamine. But I agree that it does provide moral support for the sort of puritanical-ascetic attitude re drug use and drug regulation generally.

        • vV_Vv says:

          Are there so many people who suffer from chronic pain to the extent that they need to take opioids, or is it that the pill-pushing doctors are prescribing oxycodone to people who would just need ibuprofen?

          • John Schilling says:

            Quite possibly neither. Lots of people do in fact take opioids for fun, or at least for a cheap high to obscure the fact that their life is otherwise devoid of fun, meaning, hedons, and/or utils. But admitting that requires accepting that large segments of our society are broken in ways that can’t be blamed on a convenient scapegoat, and since most of these people will at least temporarily pass through a condition where they are talking about their pain to a doctor who will prescribe them opioids, making the doctors the scapegoats is a more convenient narrative.

          • The Nybbler says:

            The current narrative in the not-so-great-state of NJ is that people take them for _acute_ pain and then get addicted, and they should take ibuprofen for broken bones or post-operative pain instead. It’s true you don’t need opioids for acute pain, you COULD just bite down on a leather strap (except if it was a dental operation). But IMO it’s really sadism at that point.

        • Garrett says:

          The part of the story nobody wants to talk about is what percentage of opiate “overdoses” are actually suicide attempts. Reporting a few months ago in talked about up to 40% being suicide attempts.

        • Lambert says:

          I hear it’s because opiates are cheap and physio is expensive.
          And insurance companies don’t have to deal with the externalities that the government does when providing healthcare.

    • Deiseach says:

      John Schilling, we just had a local murder case where a junkie beat a ninety year old man to death in a botched robbery looking for money to pay for his next fix (and then tried to pass it off as self-defence), so I have absolutely no fucking problems at all with mean ol’ no-fun proctors making sure there’s no fun to be had when getting a medical treatment based on a drug that is already being abused as a fun party substance by said shit heads.

      Sorry not sorry to harsh your mellow, man.

      • baconbits9 says:

        And this event is somehow not the cause of the current system but the cause of some hypothetical other system?

      • Theodoric says:

        How about we just enforce laws against beating people instead of subjecting people who use non-government approved recreational substances to being handcuffed, strip searched, confined in an environment that really is a rape culture, and given a record that makes it hard to find work at McDonalds?

        • Aftagley says:

          Because then you’re ignoring the causative factor drug addiction has on criminal behavior? I mean, reread her statement needing the money was a causative factor in the robbery that led to the murder. This is incredibly common. Yeah, not all drug use results in violent and illegal behavior, but a startlingly high percentage of violent and illegal behavior has a nexus to addiction.

          • Winter Shaker says:

            Deiseach didn’t actually specify what the junkie’s next fix was a fix of. If it was ketamine, then okay, this is relevant (though not sufficient proof that the policy of prohibition is justified). If it was some other drug, presumably most likely some kind of opiate, then before we can conclude that ketamine prohibition is the best policy, we really need good evidence that
            a) the policy of prohibition against opiates really is the best way to mitigate this sort of violence, despite its failure in this case, and
            b) the effects of ketamine are sufficiently similar to those of opiates, at least in the relevant sense of rewiring people’s brains so that they would become liable to resort to violence to acquire more, to justify lumping them together.

            I mean, the top level comment is also mentioning psilocybin, of all things, as likely to be caught up in the same regulatory scheme. Sadly, he’s probably right – but psilocybin and opiates are so wildly different in their effects and risk profile that any policy that treats them as basically the same for the purposes of deciding what to do to people who use them is a policy that is throwing away so much useful information that ‘proctors seeking to prevent fun’ starts to seem about as likely an explanation as ‘rational technocrats seeking to minimise harm’.

          • casuarius says:

            I don’t think it’s reasonable to even accept the premise that we should use criminal law to target indirect causal factors in crimes on the individual level. What standard will be used to demonstrate that there’s enough of a causal link to justify criminalizing behavior that isn’t causing harm by itself? How early do we want to short-circuit the possibility that a real crime with a victim will be committed, and how much of a false positive rate are we willing to accept in this kind of law enforcement?

            To reductio ad absurdum this, it’s at least a well-known cliche (for the sake of this argument, it’s not actually important whether it’s true) that many murders are committed by spouses or romantic partners of the victim. In other words, romantic involvement is a clear causal factor in murders. Should we criminalize romantic relationships?

            On a larger scale, is it reasonable to just start jailing people below a certain socio-economic status threshold, because their likelihood of committing petty crimes is high enough?

            A criminal law is a very dangerous weapon. On the individual level, any behavior you target with a criminal law should be understood as a broadening of the swath of people whose lives you are willing to ruin for the sake of everyone else, especially in the context of as punitive a system as exists in the US. On the level of whole societies, any law that defines some behavior as criminal should be seen as an expansion of the level of control and surveillance the state is allowed to subject its citizens to. The idea that we should build this kind of weapon based on the idea that there’s some less-than-total-equivalence causal link between an outcome we definitely want to prevent and a behavior that is none of our business when it doesn’t do that is a horrible tradeoff.

      • casuarius says:

        I think the irony of this exchange is that both arguments hinge on the notion that there are magically acausal non-rational bad actors that somehow taint the policy in some way or another, and thus the policy must be the opposite of what those bad actors want. You’re both expressing social disgust rather than arguments for why the policy should be one way or another.

        It seems unlikely that the state’s terminal goal in this space is to promote a weird omni-puritanical ideology about drug use. This may be how politicians convince certain segments of their base to vote for or tolerate such policies, but I’d argue the state as an entity has too many selection pressures on it to be motivated that way. If we model the state as an agent unto itself, an explanation I’ve heard that’s more convincing is the notion that universal criminality is a powerful means of control, which motivates making large swaths of controversial but fun behavior illegal. We would expect governments that successfully exert more control over their populations to be able to perpetuate themselves better than ones that exert less.

        In a similar vein, the notion that the bad behavior of some drug addicts justifies banning the drugs completely mischaracterizes the effects of the actual policy. No matter how much people want laws to be codifications of ethics or their personal aesthetics, that is not what they are in practice. A criminal law ties a behavior to a consequence, which for the most part involves armed agents of the state coming after you and locking you up. We also explicitly disenfranchise criminals, and cut off their access to various advantages of society in an explicit way, on top of the emergent social effect of being shunned and sometimes targeted by the large segment of the population who short-circuit outgroup the category of “criminals”. By creating this system of incentives, we have created the desperation that causes drug addicts to avoid detection and therefore help or treatment, seek out illegal means of feeding their addiction, and fear reporting other abuse or coercion they encounter in black markets, which allows black markets to become more abusive and coercive over time, including but not limited to their tendency to use whatever shady means they can to try to drive more addiction in their pool of potential and existing customers.

        • Christophe Biocca says:

          If we model the state as an agent unto itself

          That’d be it’s own kind of mistake. Even small groups of humans have substantial incentive misalignment, and governments much more so.

          It’s not a “terminal goal” of the state to promote anti-drug puritanism (because it has no goals of its own), but it is an explicit goal of many elected representatives for the state to be seen acting according to this view by voters (their careers depend on it).

          A view that treats the state as a monolithic agent will have a hard time explaining the mostly scattered and incoherent policy states enact in practice.

          • casuarius says:

            When decisions of such a scattered and incoherent institution are codified and enforced, it is meaningful to call them the results of that institution as a whole. Internal or temporal goal misalignment is present even in agents we’re more used to reasoning about as such, and is not an argument against modeling such an entity’s actions in aggregate as an agent, and trying to reason about what system of incentives the institution is operating under and how it’s constructed (e.g. the rules by which the scattered decisions made by several sub-agents are aggregated and codified into law).

