6 ms·

Actual ER doctors disagree with you: https://news.ycombinator.com/item?id=36336615 https://news.ycombinator.com/item?id=36336615 https://news.ycombinator.com/

by Kiro 3y ago

Actual ER doctors disagree with you: https://news.ycombinator.com/item?id=36336615 https://news.ycombinator.com/item?id=36336615 https://news.ycombinator.com/item?id=36337227 https://news.ycombinator.com/item?id=36337227 Classic HN to confidently dismiss something in areas we're actually clueless about.

notahacker 3y ago
I'm going to go out on a limb and suggest that most people on HN have some experience of the patient family side of medical problems, some of them no doubt significant and regular. We're not doubting the medical situation the ER doctor agrees sounds plausible, that GPT could have produced that response or that doctors finding it difficult to communicate with family members in intelligible and empathetic terms is a real problem. But if you're the sort of person that would "melt into calm agreeability" when basically the same explanation was offered with generic corpspeak appended (i.e. they'd trust the doctor more if he was answering your question by reading from a script so non-specific and non-empathetic it finished up with "if you have any questions or concerns please contact the medical team"!), and would also be delighted to get the same script read out to you each time you asked a different staff member the same question, I think you're very much in the minority. Certainly in my limited experience, I can assure you that I was more reassured by the empathy levels of professionals completely misunderstanding my question and thinking I was threatening to make a complaint about their standard of care than I would have been if they pulled out an index card and read that they were all doing their best, the treatment is [boilerplate], please do not hesitate to contact medical team in case of any questions and we will reread index card to you. At best, I suppose, my reaction might be to interpret the index card boilerplate repetition as a polite way of telling me to fuck off and not offer any followup questions. You could even make an argument that this is medically useful; a more specific version of "here's a leaflet" so they can get on with doing their job. But as written where everybody loves the GPT boilerplate, the story reads like a classic of the LinkedIn/politician people miraculously came round to supporting me and everybody cheered genre.
Calavar 3y ago
For what it's worth, I'm a doctor and I find this story hard to believe. It does sound to me like wishful thinking by an AI hype guy. In my experience, these sorts of situations happen when two things come together: 1. The patient's family has just enough medical knowledge to fall onto the wrong part of a Dunning-Kruger curve. 2. The family has certain beliefs about the medical establishment. Namely, that medical staff (especially doctors) are trained in medical school to have an inflexible way of thinking and that the doctors are dismissive of the family's proposed treatment because they are married to textbook thinking and rigidly following hospital protocols. #1 happens all the time - maybe even the majority of the time - but it isn't enough to cause conflict on its own. #2 is the special sauce that makes the situation boil over. What is the absolute last thing you want to do given #2? Probably something along the lines of handing every member of the staff a pre-written script to recite every time a family member asks questions. This will make them feel like they are being stonewalled, not like they are being listened to. This will confirm their fears about point #2, not ameliorate them. I can't say I've ever seen someone print off a script quite like this before, but I have seen some doctors piss off patients/families by relying too much on pat phrases or repeatedly pointing to a particular clinical guideline or hospital policy as the rationale for their decision.
- ClumsyPilot 3y ago
  > The family has certain beliefs about the medical establishment. Namely, that medical staff (especially doctors) are trained in medical school to have an inflexible way of thinking This belief saved my mother's life. We came with complainst of lung issies, for some reason doctors decided allergies are mosh likely but we wanted an X-ray. Months later they finally did a scan and found lung cancer. I don't know if they are following a flow chart, but when dealing with NHS doctors I sometimes feel like I am talking to a bot. They have the same routines, for exanple they will never ever check for Vitamin D defficiency, etc. I cant imagine what will happen once real bots / chatgpt becomes widespread.c
  - Calavar 3y ago
    Well as with most beliefs, sometimes it's wrong and sometimes it's right. I don't know anything about your mother's case, and I don’t personally have any experience with the NHS, so I can't say if it would have been appropriate to get a CT earlier in her case. Maybe it would have been. I do feel that we as doctors can be too rigid in our decision making sometimes, but my thoughts on this are quite nuanced and probably very different from the average layperson's. It's probably also a multi-page writeup, so I won't get into it here. In general though, CT scans cause cancer in about 1 out of every 1000 cases [1], so if we ordered them on everyone who asked, we could very well cause more cancer than we diagnosed. In the US, we already have system that is much more patient driven and much more aggressive on testing than most of western Europe, and we still have worse health outcomes. I agree with your concerns about Chat GPT. Patients want face to face time and I think they deserve that. Unfortunately that's low hanging fruit for C-suite executives to pick at when cutting costs. [1] https://www.bmj.com/content/346/bmj.f2360 https://www.bmj.com/content/346/bmj.f2360
sarchertech 3y ago
My wife, an ER doctor, doesn’t believe that the doctor read the script to the patient. She thought that was completely absurd. She started laughing when she got to that part of the article.
epcoa 3y ago
I don’t think that first one is an ER doctor (having ER experience is not ER doctor - at least until COVID every medical student had 4 weeks of “experience”) - you’d just say you’re an ED physician. But more pertinent, on reading it and a reply I think what is being deemed realistic is the scenario (ie patients family arguing treatment) but that’s not terribly controversial, the followup comment even admits that they would not feel good if they were presented with an AI readout: “If they are not processing the information then you have to allow for that, but I do think some relatives would react very badly if given a printout from an AI to explain a situation”. So I don’t think this makes your case as strongly as you think. The other one is hearsay agreeing with the “premise” which is a charitable way of saying the story itself is likely bullshit - and in the end disputes that ChatGPT is useful here (go actually read the comment). For what it’s worth I think the meat of the story (ie the use of ChatGPT) is farcical. One, I’ve been on the other side even before medical training, this idea that no doctor could imagine what it’s like to be on the other side is stupid. When I had questions about my dying family member prior to any medical training if the response was to pull out a canned script of what are clearly corp speak platitudes, I would have requested a different provider or a transfer. And finally, pulmonary edema. Good example for an uninformed audience that doesn’t know better, but this is waaay too common - an ED doc in a medium busy service can easily see 10 presentations a day. This is so basic that this type of conversation happens many many times over (commonly multiple times a day for this same presentation). Answering family questions and addressing objections over routine presentations is something that an attending should have been doing for nearly a decade at least. In academic medicine we talk about how to do this better but I see no evidence this ChatGPT nonsense presented has anything new to add.
EGreg 3y ago
Like Web3 smart contracts which will experience a resurgence in 2024?

