3 ms·

Honestly, Modern Medicine doesn't provide good evidence or reasoning for people around risks like surgery. From the systemic point of view, frequentist statisti

Honestly, Modern Medicine doesn't provide good evidence or reasoning for people around risks like surgery. From the systemic point of view, frequentist statistics permit doctors and policy-writers to consider risks, but an individual doesn't get to re-play the experiment 1000s of times to get results that represent the statistics. You don't get to try again if you get a complication or die on the table but your doctor and the system do. The very human desire to regain "health" and comfort combined with a not unreasonable pitch from a doctor tend to override that rationale hesitance. Risking death or further disability to be freed of pain and discomfort is an oddly acceptable risk.

whatshisface 4y ago
I agree with you overall but going after statistics as the problem is a flawed way to make that point. If you commute to work on the road you're taking accumulated risks equal to many surgeries, and if you aren't making those decisions based on statistics you're making them based on what people around you consider normal. 1% risk that everybody takes is just as dangerous as one only taken by yourself.
- blamestross 4y ago
  The statistics don't actually tell you YOUR risk. They tell if the risk a population has on average collectively. Your personal risks could be much higher or lower, and quantifying them is basically impossible. It only makes sense to use the "average" risk level for reasoning if the sampled population is particularly representative of you, which is vanishingly unlikely due to the "curse of dimensionality". It makes a lot of sense to decide public policy of how driving should be managed based on those statistics, but is hilariously inappropriate to base your decision to get in a car and drive based on them.
  - whatshisface 4y ago
    >It makes a lot of sense to decide public policy of how driving should be managed based on those statistics, but is hilariously inappropriate to base your decision to get in a car and drive based on them. Driving and surgery both involve a lot of risks you can do nothing about, making statistics (selected for your demographic, of course) the most you can know. Statistics are not a good way to predict whether one's self will buy a red shirt or a green shirt (when you can know exactly what you will do because you are deciding), but you have no say in whether or not your anesthesiologist will guess the right dose or whether your surgeon will put the sutures in exactly the right positions to stop a post-op bleed. The same goes for most accidents, which are called "accidents" because culpable negligence is actually pretty rare.
CogitoCogito 4y ago
I'm not really sure I understand your point. Isn't this true of all statistical findings? Sure for some things you can repeat a similar action, but you can never go back and change that you took a specific action. Besides health research is done on a population level. The goal is improvement of healthcare as a whole. It sucks that some people end up on the wrong end of the stick, but that's just a fact of healthcare and life.
- blamestross 4y ago
  > but that's just a fact of healthcare and life. It's a problem of different goals. In the variants of "Maximize net good/minimize net harm" frequentist statistics are appropriate to reason with. From a individual "maximize my personal quality of life" perspective, I am likely member to many intersectional groups and complexities that render that statistical reasoning meaningless. You can only generalize the results of a study to those represented by it's sampling and the cold hard reality is that a nontrivial amount of people are not represented. Public health reasoning is entirely self consistent and rational given it's goals (minus perversions due to capitalism and rent seeking). The problem is that it is actively hostile to most people who are disabled, a minority, or dealing with any long-tail health problem.
woliveirajr 4y ago
After you suffer some years, dealing with enought pain that you can't sleep, discomfort of having pills that will take out all your basic common sense (to the point that you get into crazy risky situations you can't even explain), you'll prefer to die - or, even better, to take a chance on a surgery where the worst outcome is already your daily life, and the best one is getting your life again.
- blamestross 4y ago
  Yep. I live there too. It led to the above insight. Some successes (Heart surgery helped with chronic migraine pain). I'm currently turning off large swaths of my immune system in hopes that Celiac, Crohn's and Psoriasis get better.

whatshisface 4y ago

I agree with you overall but going after statistics as the problem is a flawed way to make that point. If you commute to work on the road you're taking accumulated risks equal to many surgeries, and if you aren't making those decisions based on statistics you're making them based on what people around you consider normal. 1% risk that everybody takes is just as dangerous as one only taken by yourself.

blamestross 4y ago
The statistics don't actually tell you YOUR risk. They tell if the risk a population has on average collectively. Your personal risks could be much higher or lower, and quantifying them is basically impossible. It only makes sense to use the "average" risk level for reasoning if the sampled population is particularly representative of you, which is vanishingly unlikely due to the "curse of dimensionality". It makes a lot of sense to decide public policy of how driving should be managed based on those statistics, but is hilariously inappropriate to base your decision to get in a car and drive based on them.
- whatshisface 4y ago
  >It makes a lot of sense to decide public policy of how driving should be managed based on those statistics, but is hilariously inappropriate to base your decision to get in a car and drive based on them. Driving and surgery both involve a lot of risks you can do nothing about, making statistics (selected for your demographic, of course) the most you can know. Statistics are not a good way to predict whether one's self will buy a red shirt or a green shirt (when you can know exactly what you will do because you are deciding), but you have no say in whether or not your anesthesiologist will guess the right dose or whether your surgeon will put the sutures in exactly the right positions to stop a post-op bleed. The same goes for most accidents, which are called "accidents" because culpable negligence is actually pretty rare.

CogitoCogito 4y ago

I'm not really sure I understand your point. Isn't this true of all statistical findings? Sure for some things you can repeat a similar action, but you can never go back and change that you took a specific action. Besides health research is done on a population level. The goal is improvement of healthcare as a whole. It sucks that some people end up on the wrong end of the stick, but that's just a fact of healthcare and life.

blamestross 4y ago
> but that's just a fact of healthcare and life. It's a problem of different goals. In the variants of "Maximize net good/minimize net harm" frequentist statistics are appropriate to reason with. From a individual "maximize my personal quality of life" perspective, I am likely member to many intersectional groups and complexities that render that statistical reasoning meaningless. You can only generalize the results of a study to those represented by it's sampling and the cold hard reality is that a nontrivial amount of people are not represented. Public health reasoning is entirely self consistent and rational given it's goals (minus perversions due to capitalism and rent seeking). The problem is that it is actively hostile to most people who are disabled, a minority, or dealing with any long-tail health problem.

woliveirajr 4y ago

After you suffer some years, dealing with enought pain that you can't sleep, discomfort of having pills that will take out all your basic common sense (to the point that you get into crazy risky situations you can't even explain), you'll prefer to die - or, even better, to take a chance on a surgery where the worst outcome is already your daily life, and the best one is getting your life again.

blamestross 4y ago
Yep. I live there too. It led to the above insight. Some successes (Heart surgery helped with chronic migraine pain). I'm currently turning off large swaths of my immune system in hopes that Celiac, Crohn's and Psoriasis get better.