            The current version of the US Federal Government maintains an entire agency that bans various drugs by fiat, supposedly overriding more local legal systems, even if in practice some jurisdictions are more lenient on sentencing or tend to ignore such crimes at the local enforcement level. The same entity has codified policies which allow them to cause harm to or exert control over certain people it has deemed to be engaging in criminal activity. There are lots of reasons the entity that gets to enact exceptions to normal limitations on its power over people would seek to expand the number of exceptions, and the gestalt outcome described is no more confusing to this model than modeling a human with any kind of akrasia (that is, a human) as a monolithic agent for the purpose of explaining some decision it made.

            Regardless, I don’t think this model fully explains the state of drug policy in the US, just that it’s more plausible as a primary explanation than puritanical anti-fun cultural mores having disproportionate sway in policy-making

          • Christophe Biocca says:

            than puritanical anti-fun cultural mores having disproportionate sway in policy-making

            The “puritanical anti-fun” is probably overly specific, but drug legalization is unpopular:
            https://globalnews.ca/news/4561760/canada-legalize-all-drugs-marijuana/

            (Using a canadian poll because all US polls I could find are about Opioids/Marijuana, but I think these percentages would be roughly similar in the US. Or lower, since MJ legalization in Canada has been utterly boring, while it’s still an ongoing matter of debate in the US).

            No need to invoke fancy state-expands-its-own-power motivations for the government continuing to prosecute a war on drugs that ~80% of the population supports (in some form or other). That’s higher support than (NOT “the First Amendment goes too far”) gets.

          • casuarius says:

            I sympathize with the lack of good data, but I have to say it’s a stretch to take a Canadian poll that asks whether all drugs should be legalized to mean that ~80% of the population of a different country is for that country’s current status quo of a federal-level prohibition of a large number of substances and treating this as a matter of criminal law.

            I realize that you qualified the statement with “in some form or another”, but the section of opinion-space between “current US federal drug policy is correct” and “all drugs should be legalized” might contain the entire Overton window on this topic, so it doesn’t strike me as particularly justified by the evidence, even if we make a very charitable assumption that the results of this poll perfectly predict the rate at which Americans would answer the same question.

            In fact, the linked article itself even mentions that many of the respondents’ stated reason for marking “no” was not knowing what legalization would look like in practice.

          • Christophe Biocca says:

            but the section of opinion-space between “current US federal drug policy is correct” and “all drugs should be legalized” might contain the entire Overton window on this topic

            I think you’re drawing the Overton window quite weirdly there. If we consider 20% support to be “inside”, then “death penalty for persons convicted of selling drugs that cause a lethal overdose” has 21% support. Now I guess we can consider that “current de jure US federal drug policy is correct” (since a form of death penalty has been on the books since 1994), but that’s not quite the same.

            But more importantly the point I’m trying to make is that, on this topic, de facto federal policy (ie. most drugs are illegal, marijuana quasi-federalism, discourage doctors from prescribing painkillers, mostly prosecute distributors rather than end users, to the extent you can tell the difference) actually matches public opinion, and there’s no need for state-as-own-agent epicycles to explain why there’s still a war on drugs, still a crackdown on legal painkillers, and the FDA is playing it super-safe with esketamine distribution.

          • casuarius says:

            I’m not sure how that conclusion can even be made with the data available. Polls on this subject are generally ill-equipped to even tackle whether public opinion supports how sentencing and enforcement is done, and polls I’m able to find about particular drugs usually show that most citizens don’t even know what schedule a particular drug falls under and what that legally means, and many think that the scheduling for various drugs should be relaxed, but it’s hard to say what that means if they also don’t understand the regulations. I’m not saying I know that you’re wrong here, just that I don’t think there’s enough reliable information available for you to even draw that conclusion.

            As far as “states tend to increase their own power” arguments, I understand that this is one of those models that promotes a lot conspiracy-theory-like extrapolation and seeing bogeymen everywhere, but drug enforcement in the US is one of the cases where the model fits the best, given that drug enforcement is the explicit explanation for a lot of expansion of the reach of and resources devoted to law enforcement in the last several decades.

        • Matthias says:

          The Americans disenfranchise some convicted criminals.

          But that’s not universal. Especially not as universal as the rest of your argument tries to sound like.

          (The reality is, or course, that individual votes don’t matter too much.)

          • casuarius says:

            The American version of drug enforcement policy, so far as I can tell, is the matter being discussed. Its influence on the rest of the developed world is probably non-zero, but while the abstract claim that the real consequences of criminalizing a behavior can have counterproductive effects that look a lot like “criminals doing more bad things” certainly generalizes to some other cases, the specifics of the argument made here have to do with the consequences of drug criminalization on the federal level in the US as pertains to crimes like the one described. This is of course on top of the extra hoops (and resulting costs) imposed on medical research and treatment this blog post mentions.

            The disenfranchisement is probably not directly pushing the numbers on changing these policies, and was merely intended as an example (among others in a comma-separated non-exhaustive list) of the way in which criminalization of the flavor employed by US law enforcement tends to cut people off from the benefits of their participation in society, which in the case of recreational drug users is causing many problems without seeming to solve any definitively.

            While there are many criminal statutes where it could be argued that the benefits of adequately deterring and providing mechanisms to prevent certain harms outweigh this detrimental cluster of effects, there are many cases where the argument for the crime doing harm to anyone but the perpetrator is much clearer, on which we still probably err more on the side of punitive than continues to be useful.

      • @Deiseach:

        Do you have any stories about similar attacks by alcoholics to pay for their next fix?

        If not, isn’t the obvious conclusion that the incident you describe is a result of laws against recreational drugs pushing up their price?

        • Michael Watts says:

          I think this question is confused by the typical manner of media coverage. A junkie who attacks someone to get money for drugs is a “junkie”, because drugs are bad and therefore the bad things that drug users do are the fault of drugs. An alcoholic who attacks someone to get money for alcohol is a “mugger”, because alcohol is fine and muggings are therefore not alcohol’s fault.

          Stated another way, knowing that someone is a junkie and that they mugged someone doesn’t prove that the only use they were going to put that money to was drugs, but coverage will generally treat it that way. And symmetrically, knowing that someone committed a mugging while not obviously being a junkie doesn’t prove that they weren’t going to blow all their ill-gotten gains on booze, but coverage will (more reasonably) still treat it that way.

          Realistically, I expect alcohol-motivated muggings to occur at broadly comparable rates to illegal-drug-motivated muggings.

          • albatross11 says:

            This sounds plausible. On the other hand, the number of people getting killed over who gets to sell bathtub gin on a given street corner is *much* lower than the number killed over who gets to sell meth/crack on a given street corner.

          • Winter Shaker says:

            the number of people getting killed over who gets to sell bathtub gin on a given street corner is *much* lower than the number killed over who gets to sell meth/crack on a given street corner.

            But that too is an artifact of prohibition. The availability of legal alcohol shrinks the market for bathtub gin to the point where it’s not worth fighting over. If legal amphetamine/cocaine-based stimulants were available*, the market for street corner meth/crack would also be much smaller.

            *with all the usual caveats about not taxing them at a rate that start to make the black market stuff look attractive, etc

      • John Schilling says:

        Sorry not sorry to harsh your mellow, man.

        Well, if the villain in your little sob story had a prescription for some safe, free, and legal drugs that he could use when he felt like, then he probably wouldn’t have had to beat anyone up to get his next fix. Even if he were nothing but a fun-loving pill-seeking junkie who lied to get that prescription, the particular harm you are complaining about would have been averted by the particular policy you are objecting to.

        And don’t I recall you having repeatedly complained about doctors in Ireland being unwilling to prescribe antidepressants for you to use because you aren’t so suicidally dysfunctional to need mandatory treatment in a clinical setting and therefore don’t really need drugs?