by Kiro 3y ago

notahacker 3y ago

I'm going to go out on a limb and suggest that most people on HN have some experience of the patient family side of medical problems, some of them no doubt significant and regular. We're not doubting the medical situation the ER doctor agrees sounds plausible, that GPT could have produced that response or that doctors finding it difficult to communicate with family members in intelligible and empathetic terms is a real problem. But if you're the sort of person that would "melt into calm agreeability" when basically the same explanation was offered with generic corpspeak appended (i.e. they'd trust the doctor more if he was answering your question by reading from a script so non-specific and non-empathetic it finished up with "if you have any questions or concerns please contact the medical team"!), and would also be delighted to get the same script read out to you each time you asked a different staff member the same question, I think you're very much in the minority. Certainly in my limited experience, I can assure you that I was more reassured by the empathy levels of professionals completely misunderstanding my question and thinking I was threatening to make a complaint about their standard of care than I would have been if they pulled out an index card and read that they were all doing their best, the treatment is [boilerplate], please do not hesitate to contact medical team in case of any questions and we will reread index card to you. At best, I suppose, my reaction might be to interpret the index card boilerplate repetition as a polite way of telling me to fuck off and not offer any followup questions. You could even make an argument that this is medically useful; a more specific version of "here's a leaflet" so they can get on with doing their job. But as written where everybody loves the GPT boilerplate, the story reads like a classic of the LinkedIn/politician people miraculously came round to supporting me and everybody cheered genre.

