Scott Alexander on Wednesday: I’ve spent fifteen years not responding to [Hanson’s medicine] argument, because I worry it would be harsh and annoying to use my platform to beat up on one contrarian who nobody else listens to. But I recently learned Bryan Caplan also takes this seriously. Beating up on two contrarians who nobody else listens to is a great use of a platform!
Scott’s earlier work was awesome. He changed the way I saw the world. For five years, probably more, the quality had deteriorated. This is a new low. Such a shame. Anyway. This is a good piece and a good response.
Surprising and a bit disappointing from Scott who I usually expect to go very deep into a topic. This seems like he only skimmed for the prep. Or he probably could have just asked.
Not really sure what his motivation is unless he’s just trying to release as many words as possible and this one seemed easy.
Oh nice, a rare moment where I can almost fully agree with Robin. Medicine is in fact, a cesspool whose most likely outcomes are commonly the following two: 1) Nothing much happens. and 2) Things get worse(ultimately). That's medicine, in a nutshell. Some fields of medicine are worse than others(Psychiatry being the worst offender by far-- uh oh... a conflict of interest is already coming into view...).
What I can criticize is how Scott and Robin engaged in debating the subject. Both people, basically get lost in the details. Which is a main critique I have towards people in the "Rationalist" community-- I believe they are pathologically pragmatic, care about specifics and details and precision (which is fine to care about, but not if it's almost all you care about or how you fundamentally and philosophically structure your representation of reality).
They get lost in the details of certain studies, they get lost on certain points, all of this is utterly narrow and myopic as usual. How would I approach this? I'd start with just stating the facts of why medicine is actually a shit show: 1) Bad incentives. The whole thing, is structured as a business, for the most part, where the goal is ultimately, making money, not making people better. When everyone's healthy, you can't make money. Uh Oh! (See how this big picture stuff really has a beautiful simplicity to it?)
2) Bad people(Humans are bad in general(with loads of narratives and beliefs about them being good, or how the question itself is wrong), which is why they get confused about almost everything they do(the bugs are features, for humans). When they can use medicine to wage war or dominate each other, or dominate the bottom rungs of their hierarchy so they can more easily wage war with other hierarchies, they will do so. Like how psychiatry has historically been used and almost certainly remains today as tool of profit and social engineering in a more sophisticated form rather than anything genuinely therapeutic(There's that conflict of interest again. Hey is the CIA still doing booths at the APA's big events?)
3) Remember that detail orientation problem rationalists have? Where they miss the forest for the trees? Well that entire approach, which I like to call "autistic" with a big emphasis that I am not using this description in a pejorative sense, have nothing but well wishes for autistic people, and just find to be useful to describe a style of thinking that obsessives over precision, quantity, can be robotic, cold, and so glued to a ground-level, to superficiality, that it misses the point in a tragic way. Virtually anyone involved in a "rationalist" space will show symptoms of this flawed way of thinking, myself included.
The problem is-- this exact pathology a major foundation of the approach to Western pharmacology and most treatment approaches. This is slowly changing because we can't help but understand things slightly less superficially and narrowly with time, but we're still very dense. Let's get into the details for once. Take a skin condition. Let's say a bacteria or fungi causes it. What does Western medicine have to say? Oh, easy-- a topical antibacterial or a topical antifungal can do the trick. Of course, no one fucking asked, "Hey, why does this condition come up in the first place beyond mere appearances?" That's just not seen as very "medical". Instead, it's this autistic Cause->Effect superficial bullshit. What happens? Symptoms may reduce, but return again(Big picture problem never addressed). What also happens? The company that makes the anti-bacterial/anti-fungal makes more money. We need more ointment! What also happens? Selection pressure towards the pathogen, making it stronger. Are you... seeing the problem yet?
This is not going to perfectly represent all medical failure, but this little analogy represents an overwhelming amount of it if you understood the fundamental problem and don't get lost in the details(Zoom out for once in your life), and that is only the slightest introduction as to why medicine currently, is a cesspool.
Yes, this was an uncharacteristically weak post from Scott. I wish he had engaged with your actual arguments and applied his usual level of rigour to the evidence, rather than fighting a straw man.
I find myself sympathetic to Robin's view on the entirely flimsy basis that most medical treatment I have received have been useless. There are exceptions such as antibiotic treatment for an e. coli infection of my kidneys (very nasty and almost fatal!), some dental work (I suspect much is a waste of money), and probably flu and other vaccines to an extent (not so much Covid). I have had two spine problems for which experts provided zero help, just scans, sonograms, and worthless pain killers. I have doctor-caused damage to my left sciatic nerve which is never going to get any better (barring major new developments). I have for decades had chronic and too often really awful phlegm problems which remain unchanged despite trying probably a dozen sprays and pills and other treatments. I also continue to age without a single effective intervention.
