A preference-based item response theory model to measure health: concept and mathematics of the multi-attribute preference response model

Table 2 Marginal distribution of patients’ own classification of their health status based on the five attributes, each with three levels, of the EQ-5D-3 L instrument (n = 163)

EQ-5D-3 L attributes and levels	Radboud^a(n = 75)	MST (n = 35)	UMCG (n = 53)
Mobility
No problems (1)	45 (60.0)	20 (57.1)	29 (54.7)
Some problems (2)	30 (40.0)	15 (42.9)	18 (34.0)
Confined to bed (3)			6 (11.3)
Self-care
No problems (1)	63 (84.0)	31 (88.6)	40 (75.5)
Some problems (2)	12 (16.0)	4 (11.4)	10 (18.9)
Unable to (3)			3 (5.7)
Usual activities
No problems (1)	38 (50.7)	12 (34.3)	26 (49.1)
Some problems (2)	34 (45.3)	21 (60.0)	24 (45.3)
Unable to (3)	3 (4.0)	2 (5.7)	3 (5.7)
Pain/Discomfort
No (1)	34 (45.3)	13 (37.1)	18 (34.0)
Moderate (2)	36 (48.0)	21 (60.0)	32 (60.4)
Extreme (3)	5 (6.7)	1 (2.9)	3 (5.7)
Depression/Anxiety
Not (1)	59 (78.7)	25 (71.4)	36 (67.9)
Moderately (2)	16 (21.3)	10 (28.6)	12 (22.6)
Extremely (3)			5 (9.4)

^aRadboud Radboud University Nijmegen Medical Center, MST hospital Medisch Spectrum Twente, UMCG University Medical Center Groningen

ISSN: 1471-2288