        Pick one, please. Either people can be trusted with a bottle (nasal spray, whatever) of psychoactive drugs to take home, or they can’t. Either depressed people can be trusted to live their lives without being required to show up at the clinic every week, or they can’t. And yes, if you trust them, some of them will in fact hurt themselves or other people, and if you want to minimize the number of people who get hurt one way to do that is to make anyone with any mental health issues at all show up for a clinical evaluation every week, but pick one and live with it.

    • Godfree Roberts says:

      Good point. Ketamine, like LSD and–to a lesser extent or less interestingly–marihuana, allows people to explore dimensions of their own consciousness which are usually suppressed or overlooked. Many find that fun, though many do not.

      Our governments seem uniformly averse to such explorations–an aversion I find fascinating. Even sinister.

      • Matthias says:

        The governments seems to be happy to let you explore runner’s high or the effects of alcohol, nicotine or tobacco or prolonged fasting, though..

        • Aapje says:

          Plenty of people wanted to ban alcohol and the US actually did. It was legalized again not merely due to the immense crime it caused, but also the hypocrisy of those who punished others for violate the law, but who would often drink alcohol themselves.

          Tobacco is suppressed extremely strongly in most of the West, with high taxes, laws against using it in the workplace, bans on advertising, etc. It’s also not hallucinogenic, so it’s not apropos to this discussion in the way you used it, but merely as a counterexample to Godfree’s claim.

          In fact, is alcohol hallucinogenic? It seems to numb and suppress anxiety & such, but doesn’t really cause the kind of consciousness-expanding experience of LSD or marijuana.

          • Creutzer says:

            Alcohol isn’t in any way hallucinogenic, it allows you to explore and entirely different, and altogether less healthy corner of mindspace.

      • Aapje says:

        @Godfree Roberts

        Even sinister.

        What do left-handed people have to do with this???

      • Doctor Mist says:

        allows people to explore dimensions of their own consciousness which are usually suppressed or overlooked

        I am 100% on board with allowing people to do this, but I’m baffled by the implicit assumption that experimenting with mind-altering drugs is somehow praiseworthy. You can fiddle with your consciousness by hitting yourself over the head, too. I don’t think this is what Socrates had in mind when he said, “Know thyself”.

        • Creutzer says:

          I find two intuitive reasons why yes, I am inclined to call it praiseworthy:

          1. It’s not easy and often not pleasant, but immensely interesting. It takes a certain kind of mental toughness to fuck with your mind at the level of psychedelics.

          2. The corners of mindspace people discover with psychedelics (or MDMA) are, unlike those discovered by hitting oneself over the head, often of a positive nature. There is less suffering and anxiety in them and more goodwill.

  3. Paul Brinkley says:

    The great appeal of a ketamine nasal spray was that it was going to prevent all that.

    Instead, I’m going to have to recommend they find a ketamine clinic near them […], drive to it several times a week […] and pay through the nose

    I just had a hallucination that you chose this phrasing on purpose.

  4. Nornagest says:

    ketamine can have side effects, including hallucinations and dissociative sensations. I agree these are bad, and I urge patients only to take hallucinogens/dissociatives in an appropriate setting, such as a rave.

    I can’t help but read this in Tom Lehrer’s voice.

  5. rtypeinhell says:

    Seems like this level of regulation will ultimately restrict it to the most treatment-resistant cases anyway. If my doctor recommended me a triweekly two-hour prison stay, I’d say let’s start first with every single other treatment available, and if those don’t work I’ll self-medicate. And based on second-hand experience with ECT, I wonder how many patients will really be able to drive themselves – so add in the cost/challenge of finding a ride to whatever distant clinic, too. I can’t imagine what it’d be like for someone who has a family and kids.

    • martinw says:

      Or for someone who has a medical condition which makes it so that they have very little mental energy to spare for performing unpleasant chores. Like depression.

    • jameverywhere says:

      It seems to me that someone with a family and kids would be more likely to have the social support network necessary to actually get to the clinic and receive the treatment. Perhaps their spouse could drive them, or one of their teenage kids, or a friend or extended family member.

      I think the current setup is going to exclude poor and isolated people, but financially well-off people with a support network of friends/family will be able to navigate the inconveniences and actually get it. I do think you’re right, though, that most people with depression will try every other avenue first before turning to this one, and so it will mostly go to treatment-resistant folks in the long run.

  6. johan_larson says:

    If a drug is an approved treatment for condition A but not B, what is the mechanism that keeps physicians from prescribing it to treat B? Is doing so literally illegal? Does it raise liability concerns? Or is it just that drug plans won’t cover it?

    • yigmaw says:

      Off-label prescription is done all the time and is completely legal. The major issue is that it’s not covered by insurance. Ketamine clinics (and there are many) pretty much all charge cash, like $500 a pop for one in town here.

      • Paul Brinkley says:

        I’m guessing that insurers won’t cover it for condition B because there has been no clinical trial proving it does anything to condition B (in this case, there’s just a general wisdom that it does).

        Vague cheat sheet for clinical trials: there are six. Preclinical shows whether it will kill non-human organisms. Phase 0 shows whether you could put it in a pill without it going bad too fast. Phase I shows whether it will kill healthy people. Phase II shows whether it could have an effect. Phase III shows how much effect it will have. Phase IV shows whether it has any aftereffects after it’s reached market.

        FDA requires Phase III for practically any treatment. Ketamine presumably passed Phase III, but only as anesthesia. It never did for anti-depression. And perhaps not Phase II, either. Total price tag: about $70M.

        I have about seven recommendations to reform US healthcare. The biggest one in terms of cost control is reform of clinical trial regulations, probably abolishing phases II and III if patients are willing to sign a waiver, and letting IV be handled by the pharma to whatever level it wants to boost advertising. (They can pay the FDA to do it, too.)

      • Eric Rall says:

        The major issue is that it’s not covered by insurance.

        It would be more precise to say that it’s substantially less likely to be covered by insurance. It depends on the cost, how aggressive the insurance company is at enforcing cost controls, and how well-established the off-label use is.

        If the off-label use is universally accepted and the medication is dirt cheap, they’re very unlikely to refuse to cover it. A good example here is trazodone, which is on-label only as an antidepressant, but is very commonly prescribed for insomnia. It’s dirt cheap, as cheap or cheaper than most on-label prescription sleep aids, so insurance companies tend to cover it without a peep.

        But I’d imagine they’d be very quick to use “not FDA approved for that purpose” as grounds to refuse to cover a drug that’s considerably more expensive than on-label alternatives, such as in this scenario where IV ketamine at a clinic costs $500/treatment while lexapro or prozac costs $3-10 for month’s supply depending on the pharmacy and the dosage.

        • Matthias says:

          But if a drug is super cheap, why would you need insurance anyway?

          I don’t get paracetamol on insurance.

          • ketaburn says:

            The drug itself is super cheap. If I were allowed to buy it, take it home, and inject it, having to pay $10 for 20 doses without insurance reimbursement would not be an issue. But since the government and the medical establishment are obsessed with “abuse potential” above everything else, they will not let me take it home, and I have to get IM injections at a psychiatrist’s office and pay $250/visit. This is relatively cheap, as most psychiatrists won’t touch it, and at best will refer you to a specialized clinic doing IV ketamine for $600-$800 per treatment.

          • MilfordTrunion says:

            ” I have to get IM injections at a psychiatrist’s office and pay $250/visit. ”

            Well. You don’t pay $250 a visit, your insurance company pays $250 a visit, and you pay whatever the co-pay is plus your yearly premium. Which is the whole idea behind getting this approved by the FDA; it’s not that anyone thought it might not work, it’s about getting insurance to agree that it’s an accepted practice that they ought to pay for.

