4 ms·

> And honestly I am very pro-science but the “do no harm” bias in medicine makes me fundamentally question a lot of the recommendations. It's really not that -

> And honestly I am very pro-science but the “do no harm” bias in medicine makes me fundamentally question a lot of the recommendations. It's really not that - people have a fundamental desire to take action to mitigate tail risk even when that action poses more actual risk. You have to look at risk-reward tradeoff across the population. Treating the entire population for a disease very few of them have will almost certainly yield more (and potentially more harmful) realized side-effects than the actual primary effects being caused by the disease itself. Especially when patients aren't seeking treatment, so are most likely asymptomatic. Ditto testing low-risk populations for all sorts of conditions they mostly don't have. Even a test that is 99% accurate in a high-incidence population is going to be utterly ineffective when applied to the general population. Bayes theorem tells us this. The side effects up to and including death of investigating or treating these false positives adds up to a real net-negative situation. If it was even an eency weency bit better to treat in the US I would say they would overwhelmingly do so because they'd be sued into the ground otherwise. tl;dr: especially when it comes to treatments, action is not always better than inaction. Generally, the US medical system tends towards action.

landemva 3y ago
> Even a test that is 99% accurate in a high-incidence population is going to be utterly ineffective when applied to the general population. I read that sentence a few times. 99% accurate seems quite high. WSJ recently opined about the 1-in-5 cancer misdiagnosis rate. https://www.wsj.com/articles/are-you-sure-you-have-cancer-diagnosis-drug-treatment-trial-drug-383cb293 https://www.wsj.com/articles/are-you-sure-you-have-cancer-di... "Are You Sure You Have Cancer?"
- rovolo 3y ago
  The WSJ article is behind a paywall so I can't double check this, but you can have a 99% accurate test with a 20% misdiagnosis rate. The difference is that 99% is per-person, and 20% is per-positive-result. A rare condition will have more false-positives relative to true-positives than the test accuracy would indicate. It's a bit easier to understand if you think of the extreme case: if the disease is completely eliminated than all positives are false-positives.
- simiones 3y ago
  The point is that a test that has a guarantee like "out of 100 people who test positive, only one will turn out to be negative" (99% specificity) can have an arbitrarily high false-positive rate when applied to the whole population: if a condition affects 1 in 10k individuals, then out of 10k individuals that take that test, 100 will be positive, but only 1 will be a true positive. The rarer the condition, the more specificity you need in your test to have a low false positive rate.

landemva 3y ago

> Even a test that is 99% accurate in a high-incidence population is going to be utterly ineffective when applied to the general population. I read that sentence a few times. 99% accurate seems quite high. WSJ recently opined about the 1-in-5 cancer misdiagnosis rate. https://www.wsj.com/articles/are-you-sure-you-have-cancer-diagnosis-drug-treatment-trial-drug-383cb293 https://www.wsj.com/articles/are-you-sure-you-have-cancer-di... "Are You Sure You Have Cancer?"

rovolo 3y ago
The WSJ article is behind a paywall so I can't double check this, but you can have a 99% accurate test with a 20% misdiagnosis rate. The difference is that 99% is per-person, and 20% is per-positive-result. A rare condition will have more false-positives relative to true-positives than the test accuracy would indicate. It's a bit easier to understand if you think of the extreme case: if the disease is completely eliminated than all positives are false-positives.
simiones 3y ago
The point is that a test that has a guarantee like "out of 100 people who test positive, only one will turn out to be negative" (99% specificity) can have an arbitrarily high false-positive rate when applied to the whole population: if a condition affects 1 in 10k individuals, then out of 10k individuals that take that test, 100 will be positive, but only 1 will be a true positive. The rarer the condition, the more specificity you need in your test to have a low false positive rate.