6 ms·

I am a doc. I am also a medical researcher, but you (and most other people here) _have_ to realize that the vast majority of practicing MDs are _not_ scientists

by rscho 5y ago

I am a doc. I am also a medical researcher, but you (and most other people here) _have_ to realize that the vast majority of practicing MDs are _not_ scientists, and understand almost nothing when reading medical studies. And that's fine, because it's not their job. The profession is bound by guidelines and expert opinion. Experts do research, formulate guidelines, and guidelines get applied by practitioners. And no, that absolutely does not mean you know better than the average doc because you've read 10 studies. That's telling a highly specialized labourer that you know the job better because you've read 0.0001% of the theory.

deadpannini 5y ago
The dissonance here is intense. On the one hand, "practicing MDs ... understand almost nothing when reading medical studies", but also I don't "know better than the average doc [just] because [I've] read 10 studies." Is the idea that doctors deserve deference because they are conditioned to submit to guidelines formulated by experts? There is a lack of consensus on many long-standing, common medical questions, much less novel diseases. When I ask my GP about prostate screening, for example, he lays out the current state of research as best he can, but delegates the decision to me. Why is the COVID situation different, given that experts are still fighting through a figurative fog of war?
- nickspacek 5y ago
  Great point. Perhaps the idea is that neither you nor the avg doc can conclude much from reading 10 studies. Therefore, defer to the experts? As to why the COVID situation is different, presumably because prostate cancer screening affects only you (and perhaps your family in the case of a true/false positive/negative diagnosis), and COVID represents a threat to everyone around you, and everyone around them, etc.
  - deadpannini 5y ago
    I wasn't clear. Most of the time, as with prostrate screenings, one of the following is true: - There is no expert consensus (on treatments, etiology, etc.). - Consensus exists, but the level of consensus is overstated. - There is broad consensus, but with relatively low confidence. A generalized public health threat like a pandemic doesn't change the epistemological calculus, nor the sociology of knowledge.
    - jacquesm 5y ago
      There is a fourth option: - there a consensus, it is broad and it has good experimental support and high confidence
- rscho 5y ago
  Your GP may try to lay out the state of research before you, but that does not change the fact that he/she will do one of those things: 1. Treat you according to guidelines/expert opinion, should you choose to. 2. Give you the opportunity to participate in a relevant study. But _never ever_ will your GP formulate an experimental treatment plan based on his own research. Not GP's job, and big risk of liability. Your GP has at best a cursory view of the research happening in a certain field. COVID is different due to several good and bad reasons: hugely politicized, confusing messages from experts, and most of all novelty. 'Lack of medical consensus' usually means 'several treatments backed by experts are available', not 'do what you like'. COVID is too new for the dust around expert debates to have settled. That's what makes it different.
  - jacquesm 5y ago
    More likely: your GP will pass you on to the resident expert at the nearest hospital of your choice.
blindmute 5y ago
As someone who is actually an expert in a subfield of medicine, not only do doctors not understand it at all, but the experts making guidelines often recommend the exact opposite of what is the best treatment for the patient, usually for liability reasons, but occasionally for true lack of understanding. The quality of research is so poor that often times there are more flawed studies showing X than proper studies which would show ~X. Other times there is no research on the matter at all, and the conventional wisdom (even among experts!) is incorrect. For fields where practice and theory diverge, like an experimental surgery with sparse research, the surgeons know that the guidelines are useless. What you yourself _have_ to realize is that the expert researchers who are in a position to create and inform guidelines are themselves not some kind of 'super-experts' who know everything better than normal experts. They're just researchers in a political position in a bureaucracy. I don't exaggerate when I say that an average HN reader could write better guidelines for patient outcomes in my subfield, with no prior training, simply by reading a few studies and observing a practice. Now, does any of this apply to internal medicine, specifically covid? Probably not. But when the data is anything other than crystal clear, you should not assume the expert guidelines are anything more than some guesses by guys in a room. The consequence of knowing this is that all guidelines and consensus is suspect until you see the data yourself, like TFA outlines. In this case it seems true that Ivermectin doesn't cure covid. But with my experience, the exact opposite result could have been shown and I would not be surprised whatsoever; experts saying it doesn't work does not constitute in my eyes anything other than noise until the data is clear.
- rscho 5y ago
  I'm not saying that experts are doing a stellar job or that they're special in any way, far from it. I have slaved enough under 'true experts' to know the extent of the catastrophe. We even had a truly spectacular example of failure at the start of the pandemic with the 'intubate early and withhold steroids' expert opinion. This particular one, I won't ever forget. What I'm saying is only that: - on average, experts do spectacularly better than Joe Schmoe. Including science degree holders thinking they understand medical studies while missing the whole political and social context of the field. - it's not the job of field practitioners to prescribe experimental therapies, and those people cannot do better than follow expert opinion and guidelines. But I actually agree with you on all points, especially regarding flawed research and understanding. Which makes people basing opinions solely on papers without any practical knowledge of the field wrong all the more often. Does it make it a good idea for people to go and try on their own the thing they read about in the last issue of 'covid today'? I don't think so.

