Replies to post #128879 on Biotech Values

[iwfal]

the statement "the incidence of disease is 1% generally in the population" is not the same as "a patient referred for a diagnostic test has the chance of having the disease in question equivalent to the general incidence rate in the population".

No - I agree that they are not the same. But in fact I said neither. I stated the condition very precisely twice. I said "1% of the tested population". NOT "1% of the general population". And I made no contention of any form about the incidence in the general population. I was very very exacting in this description precisely because this kind of conflation is what gets people in trouble in the fallacy - confusion between per-tested and overall-population

If your contention in the example were true, then almost no one would ever believe a diagnosis of serious disease

If it makes you feel more comfortable think of my example as the math that a statistician at the FDA has to work through for a new test being submitted to the FDA for approval. Should he approve it or not? (And this is far from an academic debate since it is precisely the issue behind the recent debate over PSA testing etc.)