        • Matt says:

          That matches my experience. I had an acid-reflux caused inflammation in my esophagus that my ENT wanted to treat. He prescribed an inhaler for the problem. The inhaler was designed to shoot the medicine (or be inhaled into the lungs by the patient?) down the trachea, but he wanted me instead to spray it into the back of my throat and then swallow it down. It was designed for (as needed) use and he wanted me to administer it 2 times daily. Definitely helped me with my ‘food sometimes gets stuck in my throat’ issue.

          My insurance company did cover it, and in fact, when I stopped filling the prescription they panicked and got in touch right away to ask me why, because for most people who take that medicine, it’s life-critical to have it on hand when needed. I just told them that I had talked to my doctor and I did not need to continue it.

  7. Anonymous Bosch says:

    Another possibility is that everyone made a huge mistake in using left-handed ketamine, and it’s right-handed ketamine that holds the magic. Most previous research was done on a racemic mixture (an equal mix of left-handed and right-handed molecules), and at least one study suggests it was the right-handed ketamine that was driving the results. Pharma decided to pursue left-handed ketamine because it was known to have a stronger effect on NMDA receptors, but – surprise! – ketamine probably doesn’t work through NMDA after all. So there’s a chance that this is just the wrong kind of ketamine – though usually I expect big pharma to be smarter than that, and I would be surprised if this turned out to be it. I don’t know if anybody has a right-handed ketamine patent yet.

    That would be Chiba University, who has currently licensed it to ATAI.

  8. eremetic says:

    The other reason the drugs have to be given in a specially monitored clinic is because ketamine can have side effects, including hallucinations and dissociative sensations.

    When are they going to restrict Benadryl to specially monitored clinics?

    • martinw says:

      Shhh, don’t give them any ideas..

    • Deiseach says:

      That seems to be only American Benadryl, over here in the British Isles we don’t get the fun Benadryl:

      In the United States and Canada, it is the first-generation antihistamine diphenhydramine.

      In the United Kingdom, the active component of Benadryl is often the antihistamine acrivastine or cetirizine.

      • rtypeinhell says:

        As the FDA is probably well aware, there is nothing at all fun about diphenhydramine hallucinations. I won’t even take the stuff at night anymore, lest spiders start crawling out of the toilet.

      • Lambert says:

        Ask for diphenhydramine by name.
        Many chains of chemists’ sell it own-brand.
        Boots has it, if you’ve got them in the Republic.

      • zzzzort says:

        Though in the UK cough syrup with codeine doesn’t require a prescription, so swings and roundabouts I guess.

  9. yigmaw says:

    In fairness to the psilocybin bit–there is some evidence that the therapy guides the treatment effect. For example in the (admittedly preliminary) studies, people smoke less cigarettes in a psilocybin for tobacco trial, and people drink less alcohol in a psilocybin for alcohol trial, but people don’t smoke less cigarettes in the psilocybin for alcohol trial.

    Similarly MDMA for PTSD. It makes sense that approval for this would be restricted to the MDMA + therapy as it’s not clear you get the same benefit from using it on your own.

  10. Deiseach says:

    I’m as willing to knock Big Pharma and Big Government as much as the next aggrieved working-class loser with a chip on their shoulder, but I can kinda sorta see where the stumbling blocks with “let’s make ketamine legal” might lie.

    (1) “(K)etamine already existed – it’s …a popular recreational drug”. This is going to make people go “uh-oh” before ever you can say “but it will never, ever be abused honest!” since, um, it’s already being abused. This is why we can’t have nice things.

    (2) “But if ketamine is such a good antidepressant, how come it couldn’t pass the normal bar for approval? Like, people keep saying that ketamine is a real antidepressant, that works perfectly, and changes everything, unlike those bad old SSRIs which are basically just placebo.” Did anyone consider that it’s the “I’m still depressed but now I’m high as a kite so I don’t mind it as much anymore, wooooo!” effect possibly at work here? Even a mild “sorta feeling a bit less miserable” is a huge improvement over “I hate my life, I hate myself, I wish I were dead, everything in the entire universe is ultimately meaningless and I still have to drag myself out of bed to go to my stupid horrible job to earn money to keep living for no goddamn reason that makes sense”. When I was handling my depression by drinking it sure worked, but I don’t think anybody is going to try a medical study whereby they conclude “fourteen units of 40% ABV spirits over a week have a positive effect, let’s get the government to allow this as treatment”.

    (3) “We already have a system for drugs you’re worried someone will get addicted to, it’s called the Controlled Substances Act. Ketamine is less addictive than lots of chemicals that are less stringently regulated than it is. This just seems stupid and mean-spirited.” And people regularly abuse those drugs, and this causes problems, and people are currently abusing ketamine. We know this is going to happen if it’s legalised, this is not some slippery slope argument, we’re seeing it happen right now with those aforementioned drugs. Valium was and is a real problem about addiction. People sell on their methadone, even where a methadone programme is reducing opiate addiction, and other people buy it for the high and die from it. You really think somebody is not going to try “Oh hey doc, I’m depressed!” in hopes of getting a legal ketamine prescription? Like I said, this is why we can’t have nice things because we’re humans and humans are stupid and the first thing in our stupid monkey minds is “Can I get blasted off this thing?”.

    (4)”That meant going to a ketamine clinic that had nurses and anesthesiologists for IV access, then sitting there for a couple of hours hallucinating while they infused it into you” and “patients are going to have to drive to some overpriced ketamine clinic a couple of times a week and sit there for two hours per dose just because you think they’re too frail to handle a dissociative drug at home”. If you’re hallucinating in the clinic, you may also hallucinate at home. And you know as well as I do that the first person who takes the legal dose at home, has a bad trip, and does something like fall down the stairs or have an adverse mental reaction, their family will sue all round them for allowing a dangerous practice like that. Do you want to be the doctor who wrote the legal ketamine prescription for Joe Jones and then Joe broke his neck falling down the stairs after taking the dosage at home because he was too out of it to know where he was or what he was doing? I don’t think any doctor wants to take that risk, much less any government, and can you blame them?

    I’d love if ketamine really really worked for depression (and not just because “woooo! high!”) and I’d love if the pitfalls didn’t exist, but they are there and pretending only sensible people will sensibly take their medication in a safe and sensible manner and there won’t be any major adverse effects, and it’s only Big Government being Big Meanies about it stopping this from happening, is pie in the sky thinking.

    • Nornagest says:

      This is going to make people go “uh-oh” before ever you can say “but it will never, ever be abused honest!” since, um, it’s already being abused.

      It’s being used recreationally, sure, but, while plenty of people are happy to tar any off-label recreational use as abusive, what does that actually tell us about how dangerous it is? I don’t care whether someone somewhere might be having fun, I care if the fun they’re having will ruin their life. And while I don’t know how habit-forming ketamine is — most hallucinogens aren’t, but this works very differently from acid or mushrooms — I’ve never heard of anyone e.g. selling their ass for Special K.

      Did anyone consider that it’s the “I’m still depressed but now I’m high as a kite so I don’t mind it as much anymore, wooooo!” effect possibly at work here?

      We went over this last time ketamine came up here. Ketamine gets you high for maybe half an hour; the antidepressant effects allegedly last for months. If there’s anything to those effects, it’s not making you forget your depression because you’re high as shit. It can’t be, because the substance that makes you high as shit physically isn’t present in your system anymore.

      • Tarpitz says:

        I know a number of people who have taken ketamine recreationally; none of them report any lasting ill-effects, and I’ve never heard of anyone becoming addicted to it. I don’t guarantee that it can’t happen, but I’m pretty confident that in practice it happens extremely rarely.

    • thevoiceofthevoid says:

      There’s a vast difference between “must take the first dose in a controlled clinical setting” and “must take every dose in a controlled clinical setting”.

    • MilfordTrunion says:

      “Did anyone consider that it’s the “I’m still depressed but now I’m high as a kite so I don’t mind it as much anymore, wooooo!” effect possibly at work here? ”

      You’re making an interesting argument about the philosophy of thought but when it comes to keeping people from jumping off bridges maybe “not-depressed” versus “you just think you’re not depressed” is not a meaningful distinction.

      “I don’t think anybody is going to try a medical study whereby they conclude “fourteen units of 40% ABV spirits over a week have a positive effect, let’s get the government to allow this as treatment”.”

      If only there were some form of Trial we could do, perhaps in a Clinical setting, where we could identify the adverse effects of a proposed pharmaceutical regimen.

      “If you’re hallucinating in the clinic, you may also hallucinate at home.”

      You are certainly correct here.

  11. mobile says:

    I urge patients only to take hallucinogens/dissociatives in an appropriate setting, such as a rave

    Since the clinical setting is already going to be expensive, why don’t you go all out and hire a DJ?

    • MilfordTrunion says:

      haw. “our study protocol found that synthwave produced the best clinical outcomes, with melodic trance and chillstep also positive…”

  12. ManyCookies says:

    Another possibility is that clinical trials are just really tough on antidepressants for some reason. I’ve mentioned this before in the context of SSRIs. Patients love them. Doctors love them. Clinical trials say they barely have any effect. Well, now patients love ketamine. Doctors love ketamine. And now there’s a clinical trial showing barely any effect. This isn’t really a solution to esketamine’s misery, but at least it has company.

    Are SSRI studies looking at one particular variant at a time, or are they allowing the patients to switch between variants during the trial period? It’s possible that sticking to one SSRI regardless of patient reaction isn’t very effective, but the process of trying multiple SSRIs – which is what psychs actually do – is.

  13. Maxander says:

    The current drug-certification system is roughly reasonable as a way of getting capitalism to pay tens of millions of dollars to run the trials proving a drug is safe and effective, and obviously we don’t want to do without trials that show drugs are safe and effective, but why do these trials cost prohibitive amounts of money in the first place?

    Ketamine, like a bunch of other “available but not certified” drugs, is really cheap. Measuring depressive symptoms can be difficult, and requires some real expertise, but (AFAIK) it generally comes down to questionnaires, which aren’t expensive. Measuring adverse effects is hard, and at that point full-on doctors are probably actually necessary, but a handful of checkups per subject won’t get you to the standard clinical trial price tag. Finally, data collection, analysis, and writeup can be done by postdocs, a resource which is too cheap to meter.

    If each subject requires $200 worth of experimental drug, ten hours of questionnaire-filling (say $50/hour for overpaid attendants), ten hours of physical exams (say $300/hour for overpaid doctors), and a person-month of clinical scientist analysis time (say $10k, but this a single-time fee rather than per-patient), a typical N=1000 study costs… $3,710,000. A hefty sum, but an order of ten less than estimates for a typical clinical trial in most fields (aside from phase I, which is typically N<100.)

    What am I missing?

    • Uribe says:

      I’ve heard people argue that the expensive approval process was essentially set up by Big Pharma as an economic moat against Small Pharma. Twenty years ago, when Big Pharma was a lot more profitable, this seemed like a sensible theory, but these days you would think they’d want the process to be cheaper. Is it possible they can’t kill the monster they created? Or was the conspiracy theory false to begin with? (Can lobbyists accurately be called conspirators?)

      • MilfordTrunion says:

        The people who argue that don’t know what they’re talking about.

        Big Pharma has enough capital to eat any small companies that look like they might be a threat. Shit, that’s the plan of these small companies! Buy some molecules or a professor’s wacky device idea from a university, run the clinical trials, file an NDA, get bought by Pfizer.

        There’s no anticompetitive conspiracy involved in clinical-trial regulations. I mean, sure, regulations are a fixed cost, which provides a competitive benefit to larger firms, but it’s not like the larger firms are sitting around saying “no, please, don’t let us make new drugs, we definitely don’t want to have more exclusive patents in our portfolio”

    • MilfordTrunion says:

      “If each subject requires $200 worth of experimental drug, ten hours of questionnaire-filling (say $50/hour for overpaid attendants), ten hours of physical exams (say $300/hour for overpaid doctors), and a person-month of clinical scientist analysis time (say $10k, but this a single-time fee rather than per-patient), a typical N=1000 study costs… $3,710,000. A hefty sum, but an order of ten less than estimates for a typical clinical trial in most fields (aside from phase I, which is typically N<100.)"

      That's one site. It costs a million or two per site–and that's if you can get the sites cheap. And the FDA wants you to have five or six sites, preferably ten, all in different parts of the country. And you don't get 1000 patients in the door all at once; usually the study criteria are so stringent that you'll get maybe one a week, sometimes one a month or even fewer, and you have to pay to keep the study rolling during that time.

      Now you've *got* all that data, but you still have to do the analysis and write up the report, and that means you need at least a statistician who's making one-fifty K a year if they're worth having, and a tech writer pulling similar numbers to produce a report that the FDA won't laugh off the table. (And you've been paying these people for the last three years of writing the study protocol, right?) Oh, right–when you're done with all this you have to file your NDA. That's about one-point-eight million right there.

      So, y'know. Maybe not an order of magnitude up, but I've tripled your cost just slinging hash here, and that's for a very simple trial with only two people running it and only a couple of sites that get lots of patients right away, none of which is true outside of comments boxes.

  14. Hackworth says:

    “If you look at Table 9 in the FDA report, ketamine did notably worse than most of the other antidepressants the FDA has approved recently – including vortioxetine, an SSRI-like medication.”

    Should that read “esketamine” rather than “ketamine”, or did I just totally lose you?

    • thevoiceofthevoid says:

      Scott argues that he suspects there’s no significant difference between the two, one simply contains 50% mirror image (unless of course it turns out that the mirror image happens to make it fit one receptor or not fit another).

    • Scott Alexander says:

      Yes, good catch.

  15. tmk says:

    > But patients bring prescriptions of OxyContin and Xanax home every day. Come on, FDA. We already have a system for drugs you’re worried someone will get addicted to, it’s called the Controlled Substances Act.

    I’m going to jump on this because it’s a form of bad argument you keep making. Yes, we have a system, and it works really badly. Opioid crisis. Ketamine may be less addictive than opioids, but it’s too late to withdraw OxyContin. People are already addicted and would start buying from the black market and cause lots of crime. That’s no reason to repeat the same mistake with Ketamine.

    • baconbits9 says:

      It’s a little disingenuous to claim he’s making a bad argument and bolstering with “opioid crisis”, the causes of the opioid crisis are not firmly established and its on just as firm a footing to rebut your claim with a “hey, there is a depression crisis” argument.

      • tmk says:

        I don’t see how mentioning the opioid crisis is disingenuous. It’s rather strange to talk about the current system for drugs like OxyContin and not mention it. I’m sure the causes will be debated forever, but “we gave out opioids too easily” seems like a top contender.

        • baconbits9 says:

          It would be strange to talk about a handful of drugs when the schedule is used to regulate many drugs. The problems with your statement are (non exhaustively)

          1. If the FDA is worried about abuse it should use the CSA, if it doesn’t think the CSA is a good framework then it should be aggressively trying to update/repair/replace the CSA, not refusing to let other drugs enter the system because the CSA doesn’t work perfectly.

          2. It is entirely possible that part of the opioid crisis is due to people unable to get access to functional depression treatments and self medicating.

          3. The suicide rate in the US are 5-10x higher than all overdoses for all drugs, the depression crisis is much worse than the opioid crisis.