Calavar 3y ago

For what it's worth, I'm a doctor and I find this story hard to believe. It does sound to me like wishful thinking by an AI hype guy. In my experience, these sorts of situations happen when two things come together: 1. The patient's family has just enough medical knowledge to fall onto the wrong part of a Dunning-Kruger curve. 2. The family has certain beliefs about the medical establishment. Namely, that medical staff (especially doctors) are trained in medical school to have an inflexible way of thinking and that the doctors are dismissive of the family's proposed treatment because they are married to textbook thinking and rigidly following hospital protocols. #1 happens all the time - maybe even the majority of the time - but it isn't enough to cause conflict on its own. #2 is the special sauce that makes the situation boil over. What is the absolute last thing you want to do given #2? Probably something along the lines of handing every member of the staff a pre-written script to recite every time a family member asks questions. This will make them feel like they are being stonewalled, not like they are being listened to. This will confirm their fears about point #2, not ameliorate them. I can't say I've ever seen someone print off a script quite like this before, but I have seen some doctors piss off patients/families by relying too much on pat phrases or repeatedly pointing to a particular clinical guideline or hospital policy as the rationale for their decision.

ClumsyPilot 3y ago
> The family has certain beliefs about the medical establishment. Namely, that medical staff (especially doctors) are trained in medical school to have an inflexible way of thinking This belief saved my mother's life. We came with complainst of lung issies, for some reason doctors decided allergies are mosh likely but we wanted an X-ray. Months later they finally did a scan and found lung cancer. I don't know if they are following a flow chart, but when dealing with NHS doctors I sometimes feel like I am talking to a bot. They have the same routines, for exanple they will never ever check for Vitamin D defficiency, etc. I cant imagine what will happen once real bots / chatgpt becomes widespread.c
- Calavar 3y ago
  Well as with most beliefs, sometimes it's wrong and sometimes it's right. I don't know anything about your mother's case, and I don’t personally have any experience with the NHS, so I can't say if it would have been appropriate to get a CT earlier in her case. Maybe it would have been. I do feel that we as doctors can be too rigid in our decision making sometimes, but my thoughts on this are quite nuanced and probably very different from the average layperson's. It's probably also a multi-page writeup, so I won't get into it here. In general though, CT scans cause cancer in about 1 out of every 1000 cases [1], so if we ordered them on everyone who asked, we could very well cause more cancer than we diagnosed. In the US, we already have system that is much more patient driven and much more aggressive on testing than most of western Europe, and we still have worse health outcomes. I agree with your concerns about Chat GPT. Patients want face to face time and I think they deserve that. Unfortunately that's low hanging fruit for C-suite executives to pick at when cutting costs. [1] https://www.bmj.com/content/346/bmj.f2360 https://www.bmj.com/content/346/bmj.f2360

sarchertech 3y ago

My wife, an ER doctor, doesn’t believe that the doctor read the script to the patient. She thought that was completely absurd. She started laughing when she got to that part of the article.

epcoa 3y ago

I don’t think that first one is an ER doctor (having ER experience is not ER doctor - at least until COVID every medical student had 4 weeks of “experience”) - you’d just say you’re an ED physician. But more pertinent, on reading it and a reply I think what is being deemed realistic is the scenario (ie patients family arguing treatment) but that’s not terribly controversial, the followup comment even admits that they would not feel good if they were presented with an AI readout: “If they are not processing the information then you have to allow for that, but I do think some relatives would react very badly if given a printout from an AI to explain a situation”. So I don’t think this makes your case as strongly as you think. The other one is hearsay agreeing with the “premise” which is a charitable way of saying the story itself is likely bullshit - and in the end disputes that ChatGPT is useful here (go actually read the comment). For what it’s worth I think the meat of the story (ie the use of ChatGPT) is farcical. One, I’ve been on the other side even before medical training, this idea that no doctor could imagine what it’s like to be on the other side is stupid. When I had questions about my dying family member prior to any medical training if the response was to pull out a canned script of what are clearly corp speak platitudes, I would have requested a different provider or a transfer. And finally, pulmonary edema. Good example for an uninformed audience that doesn’t know better, but this is waaay too common - an ED doc in a medium busy service can easily see 10 presentations a day. This is so basic that this type of conversation happens many many times over (commonly multiple times a day for this same presentation). Answering family questions and addressing objections over routine presentations is something that an attending should have been doing for nearly a decade at least. In academic medicine we talk about how to do this better but I see no evidence this ChatGPT nonsense presented has anything new to add.

EGreg 3y ago

Like Web3 smart contracts which will experience a resurgence in 2024?