Problems in Conducting Multivariate Logistic Regression Analysis

Abstract

We read with great interest the original article “The Relationship Between Procalcitonin Level and Short Term Mortality in Emergency Department” by Cehreli et al. (1). A high initial level of procalcitonin (PCT) is a sensitive marker of bacterial infection, and the degree of PCT elevation can determine which patients are at greater risk of adverse outcomes. In this study, the authors wanted to analyse whether there was an association between PCT levels and seven-day mortality in patients whom PCT levels were measured in the emergency department. However, we think that there are some serious methodological mistakes.