          4. Your post implies a very weak substitution effect where most of the people who would end up abusing legal ketamine wouldn’t have ended up abusing legal opioids or illegal ketamine.

    • MilfordTrunion says:

      You’re also assuming that the opioid crisis is driven by people reselling over-prescribed Oxy. It isn’t, it’s mostly Chinese Fentanyl bought from the same tattooed guy who sells you Ketamine.

  16. simbalimsi says:

    I think all medical professionals have to compulsorily work for a central government lab that performs those FDA trials and releases it as open source as a public service at least 1 or 2 years after completing their studies before receiving their licenses. This will solve the majority of this problem.

    Also, aren’t public universities doing any research? Shouldn’t that also be open to public? I mean aren’t there any public medicine schools in America?

    • Christophe Biocca says:

      Even assuming that the government could competently run pharmaceutical trials and efficiently choose which substances to test, using the medical equivalent of the draft is a bad idea. Even the military is staffed with volunteers nowadays, because it’s more efficient to raise money through taxes and hire those people who are best suited to the task and/or want the least money to do it.

    • Corey says:

      There are simpler ways if you want to go the public route. The whole industry’s R&D expense is about half of what the governments of these United States already spend buying drugs. (somewhat dated numbers) So we could solve this class of problem while saving money (probably for just the US government itself, definitely for the country as a whole and the rest of the world).

      Would just publicly funding it all be less dysfunctional than the current system? Most people’s priors are p=1 “of course!” so it’s politically impossible, but a fella can dream.

      • Scott Alexander says:

        Wait, why does everyone thinking it would work make it impossible?

        • Corey says:

          My own priors getting the best of me 🙂

        • Corey says:

          I see now, I worded it wrong. I thought everyone would think it’d be *more* dysfunctional, because government. My guess is there would be some exciting new dysfunctions, but probably fewer on net.

          E.g. in a publicly-funded-R&D world trying to approve ketamine, you wouldn’t have this gratuitous enantiomer patent issue, but the anti-fun resistance would be stronger.

  17. TheFlyingFish says:

    How bad would it be to loosen up the rules on drug-testing a bit?

    I agree that we definitely want the drugs we take to be heavily tested, because otherwise people will die. But people with serious medical issues often die anyway, in some cases because they don’t have access to a drug that could potentially save their life only it hasn’t been through enough testing yet.

    What if there were a new intermediary phase, requiring only minimal testing, in which doctors could prescribe a medication but were required to warn the patient that the medication is experimental and may have negative side effects. Then people could decide for themselves whether they were desperate enough to try such a thing. Then, and here’s the kicker, if the patient did experience negative side effects, they wouldn’t be allowed to sue the doctor for malpractice, because they knew the risks from the get-go.

    I’m not terribly familiar with the landscape of drug trials and Big Pharma, so it’s possible there there’s something sort of like this already in place (certainly I’ve heard the term “clinical trials” tossed around a lot, and that sounds sort of like what I’m describing here.) If that’s the case then I guess I’m just advocating a relaxation of the requirements for letting a drug into clinical trials.

    Would this be criminally irresponsible, or could it serve to stimulate more and better research into new drugs?

    • Ghillie Dhu says:

      Perhaps, given the on-label/off-label split that’s already a feature of reality, drugs that pass the min bar for safety (a much less expensive proposition for the pharma company) could be approved with no on-label uses; they’re available for physicians to prescribe off-label, albeit in a somewhat caveat emptor realm.

      Only those drugs whose producers want it to be government certified for particular on-label use need go through the much more time-consuming & expensive process; probably anything that’s not inherently safe but better than the disease (assuming it is effective) would still be held to this standard as well (chemotherapy is the obvious example).

      • beleester says:

        Scott mentioned that insurance won’t pay for off-label uses.

        • Ghillie Dhu says:

          Which would provide the incentive for the producer to bother with the efficacy studies.

      • Evan Þ says:

        According to my dad who’s a statistician analyzing these clinical trials, there’re four stages in the existing process:

        1) Give the drug to healthy volunteers at different doses; see if it hurts them (and if so, how little’s safe).
        2) Give the drug to sick volunteers at different (lower) doses; see if it helps them (and if so, how much’s needed).
        3) Give the drug to sick volunteers at the one dose you decided on with extra randomization and controls; see if it actually does help, and if so if it helps at least as much as existing treatments. This’s the one that takes most of the time and money.
        4) If you’ve gotten through the previous stages and gotten your drug approved (yay!) and in use, keep monitoring for side effects.

        The first three stages here are all called by the name “clinical trials.” When you hear of someone joining a clinical trial, it’s usually stage (3), since that’s the one that takes the most volunteers and the most time.

        What you’re probably asking for is approval-with-no-on-label-uses that skips stage (3)? That would almost certainly get a lot more drugs into circulation. The problem is that companies would have a lot less incentive to go through stage (3) (which, again, involves a whole lot of expense and time) and that means we’d have a lot less information on whether these drugs actually work.

        Perhaps to get their drugs into this approval-but-not-on-label program, companies would be required to start a stage (3) trial anyway? That’d get us the scientific data we want, but companies might still hold back some drugs where stage (2) results don’t seem super-hopeful (did I mention stage (3) involves a whole lot of expense and time?) On the gripping hand, it’s still better than the current system.

        • Christophe Biocca says:

          and if so if it helps at least as much as existing treatments

          That part seems like the biggest mistake to me. Allowing less-effective-but-still-useful drugs to gain approval would help reduce the financial risk in putting drugs through stage III. Sure, if insurance is paying or the disease is severe, patients will probably still pay extra for the better one, but if you’re uninsured in the price difference is 10x, it’d be nice to have the option? This seems like an easy-ish way to put downward pressure on drug prices, without decreasing the incentives to research/trial new drugs.

          • This was the issue in Peltzman’s old article. He found that adding the requirement that the drug be effective as well as safe reduced roughly in half the number of drugs produced while having no detectable effect on their average quality, as measured by several proxies.

          • MilfordTrunion says:

            ” Allowing less-effective-but-still-useful drugs to gain approval ”

            …would lead to a sharp uptick in articles on Mother Jones and Buzzfeed and Slate about how Drug Companies Got This Drug Approved But It Isn’t Any Better And It Costs 10000X As Much And Here’s Some Kid Who Died, FUCK CAPITALISM

          • Garrett says:

            Allowing less-effective-but-still-useful drugs to gain approval

            I’m not certain that makes sense for the drug companies to do.
            Consider two possible outcomes:
            1) New drug is better than placebo.
            2) New drug is better than the current standard of care.

            You are advocating that drugs meeting (1) be approved even if they don’t meet (2). Fair enough.
            But you run into a few problems doing this.
            A) You need to design a study which is set up and powered sufficiently such that it can make the determination of (1) even if it fails to demonstrate (2).
            B) The Expected Value of sales from a drug meeting (1) but not (2) has to be above the cost of a trial which can demonstrate (1) and (2) over just (2).
            C) A drug which is better than the current standard of care is likely to get adopted quickly and widely (obviously depending upon the differences in effects and pricing). This means that the patent protection is worth something.
            D) A drug which is comparable to the current standard of care will make a bit of money, but ultimately is likely to merely drive down the costs of both drugs. Doctors are also less likely to switch existing users or prescribe your treatment as a first-line choice, but at least you have a decent shot at volume if you play your cards right. This means that you don’t make nearly as much money.
            E) A drug which is less good than the current standard of care is unlikely to be selected by anybody unless the current standard of care is found to be ineffective or have idiosyncratic side-effects in a particular patient. The result is that even if you got approval for free, it probably isn’t even worth trying to manufacture.

            I agree that having a larger compendium of pharmaceuticals would be great. But I don’t think the economics and the industry work out that well.

