Abstract
A high frequency of "false positive" conclusions is often the outcome when several laboratory test results are interpreted by comparing each single test result with the corresponding reference interval, i.e., by multiple univariate comparisons. The adequate solution to this problem is the application of a multivariate reference region for the test profile. This region is a straightforward extension of the univariate reference interval to the multidimensional situation.

This publication has 0 references indexed in Scilit: