Back

Inter-diagnostician Agreement and Inter-subject Variability for Binary Diagnostic Decisions in Psychiatry

Aiken, R. C.

2020-08-04 psychiatry and clinical psychology
10.1101/2020.07.31.20166074 medRxiv
Show abstract

SO_SCPLOWUMMARYC_SCPLOWAmong the many challenges in arriving at a valid psychiatric diagnosis are the variations in biologic, genetic/ epigenetic, psychological, cognitive, and behavioral attributes of the subject population: inter-subject variability. Study of nosologic classification in psychiatry has centered primarily on inter-diagnostician variability, reliability, while neglecting inter-subject variation. We present an analysis of the diagnostic process that considers variability not only in the diagnosticians but also the subjects. The conditions for which the percent agreement statistic is the preferred measure of agreement is detailed for both a general case as well as one for which the inter-subject variation is explicitly considered. Specifics are given as to the non-intuitive fact that the smaller the inter-subject variability, the poorer can be the agreement, suggesting that an agreement statistic alone may be misleading depending on the nature of the subject population. Existing agreement statistics do not measure inter-diagnostician reliability but the agreement between two diagnosticians influenced by the inter-subject variation.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.