        • Ghillie Dhu says:

          What you’re probably asking for is approval-with-no-on-label-uses that skips stage (3)?

          Not just (3), but (2) as well.

          The problem is that companies would have a lot less incentive to go through stage (3)

          The differential insurance reimbursement rate (plus physician’s own (possibly liability-driven) reticence to prescribe off-label) would provide more incentive than you might guess.

          …we’d have a lot less information on whether these drugs actually work.

          It’s at least conceivable that, with much more immediate broad availability, there’d actually be more data produced; whether the data could be usefully collected for analysis is a separate question.

    • Mary says:

      That’s “compassionate use.” What you may want to question is why it isn’t more widespread.

  18. ketaburn says:

    Rather peripheral to the main point, but you’re underselling it when you say ketamine costs only $10/dose – the links you give are for 20ml vials of 50mg/ml, so a standard IV/IM dose (50mg for a 100kg adult male) costs 1/20 of that, or $0.50. (Or less for smaller people, or 5-10x that from the tattooed man.)

    • One problem mentioned is that it’s injected. But diabetics routinely inject themselves, so presumably someone using Ketamine could.

      • zakamutt says:

        Insulin injections are subcutaneous rather than intravenous, which makes the whole thing significantly easier as there’s less need to aim (I’ve never tried an intravenous injection and don’t know how hard they are, however.) The administration pens most people use (as far as I know) also help, as they are easier to use than your usual syringe.

        • RAPl says:

          Recreational ket is almost never injected IV, because it’s so fast acting you’d black out or go into K-hole instantly with a needle in your arm. If people inject, they usually do it IM. But there is nothing to suggest other routes of administration can’t be used, the most common recreative route is snorting the powder. But you can drink the solution as well, it just requires a higher dose.

  19. Andrew Hunter says:

    The thing that excited me a lot about ketamine, as I understood it, is that it was a (semi?)-permanent cure if it worked at all; take it a few times and you were basically just not a depressed person anymore. (Correct me if this is a wrong impression.)

    The biggest objection I have to traditional antidepressants is that they’re an incredibly brittle solution: you’re now on this drug, basically forever and certainly for months or years [1]. If you develop tolerance, you fall back tto your old condition with no breathing room or margin to deal with the shock, and you’re done. If you have a side effect you don’t like, first off, too bad, you can’t stop taking it without disastrous discontinuation syndrome, and even if you could, congratulations, you’re miserable again!

    This is not a good solution. Even if SSRIs work, which, again, not clear it does, and they didn’t have serious, permanent side effects (which our gracious host has discussed having in his thirties after taking them as a teenager!) this is just bad engineering that I would reject building any system over, because if anything goes wrong in your life–which it will–you are worse than where you started. Unless antidepressants are the only ways you can make progress towards fixing the actual problems in your life that make you miserable, or you really are depressed for no reason having to do with your actual life experience (I have met these people, but they’re rare), this is just papering over a giant sinkhole and waiting for an accident.

    My point, such as it is, is that if nasal ketamine works as I think it might, it might be still a huge improvement over ssris even if it is hugely inconvenient to take any given dose of it. If J&J screwed the pooch and this doesn’t work like ketamine should, that’s bad, but if not, isn’t this still a huge win?

    [1] I’ve heard people say “Oh, no, you only take it for a while and then taper off and recover!” Then why, of the dozens of people on antidepressants I know [2], does precisely one believe she would ever stop taking it, and her only if she gets pregnant, a condition with many negative medical effects that she dreaded precisely and only because she couldn’t take her meds?

    [2] Which, again, something’s clearly a bit wrong with society.

    • ketaburn says:

      It’s not uncommon for people to receive followup treatments. The most common protocol I’ve seen for IM/IV ketamine is six treatments within two weeks, then followups every month or two if needed. Some (like me) are still taking their previous partially-effective antidepressants alongside the ketamine.

      But then, it’s also not uncommon for people not to need followup treatments.

      The most novel part of the ketamine antidepressant experience is how quickly it works. When my psychiatrist asked if my mood was improved an hour after the injection, while I was still shaking off the dizziness, I thought he was joking. But even on the drive home, my partner was commenting on how I was smiling and laughing more freely than she’d seen in years. It was so dramatic and sudden that it gave me greater perspective on just how much my depression had been crippling me; it’s easier to overlook with a drug that takes weeks to work.

    • TheFlyingFish says:

      The biggest objection I have to traditional antidepressants is that they’re an incredibly brittle solution: you’re now on this drug, basically forever and certainly for months or years [1].

      Not only that, but you’re living in a society which has conditioned you to think chemical dependency = bad. Regardless of the objective truth of that statement, if that’s your gut feeling, it’s going to make it pretty difficult not to feel at least a little leery about the prospect of having a chemical dependency for the rest of your life.

    • Scott Alexander says:

      “The thing that excited me a lot about ketamine, as I understood it, is that it was a (semi?)-permanent cure if it worked at all; take it a few times and you were basically just not a depressed person anymore. (Correct me if this is a wrong impression.)”

      I think this is a wrong impression; a ketamine dose lasts a few days; most people get a few weeks’ worth of doses. I think (not sure, very little experience with this drug and few long-term studies) that then you hope it treats the current episode, but sometimes it doesn’t, and people can always have more episodes.

  20. Murphy says:

    This removes most of the advantages of having it be intranasal, so why are they doing this? They give two reasons. First, they want to make sure no patient can ever bring ketamine home, because they might get addicted to it. Okay, I agree addiction is bad. But patients bring prescriptions of OxyContin and Xanax home every day. Come on, FDA. We already have a system for drugs you’re worried someone will get addicted to, it’s called the Controlled Substances Act. Ketamine is less addictive than lots of chemicals that are less stringently regulated than it is. This just seems stupid and mean-spirited.

    I remember getting pissed about the same sort of thing a few years back.

    I eventually came to understand the logic a bit better.

    So, us geeklike people with a tendency to prefer letting people do their own thing and run their own lives when it isn’t too dramatically disastrous basically tend to assume there’s a line.

    A line in the sand, some level of harm which society won’t tolerate going beyond for various reasons like the externalities that drug addicts cause etc. And the and that things beyond the line get banned and the things on the near side don’t.

    But in reality while there’s sort of a line in the sand that many doctors and medical types support …. a huge fraction of normal things in modern society are on the far side of that line.

    I remember an argument with a med student (side note: one who you may have shared classes with scott) where, after much argument it became clear to me that his actual position… that he didn’t want to say out loud because he kinda knew it would be unpopular was that he would absolutely love to get the chance to use the legal system to effectively force treatment on people who, as far as he’s concerned, don’t know any better and thus ban people from drinking alcohol, eating junk food, smoking, failing to eat their greens etc… and his worldview isn’t remotely rare in the medical profession. Because on average people would be healthier. He just wanted what’s best for people.

    As far as he was concerned the line in the sand was a long way back from a lot of things currently socially acceptable.

    Throw in on top of this that the medical associations (read, doctors guilds) have a strong incentive to create legally protected make-work for doctors and you get a situation where one of the big lobbying groups is prestigious medical professionals who are very keen to restrict things as tightly as possible.

  21. Virriman says:

    So there’s a chance that this is just the wrong kind of ketamine – though usually I expect big pharma to be smarter than that, and I would be surprised if this turned out to be it. I don’t know if anybody has a right-handed ketamine patent yet.

    Anyone who’s played Tetris knows that the S shape and the Z shape do not fit into the same spaces. They are different shapes!

    Same thing holds on the molecular level. Wikipedia gives the example of the terpenoids L-Carvone smelling sweet and minty while R-carvone smells spicy with hints of rye. Likewise D-glucose and L-glucose are sweet tasting sugars, but we can only metabolize D-glucose.

