As my learned colleague, mentor (from afar) and friend, Len Calabrese said, Bayesian reasoning is the key. This seems to have been almost completely abandoned by most if any discussion in our and other fields. As someone who runs a very large immunology laboratory I have seen and had to deal with the gross misuse of testing , instead of, as meant, tests which stratify diagnosed patients to screening tests where pre-test probability is close to zero. Simple maths says if a patient has less than 1% of having the disease , that even a 99% specific test ill yield more false positives than true.

Recently I was shocked to see in an ACR supported publication, the call to investigate young women with an ANA of 1:320 for possible Sjögren’s syndrome. This is despite lack of symptoms. The yield will damage patients in the desire for early diagnosis, despite total lack of evidence its even possible in real life.

Happy to be debated on any of the above. As an overseas ACR member, it’s dismaying to see what I have spent 25 years teaching about ignored.

Daman