In the attached paper, I argue, the prescribed Bayesian method is not employed for statistical inference during diagnosis in practice as it leads to diagnostic errors. This method has been prescribed due to its rationality based on its coherence and not due to its inferential accuracy (I quote a well known Bayesian statistician).
Like Feinstein in 1977, I cannot find a single published diagnostic exercise in a real patient such as CPC etc. in which this method is employed. For example, in none of these exercises is the Bayesian notion of a prior probability as prior evidence giving rise to a degree of belief employed.
The method employed in practice, I point out, is the confidence frequentist method, which is the other major method of statistical inference. In this method, for example, a prior probability does not represent prior evidence which prevents diagnostic errors in case of diseases with atypical presentations.
I have communicated with several statisticians and philosophers of science in order to understand this discrepancy between theory and practice without any success as they seem to lack knowledge about the process of diagnosis.
This is a problem that exists in a field, diagnosis, with which we as practicing physicians are very familiar. I request the worldwide members of SIDM to review and comment on this paper to find weaknesses in its reasoning if they are present, so that we can learn more about the diagnostic process which may minimize errors.
Bimal
Bimal Jain MD
Northshore Medical Center
Salem MA 01970
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.