Obstetrics is a good example of this. The USA is suffering from too much medical care. Countries that use midwives get better outcomes. We get too many unnecessary c-sections because that makes the doctor more money. It’s been a few years since I’ve looked at the research but of my memory is correct, the often cited stat that black women have worse outcomes is a result (in part) of increased rates of Medicaid. Doctors are more likely to over-treat Medicaid patients because they pay per service while a private insurance company will decline to pay for unnecessary procedures. Episiotomies are an excellent example, they are harmful and no study has ever found a positive effect. They should never be done. But Medicaid pays for it and private insurance won’t.
My first child was taken to the NICU for 4 days even though there was nothing wrong with her. The doctor kept giving us bs reasons that she had to stay, and even when she gained weight and did all the things he wanted, he insisted she stay. When the doctor changed out we were sent straight home. My private insurance found it unnecessary and refused to pay for the NICU stay at all.
I am glad you wrote up this response. I was really struggling to understand Scott's position since every time I had spoken with you or heard you speak on the matter your position had pretty consistently been "Some medicine is good, but most is a combination of useless, harmful, or generally the cost is greater than the benefit." I can see how the casino example is confusing if one doesn't know quite what your arguments have been over all, but having an inkling of those it seemed like a strange case to latch on to. Unless one assumes you had a significant stroke or something that wildly changed your position in a strange way.
I think this is an explained variance problem. Variance in medical care use explains negligible variance in health outcomes between individuals, but that doesn't mean that it doesn't have an effect in the aggregate.
Apparently he's fighting a straw argument because you once made a certain claim and he's afraid lots of people are going to stand by that decontextualized claim.
Thanks for the detailed response. I broadly agree with the stochastic concept of aging/dying. I have also often wondered how much of the (substantial) effect of say exercise on longevity is in fact genetic. That said, implicit in that very model is there is a certain rate of decay of our bodies. Based on almost every system we have ever seen we should be able to accelerate or slow down that rate, which is exactly what for example smoking and drinking, on one hand, and exercise and sauna with ice baths on the other, appear to do.
Accordingly, I do not follow quite how you go from that hypothesis to what appears to be a conclusion that there is nothing else to be done but wait and die. Some specific comments:
1. You appear not to value healthspan as opposed to lifespan. I think both are important.
2. I assume that you agree that exercise affects the shape of the curve? I agree that observational results from exercise are likely confounded in part with genetics, but it is a very strong claim that all of it is. I would see the burden of evidence there being on you not the rest of the world.
3. The role of vaccines in this is complicated. One can hypothesise a world in which, had the vaccines I mentioned not existed, we would have evolved to be resistant (in some ways a “better” species, except that not necessarily on the axes that we are most interested in such as fertility or intelligence). But that is pure supposition. Much more likely, and indeed consistent with the data we see (some of which you cite!) those vaccines effectively push the stochastic curve out to the right, if not its end point. That seems like a win for medical science.
4. I kinda do agree with regard to most cancers, flu, etc, because by and large these are the symptoms of a degraded metabolism. But that implies my point 1. It also implies that specific treatments to alleviate these degraded metabolisms (i.e. semaglutide, metformin, insulin) are in fact useful at moving the curve to the right.
5. It seems very hard to disagree with the effectiveness of A&E medicine: lots of people have accidents that would kill them or significantly shorten their lifespans, let alone health-spans, and are restored to nearly original function thanks to effective medical interventions. These may not increase total lifespan, but it does push the curve out to the right again.
I am certainly skeptical about medicine but I am even more skeptical about dramatic absolute claims.
Referring to you dismissively as someone "nobody listens to" is really surprising. I've always admired Scott as somebody who rises above petty rhetoric like that.
I've said this before, but your single minded focus on years of life when it comes to this issue is bizarre. Medical care has at least two other very important goals and these sometimes trade off with life extension.
The first is reduction in pain, and in discomfort broadly construed. Your link to the piece about prescription drug deaths is telling. It notes that some of the biggest killers are pain medications including NSAIDs.
Well maybe if you live with chronic pain, a shorter lifespan plus relief of your pain is worth it. Same goes for the deaths from falling later in life due to antidepressants.
Another aim of medicine is the provision of information that people want. They want to know if they have cancer, even if the cancer isn't treatable. A lot of medicine works to provide this benefit.
How old are Scott's infant twins? As someone barely able to do my job in the midst of the chronic sleep deprivation and distraction imposed my by own two babies, that's where my head goes.