    While there might be exceptions (such as when a molecule frequently coverts back and forth with its mirror image naturally such that a pure sample rapidly becomes a mixed sample), there is zero reason to expect mirror image molecules to be equivalent in the body. I’d be interested to know why big pharma decided this would be one of the exceptions, especially if nobody has pinpointed the mechanism by which ketamine treats depression.

  22. Manx says:

    For whom would you recommend ketamine at all? Straight up treatment resistant depression? Are there phenotypes you think could benefit more strongly from it than others? Have you seen patients recover on it?

  23. JulieK says:

    Would it be possible for government or charity to fund clinical trials in cases like this, where we have god reason to expect positive result, but not much hope of profit?

    • Murphy says:

      Possible? absolutely!

      But politically hard.

      Currently companies try their best to avoid wasting money on things that won’t work but it’s not easy. There’s almost no such things as a “sure thing”. Your government office isn’t going to be better at picking a winning horse than a pharma company.

      So some of the time you’re gonna run your phase 3 trials… and the drug is gonna fail.

      With a company they just write it off as shit happens. Some investment are expected to fail.

      But if it happens with the government then you’ll get a long line of clueless nobodies making political hay over the “waste” of hundreds of millions of dollars down the drain and they’re gonna do their best to paint it as “obvious” that the drug would fail and most of the know-nothings listening will believe them.

      And that will be the end of the careers of a bunch of politicians who touched it or anything associated with it.

      Meanwhile ,if it succeeds, there’s little glory in being the minister who rubber stamped the project to assess drug X.

      Success has many fathers, failure is an orphan so there will be lots of other people claiming credit for success and some will likely have better claim like the scientists involved.

      So the political payoff matrix is very poor.

    • Edward Scizorhands says:

      Bill & Melinda Gates Foundation has funded a bunch of clinical trials for things where they just want the end drug. I’m not sure how effective they are, but I know they’ve done it more than once.

  24. Garrett says:

    Are there any studies for depression involving IM instead of IV administration. In the EMS world in my State we’re using it IM for excited delirium. (Technically, we can give it IV, but my medical director notes that if you can get an IV established in someone it’s not excited delirium).

    If IM administration is comparably effective, there are probably other ways to make something like this work, especially for follow-on administration where allergic reactions are unlikely. We’re already approving pharmacists to give flu shots, and we’re approving pharmacists to handle controlled substances. I’m sure we could find a way to combine those two effectively.

    • RAPl says:

      Data from Erowid and other sources strongly suggests there is no difference in nature of effect in any method of administration per se, what differs is the amount of ketamine you need, with taking it orally being the most wasteful route of administration. On streets, it’s most commonly snorted.

  25. Corey says:

    Are the poor showings for antidepressants in clinical trials because placebo works well for treating depression?

  26. Douglas Knight says:

    In this case, whoever spends tens of millions of dollars proving it works to the FDA may not get a monopoly on the drug and the right to sell it for ridiculous prices.

    This refers to the separation of FDA approval from patent monopolies. But, for completeness, note that the FDA does have the power to grant monopolies itself, though usually only 3 years.

  27. MilfordTrunion says:

    The Terrible Awful Truth interpretation:
    “If the FDA approves psilocybin for depression (and it’s currently in Phase 2 trials, so watch this space!) you can bet you’re going to have to go to a special psilocybin clinic if you want to get it. Psychedelic medicine is potentially the future of psychiatry, and there’s every indication that it will be as inconvenient and red-tape-filled a future as possible. If you thought it was tough getting your Adderall prescription refilled every month, just wait.”

    Yep. The FDA knows perfectly well that people who want ketamine can get it. They don’t need to approve it for safety reasons; they only would need to approve it so that patients who want it for mental-health purposes can get their insurance to pay for that usage, and the FDA doesn’t really give two shits about healthcare costs.

    No, the purpose of this is to see gauge patients’ willingness to jump through hoops to get these therapies. Because there’s no way the hundred-year-old fun-haters in Congress are going to allow anyone to just go buy LSD–or THC–any time they want it; so we use something safe and de facto approved, like Ketamine, as a trial run.

    (And besides, if the patients decide not to bother, or the approved regimen is sufficiently shitty as to be useless and everyone stops doing it, that’s a win for Papa Gummint as well, because the FDA can go to all these LSD and THC studies and say “see, we approved Ketamine and it did jack shit, maybe we don’t need to approve acid and pot after all”).

  28. samuel846 says:

    esketamine/S+ isomer has been for sale from Pfizer for a long time under ‘ketanest’, has it not? (I had that and psychadelic wise its a lot better, in fact, all the R- stuff is what you want gone: the groggy, confusing, slowness, having supposedly tried R- alone as well i can confirm this, however that was not from a pharma ampule so I cannot confirm whereas the ketanest was cooked from ampules.

    I’ve never tried IV and often debate it, but this is quite irresponsible as its recreational use.

    Racemic I’ve had from Rotex ampules as well as just plenty of it from ‘street corners’. I’d be really surprised with my uplifting expriences with S+ and racemic if it turned out that R- was the golden one for antidepressant. Though there’s no reason – necessarily – that enjoyment derived from ‘recreational doses’ would be the same enjoyment i assume antidepressants give you.. though I have never taken antidepressants (other than K??)

  29. themistocles says:

    I wonder if you could technically repurpose a large van/old ambulance as a “certified medical office” and drive to where patients are to administer it? Or if you have physicians/personnel go to a patient’s home and monitor, would that fit the letter of the law?

  30. Jan_Rzymkowski says:

    Wake therapy is one of things that visibly help my atypical depression that grew for last 3 years (I visit new famed expert now, but it might not work and I might set some 500$ charity bounty in the near future in my desperate attempt). Now I read that it might work through the same mechanism as ketamine – I now think of trying it to get more data on my illness. (Sulpiride was a miracle cure for me, that lasted over a month, but led to confusion; I still might try getting back on it, though. BTW., anybody think there might be some connection?)

    But back to the wake therapy. I read piece after piece where scientists complain that it’s hard to devise a placebo to wake therapy. Damn, what utter fools! (I’m pulling an all-nighter, so I’m a little over the edge. Also, their incompetence directly influence my misery, so forgive me the anger) You don’t need placebo to do a trial – what you need is ability to do data analysis above mere basics. When treating depression you run into an issue of how much self-report mirrors the actual improvement of the condition and how persistent would the improvement be. IT HAS NOTHING TO DO WITH PLACEBOS. Not to mention they’re like in 80% just regression to the mean. Damn, if you devised a way to make patient’s mere awareness of therapy cure his drug-resistant depression, THEN HOW ON EARTH IS THAT NOT A WORLD CHANGING NEWS?! I put sincere faith in 10+ different chemical compounds and it didn’t help me – and you worry that maybe I’ll be cured not by act, but your words alone?! IT’S MATTHEW 8:6-13 ALL OVER AGAIN!!!

  31. Yurgh says:

    Looking at this patent application (nasal, racemic) it seems like marketing this as an analgesic is a big part of the goal. It’s nice to think that the pharm industry is nobly taking on the plight of the small percentage of the population with treatment resistant depression but the market for less addictive painkillers has gotta be more of a financial motivator. Which may be why they chose esketamine – according to this patent it’s 2x as analgesic, but it also has more of the lousy you gotta take this in a clinic cause we’re scared to let you out side effects than the more anti-depressant less wooze inducing arketamine.

    http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=1&f=G&l=50&co1=AND&d=PG01&s1=15%2F849587&OS=15/849587&RS=15/849587

    ATAI Life, a German co, via Perception Neuroscience out of NY, seem to have the patent on arketamine.

    https://www.thepharmaletter.com/article/atai-pins-hopes-on-arketamine