I have a score and a dichotomous criterion for n observations. Using a ROC-Analysis, I found some optimal cut-off, say c. Now, literature (e.g., author of scale with which the score was determined) says, optimal cut-off should be c+2. Now, I want to test, whether sensitivity, specifity, etc. differ significantly if using c vs. c+2 (actually, the reviewer requests that). I did compare CIs, but is there a specific test available?
Online recherche always ends up in comparing 2 ROC-curves (paired or not). Wrong search terms? Or did I overlook an embarrassingly simple solution?
