Skip to main content

Table 3 Multivariable penalized logistic regression using Firth’s correction

From: A retrospective pooled analysis of response patterns and risk factors in recurrent malignant glioma patients receiving a nitrosourea-based chemotherapy

Response variable = Disease control (CR + PR + SD)
Effect Odds ratio Lower 95% Confidence Limit Upper 95% Confidence Limit P-Value
Gender - Female vs Male 0.906 0.452 1.800 0.7806
Second Surgery 2.802 1.152 7.334 0.0306
Time from radiotherapy 1.075 1.019 1.147 0.0161
Number of TMZ cycles 0.925 0.838 1.013 0.1053
  1. Response variable is Disease control (CR + PR + SD). Odds ratios for time from radiotherapy and number of TMZ cycles show an improvement ratio (OR > 1) associated with an increase of 1 month in time from radiotherapy and the deterioration ratio (OR < 1) associated with an increase of 1 cycle in number of TMZ.