3 ms·
It's has nothing to do with the Bayesian Base Rate fallacy. In this case we ABSOLUTELY KNOW the base rate is lower, since it was a preventative scan. The falla
by drcode 3y ago
It's has nothing to do with the Bayesian Base Rate fallacy.
In this case we ABSOLUTELY KNOW the base rate is lower, since it was a preventative scan. The fallacy only happens if you then subsequently fail to adjust for the differing base rate.
If doctors later in the diagnostic process fail to take the differing base rate into account, that is not the fault of the preventative test, that is an error in determining what the correct follow up care is.
- Calavar 3y ago> If doctors later in the diagnostic process fail to take the differing base rate into account, that is not the fault of the preventative test, that is an error in determining what the correct follow up care is. In today's medical malpractice climate, many physicians intentionally ignore the differing base rate.
- drcode 3y ago100% correct But now we're no longer having a conversation about medicine we're having a conversation about our broken legal system and our broken regulatory frameworks
- tptacek 3y agoSounds like it does in fact have a bunch to do with the base rate fallacy.
- drcode 3y agoI suppose that's fair way to look at it, but people are being left with the impression that the problem is inherent in the tests, when the problem is entirely separate from the tests themselves.
- tptacek 3y agoI suppose it's a fair way to look at it, too! It's good when we can reach a shared understanding of what's happening; so often, the dynamics of message boards work against that.
- lozenge 3y agoMost people aren't going to be happy with "the scan shows something unusual that would normally be concerning, but you say you feel fine so I'll trust you instead of the scan".
- n8henrie 3y agoWhat is the base rate of a not-very-suspicious-but-maybe-a-little-bit finding on imaging of an asymptomatic person in the general not-sick population? I think pretending that medicine is equipped to deal with this in a healthy way is stretching the limits of well established medical science, and beyond what I would expect even a well intentioned and astute physician to know how to approach (or at least have been trained to do so expertly). Not that you're wrong, you're not, strictly speaking. Relevant piece: https://dynomight.net/diagnostics/ https://dynomight.net/diagnostics/