Table 1.

Characteristics of the five survey datasets

Survey places	EuroQol version	Self-rating of EQ-5D profile and EQ-VAS	Valuation data	No. of health states in the TTO	Background variablesa	Year of survey	No. of respondents	Exclusion criteria for the valuation study	No. of respondents in the valuation data setb	Softwarec	Health state valuation method
Japan	EQ-5D-3L	Availabled	Available	17	Available	1998	621	Respondents completely missing TTO data; valued 1 or 2 states only; giving all states the same value; valued all states worse than dead	543	STATA	Conventional TTO
UK	EQ-5D-3L	Available	Available	42	Available	1993	3395	Missing in valuation data	2997	LIMDEP	Conventional TTO
Japan	EQ-5D-5L	Available	Available	86	Available	2014	1098	Three of the investigators (involving 72 respondents) did not follow the survey manual procedures	1026	STATA, SAS, and R	Lead time TTO
England	EQ-5D-5L	Available	Available	86	Available	2012–2013	996	Respondents who gave the same TTO value for all health states; giving health state 55555 a value no lower than the value for the mildest health state in their block	912	R and WinBugs	Lead time TTO and DCE
Spain	EQ-5D-5L	Available	N/A	86	Available	2012	1000	N/A	N/A	N/A	N/A

N/A not available, TTO time trade-off, DCE discrete choice experiments

^aThe background variables include age, gender, experience with serious illness by respondents themselves, their families, and others

^bRepresents the number of respondents after applying the exclusion criteria. For the Japanese EQ-5D-3L dataset, this study only has access to the sample that is included in the valuation study (N = 543)

^cThe software used for the original analysis in the valuation studies

^d‘Available’ and ‘N/A’ refer to the availability of data for this study