While that is true, the GP (and even specialists) use a different optimization criterion than yourself (proverbial you): You prefer to minimize false negatives

While that is true, the GP (and even specialists) use a different optimization criterion than yourself (proverbial you): You prefer to minimize false negatives with some exceptions about false positives, whereas they prefer to minimize overall error rate with some exceptions -- and that goes very deep into their training, to the point of being blind to this preference. E.g., while at med school, they are warned against Zebras[0] so much that while practicing they will often ignore anything with a 1-in-a-million-chance unless the false-positive rate is zero. It does minimize the overall error rate, but for the person who has the 1-in-a-million condition, this is no comfort -- and that person is often in a position to evaluate that condition better than the GP (if they are scientifically literate). I have been told by a senior doctor at one of the leading US hospitals to "cut the internet cord so that I don't confuse myself as I am not trained to understand the articles I find". Turned out said doctor did not understand the statistical inference criterion, and I did -- unsurprisingly, since I have the mathematical background, and they do not. [0] https://en.wikipedia.org/wiki/Zebra_%28medicine%29 https://en.wikipedia.org/wiki/Zebra_%28medicine%29