A New Procedure for Discriminating Between Two Patient Populations Using Multivariate Decision Limits: Application in the Detection and Exclusion of Alcoholism Based on Clinical Laboratory Findings

摘要: A simple, nonparametric discrimination procedure was developed and tested for use in discriminating between two populations, especially those cases where the properties of distributions lead to unsatisfactory results by classical procedures. The is based on transferring conventional concept a decision limit from one several variables, with combinations (in set theory terms: unions intersections) multidimensional intervals serving as discriminant regions (referred here simple regions). applied groups patients, alcoholics non-alcoholics. level efficiency 20% higher than that attainable procedures obtained using only six clinical chemical parameters. In addition, more flexible It enables solution optimization problems each three common criteria (sensitivity, specificity efficiency) no fundamental difficulties, it still possible select at will range(s) point order regions. For all parameters studied patient samples overlapped very markedly. If, example, sensitivity 100% specified, best discriminated about 53%, if required, 56%. analogy principle statistical test (the relationship detection exclusion alcoholism) also discussed.