Scott’s earlier work was awesome. He changed the way I saw the world. For five years, probably more, the quality had deteriorated. This is a new low. Such a shame. Anyway. This is a good piece and a good response.
Surprising and a bit disappointing from Scott who I usually expect to go very deep into a topic. This seems like he only skimmed for the prep. Or he probably could have just asked.
Not really sure what his motivation is unless he’s just trying to release as many words as possible and this one seemed easy.
Oh nice, a rare moment where I can almost fully agree with Robin. Medicine is in fact, a cesspool whose most likely outcomes are commonly the following two: 1) Nothing much happens. and 2) Things get worse(ultimately). That's medicine, in a nutshell. Some fields of medicine are worse than others(Psychiatry being the worst offender by far-- uh oh... a conflict of interest is already coming into view...).
What I can criticize is how Scott and Robin engaged in debating the subject. Both people, basically get lost in the details. Which is a main critique I have towards people in the "Rationalist" community-- I believe they are pathologically pragmatic, care about specifics and details and precision (which is fine to care about, but not if it's almost all you care about or how you fundamentally and philosophically structure your representation of reality).
They get lost in the details of certain studies, they get lost on certain points, all of this is utterly narrow and myopic as usual. How would I approach this? I'd start with just stating the facts of why medicine is actually a shit show: 1) Bad incentives. The whole thing, is structured as a business, for the most part, where the goal is ultimately, making money, not making people better. When everyone's healthy, you can't make money. Uh Oh! (See how this big picture stuff really has a beautiful simplicity to it?)
2) Bad people(Humans are bad in general(with loads of narratives and beliefs about them being good, or how the question itself is wrong), which is why they get confused about almost everything they do(the bugs are features, for humans). When they can use medicine to wage war or dominate each other, or dominate the bottom rungs of their hierarchy so they can more easily wage war with other hierarchies, they will do so. Like how psychiatry has historically been used and almost certainly remains today as tool of profit and social engineering in a more sophisticated form rather than anything genuinely therapeutic(There's that conflict of interest again. Hey is the CIA still doing booths at the APA's big events?)
3) Remember that detail orientation problem rationalists have? Where they miss the forest for the trees? Well that entire approach, which I like to call "autistic" with a big emphasis that I am not using this description in a pejorative sense, have nothing but well wishes for autistic people, and just find to be useful to describe a style of thinking that obsessives over precision, quantity, can be robotic, cold, and so glued to a ground-level, to superficiality, that it misses the point in a tragic way. Virtually anyone involved in a "rationalist" space will show symptoms of this flawed way of thinking, myself included.
The problem is-- this exact pathology a major foundation of the approach to Western pharmacology and most treatment approaches. This is slowly changing because we can't help but understand things slightly less superficially and narrowly with time, but we're still very dense. Let's get into the details for once. Take a skin condition. Let's say a bacteria or fungi causes it. What does Western medicine have to say? Oh, easy-- a topical antibacterial or a topical antifungal can do the trick. Of course, no one fucking asked, "Hey, why does this condition come up in the first place beyond mere appearances?" That's just not seen as very "medical". Instead, it's this autistic Cause->Effect superficial bullshit. What happens? Symptoms may reduce, but return again(Big picture problem never addressed). What also happens? The company that makes the anti-bacterial/anti-fungal makes more money. We need more ointment! What also happens? Selection pressure towards the pathogen, making it stronger. Are you... seeing the problem yet?
This is not going to perfectly represent all medical failure, but this little analogy represents an overwhelming amount of it if you understood the fundamental problem and don't get lost in the details(Zoom out for once in your life), and that is only the slightest introduction as to why medicine currently, is a cesspool.
Yes, this was an uncharacteristically weak post from Scott. I wish he had engaged with your actual arguments and applied his usual level of rigour to the evidence, rather than fighting a straw man.
I find myself sympathetic to Robin's view on the entirely flimsy basis that most medical treatment I have received have been useless. There are exceptions such as antibiotic treatment for an e. coli infection of my kidneys (very nasty and almost fatal!), some dental work (I suspect much is a waste of money), and probably flu and other vaccines to an extent (not so much Covid). I have had two spine problems for which experts provided zero help, just scans, sonograms, and worthless pain killers. I have doctor-caused damage to my left sciatic nerve which is never going to get any better (barring major new developments). I have for decades had chronic and too often really awful phlegm problems which remain unchanged despite trying probably a dozen sprays and pills and other treatments. I also continue to age without a single effective intervention.
