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

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

		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

SD standard deviation, IQR interquartile range, N/A not applicable
^*Distributions are significantly different between training and validation sets by two-tailed Student’s t-test
^aDenotes feature of 3-month model
^bDenotes feature of 12-month model

ISSN: 1471-2407