by rscho 5y ago

deadpannini 5y ago

The dissonance here is intense. On the one hand, "practicing MDs ... understand almost nothing when reading medical studies", but also I don't "know better than the average doc [just] because [I've] read 10 studies." Is the idea that doctors deserve deference because they are conditioned to submit to guidelines formulated by experts? There is a lack of consensus on many long-standing, common medical questions, much less novel diseases. When I ask my GP about prostate screening, for example, he lays out the current state of research as best he can, but delegates the decision to me. Why is the COVID situation different, given that experts are still fighting through a figurative fog of war?

nickspacek 5y ago
Great point. Perhaps the idea is that neither you nor the avg doc can conclude much from reading 10 studies. Therefore, defer to the experts? As to why the COVID situation is different, presumably because prostate cancer screening affects only you (and perhaps your family in the case of a true/false positive/negative diagnosis), and COVID represents a threat to everyone around you, and everyone around them, etc.
- deadpannini 5y ago
  I wasn't clear. Most of the time, as with prostrate screenings, one of the following is true: - There is no expert consensus (on treatments, etiology, etc.). - Consensus exists, but the level of consensus is overstated. - There is broad consensus, but with relatively low confidence. A generalized public health threat like a pandemic doesn't change the epistemological calculus, nor the sociology of knowledge.
  - jacquesm 5y ago
    There is a fourth option: - there a consensus, it is broad and it has good experimental support and high confidence
rscho 5y ago
Your GP may try to lay out the state of research before you, but that does not change the fact that he/she will do one of those things: 1. Treat you according to guidelines/expert opinion, should you choose to. 2. Give you the opportunity to participate in a relevant study. But _never ever_ will your GP formulate an experimental treatment plan based on his own research. Not GP's job, and big risk of liability. Your GP has at best a cursory view of the research happening in a certain field. COVID is different due to several good and bad reasons: hugely politicized, confusing messages from experts, and most of all novelty. 'Lack of medical consensus' usually means 'several treatments backed by experts are available', not 'do what you like'. COVID is too new for the dust around expert debates to have settled. That's what makes it different.
- jacquesm 5y ago
  More likely: your GP will pass you on to the resident expert at the nearest hospital of your choice.

blindmute 5y ago

As someone who is actually an expert in a subfield of medicine, not only do doctors not understand it at all, but the experts making guidelines often recommend the exact opposite of what is the best treatment for the patient, usually for liability reasons, but occasionally for true lack of understanding. The quality of research is so poor that often times there are more flawed studies showing X than proper studies which would show ~X. Other times there is no research on the matter at all, and the conventional wisdom (even among experts!) is incorrect. For fields where practice and theory diverge, like an experimental surgery with sparse research, the surgeons know that the guidelines are useless. What you yourself _have_ to realize is that the expert researchers who are in a position to create and inform guidelines are themselves not some kind of 'super-experts' who know everything better than normal experts. They're just researchers in a political position in a bureaucracy. I don't exaggerate when I say that an average HN reader could write better guidelines for patient outcomes in my subfield, with no prior training, simply by reading a few studies and observing a practice. Now, does any of this apply to internal medicine, specifically covid? Probably not. But when the data is anything other than crystal clear, you should not assume the expert guidelines are anything more than some guesses by guys in a room. The consequence of knowing this is that all guidelines and consensus is suspect until you see the data yourself, like TFA outlines. In this case it seems true that Ivermectin doesn't cure covid. But with my experience, the exact opposite result could have been shown and I would not be surprised whatsoever; experts saying it doesn't work does not constitute in my eyes anything other than noise until the data is clear.

rscho 5y ago
I'm not saying that experts are doing a stellar job or that they're special in any way, far from it. I have slaved enough under 'true experts' to know the extent of the catastrophe. We even had a truly spectacular example of failure at the start of the pandemic with the 'intubate early and withhold steroids' expert opinion. This particular one, I won't ever forget. What I'm saying is only that: - on average, experts do spectacularly better than Joe Schmoe. Including science degree holders thinking they understand medical studies while missing the whole political and social context of the field. - it's not the job of field practitioners to prescribe experimental therapies, and those people cannot do better than follow expert opinion and guidelines. But I actually agree with you on all points, especially regarding flawed research and understanding. Which makes people basing opinions solely on papers without any practical knowledge of the field wrong all the more often. Does it make it a good idea for people to go and try on their own the thing they read about in the last issue of 'covid today'? I don't think so.