A caution on the use of cut-points applied to screening instruments or diagnostic criteria.

Although cut-points or threshold scores are commonly used in screening instruments as well as in psychiatric diagnostic practice, much is still to be learned about their functional characteristics. This study demonstrates that, for a cut-off of y items on a screening instrument, there is great variability in the identification of caseness between different combinations of y items. The results suggest that the use of cut-points is perhaps not as reliable as is sometimes thought.