Skip to main content

Table 2 Individual biomarker evaluation including descriptive statistics, unadjusted p-values, and logistic regression model outcomes (Standard, Bayesian with flat prior, and Bayes with horseshoe prior), including age and gender (except univariate age and gender models)

From: Comparing conventional and Bayesian workflows for clinical outcome prediction modelling with an exemplar cohort study of severe COVID-19 infection incorporating clinical biomarker test results

  1. *Biomarkers not included in subsequent models due to small sample size, and recorded only in ICU (PCT)
  2. The True and False columns describe the number/percentage of severe outcomes for cases where the particular biomarker or demographic reading is true or false. For example, there were 257 patients who were women who had a severe outcome, and conversely there were 333 patients who were not women (i.e. men) who had a severe outcome. Regressions were fit using all associated dummy variables for a given biomarker (e.g. normal, mild, moderate, severe) and using only complete cases of training data, i.e. not using a variable for ‘Test not taken’. Categorical variables use a reading of ‘Normal’ as a reference in the fitted model, except ‘Male’ used as the reference category for gender
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us