Assessing outcomes for cost-utility analysis in mental health interventions: mapping mental health specific outcome measure GHQ-12 onto EQ-5D-3L

Table 3 Characteristics of the individuals (Total sample, n = 32548^a)

	Estimation sample	Validation sample	p-value^b	Effect size^c
	Model building n = 17,101	Capacity checking n = 15,447	p-value^b	Effect size^c
Sex, n (%)
Men	7874 (46.0)	7009 (45.4)	0.226
Women	9227 (54.0)	8438 (54.6)
Self- rated health, n (%)
Very good	3048 (18.3)	2693 18.0)	0.932
Good	8191 (49.2)	7401 (49.6)
Neither nor	4354 (26.2)	3901 (26.1)
Bad	893 (5.4)	782 (5.2)
Very bad	159 (1.0)	147 (1.0)
Age, Mean (SD)	55.44 (19.44)	55.48 (18.52)	0.871
EQ-5D-UK, Mean (SD)	0.80 (0.22)	0.80 (0.22)	0.985	0.00
EQ-5D-SW, Mean (SD)	0.90 (0.09)	0.90 (0.09)	0.184	−0.01
GHQ-12, Mean (SD)	1.14 (2.46)	1.05 (2.38)	0.001	0.04

^aSelf-rated health 3.0 % missing, EQ-5D-UK 8.0 % missing, EQ-5D-SW 8.0 % missing, GHQ-12 3.1 % missing
^bPearson Chi-square test for categorical variables and independent samples t-test for quantitative variables
^cCohen’s effect size measure

ISSN: 1477-7525