Obstetrics is a good example of this. The USA is suffering from too much medical care. Countries that use midwives get better outcomes. We get too many unnecessary c-sections because that makes the doctor more money. It’s been a few years since I’ve looked at the research but of my memory is correct, the often cited stat that black women have worse outcomes is a result (in part) of increased rates of Medicaid. Doctors are more likely to over-treat Medicaid patients because they pay per service while a private insurance company will decline to pay for unnecessary procedures. Episiotomies are an excellent example, they are harmful and no study has ever found a positive effect. They should never be done. But Medicaid pays for it and private insurance won’t.
My first child was taken to the NICU for 4 days even though there was nothing wrong with her. The doctor kept giving us bs reasons that she had to stay, and even when she gained weight and did all the things he wanted, he insisted she stay. When the doctor changed out we were sent straight home. My private insurance found it unnecessary and refused to pay for the NICU stay at all.
I am glad you wrote up this response. I was really struggling to understand Scott's position since every time I had spoken with you or heard you speak on the matter your position had pretty consistently been "Some medicine is good, but most is a combination of useless, harmful, or generally the cost is greater than the benefit." I can see how the casino example is confusing if one doesn't know quite what your arguments have been over all, but having an inkling of those it seemed like a strange case to latch on to. Unless one assumes you had a significant stroke or something that wildly changed your position in a strange way.
All I got away was that medicine is effective conditional on one actually needing it. And a bunch of stuff about insurance
I think this is an explained variance problem. Variance in medical care use explains negligible variance in health outcomes between individuals, but that doesn't mean that it doesn't have an effect in the aggregate.
According to these article, prescription drugs are not that helpful:
https://brownstone.org/articles/prescription-drugs-are-the-leading-cause-of-death/
...
Also I like very much some pertinent articles from https://www.midwesterndoctor.com/, like osteoporosis, statins, antidepressants...
Apparently he's fighting a straw argument because you once made a certain claim and he's afraid lots of people are going to stand by that decontextualized claim.
Apart from getting older, Scott has also gotten noticeably more cowardly since his real identity was revealed.
Thanks for the detailed response. I broadly agree with the stochastic concept of aging/dying. I have also often wondered how much of the (substantial) effect of say exercise on longevity is in fact genetic. That said, implicit in that very model is there is a certain rate of decay of our bodies. Based on almost every system we have ever seen we should be able to accelerate or slow down that rate, which is exactly what for example smoking and drinking, on one hand, and exercise and sauna with ice baths on the other, appear to do.
Accordingly, I do not follow quite how you go from that hypothesis to what appears to be a conclusion that there is nothing else to be done but wait and die. Some specific comments:
1. You appear not to value healthspan as opposed to lifespan. I think both are important.
2. I assume that you agree that exercise affects the shape of the curve? I agree that observational results from exercise are likely confounded in part with genetics, but it is a very strong claim that all of it is. I would see the burden of evidence there being on you not the rest of the world.
3. The role of vaccines in this is complicated. One can hypothesise a world in which, had the vaccines I mentioned not existed, we would have evolved to be resistant (in some ways a “better” species, except that not necessarily on the axes that we are most interested in such as fertility or intelligence). But that is pure supposition. Much more likely, and indeed consistent with the data we see (some of which you cite!) those vaccines effectively push the stochastic curve out to the right, if not its end point. That seems like a win for medical science.
4. I kinda do agree with regard to most cancers, flu, etc, because by and large these are the symptoms of a degraded metabolism. But that implies my point 1. It also implies that specific treatments to alleviate these degraded metabolisms (i.e. semaglutide, metformin, insulin) are in fact useful at moving the curve to the right.
5. It seems very hard to disagree with the effectiveness of A&E medicine: lots of people have accidents that would kill them or significantly shorten their lifespans, let alone health-spans, and are restored to nearly original function thanks to effective medical interventions. These may not increase total lifespan, but it does push the curve out to the right again.
I am certainly skeptical about medicine but I am even more skeptical about dramatic absolute claims.
Referring to you dismissively as someone "nobody listens to" is really surprising. I've always admired Scott as somebody who rises above petty rhetoric like that.
I've said this before, but your single minded focus on years of life when it comes to this issue is bizarre. Medical care has at least two other very important goals and these sometimes trade off with life extension.
The first is reduction in pain, and in discomfort broadly construed. Your link to the piece about prescription drug deaths is telling. It notes that some of the biggest killers are pain medications including NSAIDs.
Well maybe if you live with chronic pain, a shorter lifespan plus relief of your pain is worth it. Same goes for the deaths from falling later in life due to antidepressants.
Another aim of medicine is the provision of information that people want. They want to know if they have cancer, even if the cancer isn't treatable. A lot of medicine works to provide this benefit.
How old are Scott's infant twins? As someone barely able to do my job in the midst of the chronic sleep deprivation and distraction imposed my by own two babies, that's where my head goes.