OPTIMAL CUT-POINTS WHEN SCREENING FOR MORE THAN ONE DISEASE STATE - AN EXAMPLE FROM THE CANADIAN STUDY OF HEALTH AND AGING

A method is described for choosing optimal cut points in a screening t est with a continuous score in order to divide people into any number of disease categories. The cut points are chosen to minimize the overa ll expected loss, given a matrix of losses incurred by misclassificati ons. It is shown that the optimal cut-points depend on the column diff erences of the loss matrix, The effect of changing the loss matrix is examined. For the error counting matrix, where misclassifications are given equal weight, it is shown that the relationship between the mini mal loss cut-point and the receiver operating characteristic (ROC) cur ve for two disease categories extends to the general case. The results of a study in which elderly people were screened for cognitive impair ment and dementia are used to illustrate the method. Copyright (C) 199 6 Elsevier Science Inc.