Table 5.

Mapping equations from FIQR score to multi-attribute utility instruments

	AQoL-8D		15D		EQ-5D-5 L		SF-6D
	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE
PANEL A
OLS
Constant	.778	.027***	.997	.039***	1.479	.024***	.794	.022***
FIQR/100	−.564	.046***	−.420	.030***			−.383	.035***
Age			−.001	.001*
Squared-FIQR					−.898	.056***
GLM
Constant	−.144	.060*	.068	.058	.425	.025***	−.179	.039***
FIQR/100	-1.175	.101***	−.600	.046***			−.670	.061***
Age			−.002	.001*
Squared-FIQR					−.824	.058***
PANEL B
OLS
Constant	.787	.031***	.950	.021***	1.496	.027***	.786	.025***
Age
FIQR_D1			−.138	.036***			−.137	.043**
FIQR_D2	−.144	.043**					−.089	.034**
FIQR_D3	−.415	.065***	−.305	.042***			−.149	.049**
Squared FIQR_D1					−.335	.081***
Squared FIQR_D2					−.167	.064**
Squared FIQR_D3					−.415	.084***
GLM
Constant	−.132	.068*	.078	.059	0.438	.028***	−.192	.044***
Age			−.002	.001*
FIQR_D1			−.156	.058**			−.241	.075**
FIQR_D2	−.318	.096**	−.096	.048*			−.153	.059**
FIQR_D3	−.840	.147***	−.359	.070***			−.264	.085**
Squared FIQR_D1					−.304	.081***
Squared FIQ_D2					−.167	.065**
Squared FIQR_D3					−.368	.086***

Coeff. and SE: Beta and Standard error of significant variables (*p < 0.05, **p < 0.01; ***p < 0.001) in the equations

Panel A: FIQR, Squared-FIQR, Age

Panel B: FIQR function domain (FIQR_D1); FIQR overall impact domain (FIQR_D2); FIQR symptom domain (FIQR_D2); Squared FIQR function domain; Squared FIQR overall impact domain; Squared FIQR symptom domain; Age

Gamma GLM: Generalized linear model Gamma log-link. EQ-5D was pre-transformed as EQ-5D + 0.66 in order to calculate the Gamma GLM with all scores (including negative values). It was post-transformed as EQ-5D – 0.66

Coefficients with best fitting for equations are in bold