Skip to main content

Table 1 Comparison of continuous features for training (U.S.), first validation (Scandinavian) and second validation (Italian) datasets

From: How do we estimate survival? External validation of a tool for survival estimation in patients with metastatic bone disease—decision analysis and comparison of three international patient populations

   Training set n = 189 Scandinavian set n = 815 Italian set n = 287
Feature     % missing   % missing vs. Training set p vs. Scandinavian set p
Age at surgery (years)b Mean 62.4 66.3 0 63.1 0 0.54 <0.005*
SD 13.7 12.8   11.7    
Median 62.7 67   64    
IQR 54.4, 72.2 58, 76   56, 72    
Hemoglobin concentration (mg/dL) a,b Mean 11.5 11.5 0.6 11.5 10 0.83 0.90
SD 1.9 3.5   1.4    
Median 11.4 11.3   12    
IQR 10.1, 12.9 10.3, 12.6   11,13    
Absolute lymphocyte count (K/μL)a,b Mean 1.2 1.2 83.8 1.3 23 0.59 0.40
SD 1.3 0.74   0.50    
Median 1.0 1.2   1.5    
IQR 0.6, 1.5 0.8, 1.6   1.0, 2.0    
Senior surgeon’s estimate of survival (months)a,b Mean 10.3 N/A 100 11.2 87 0.56 N/A
SD 8.6    7.0    
Median 6.0    10    
IQR 4.0, 12.0    5, 20    
  1. SD standard deviation, IQR interquartile range, N/A not applicable
  2. *Distributions are significantly different between training and validation sets by two-tailed Student’s t-test
  3. aDenotes feature of 3-month model
  4. bDenotes feature of 12-month model