Evaluation of the performance of algorithms mapping EORTC QLQ-C30 onto the EQ-5D index in a metastatic colorectal cancer cost-effectiveness model

Table 8 Mean, standard deviation, minimum and maximum of utility values, RMSE and MAE for the predicted utilities for incomplete questionnaires (n = 120) with algorithms using domain scores for utility prediction (model 1, 4, 5 and 6)

	Observed utility NL	Model 1 RE functional scale scores*	Model 4 Ordered logit	Model 5 Beta regression	Model 6 Separate equations subgroup approach
Mean	0.760	0.764	0.756	0.764	0.767
St.Dev	0.232	0.177	0.222	0.222	0.178
Min	−0.086	0.108	−0.178	0.032	0.232
Max	1	0.982	0.971	0.959	0.985
RMSE	–	0.128	0.139	0.131	0.128
MAE	–	0.085	0.086	0.088	0.085
Spearman correlation	–	0.843	0.827	0.828	0.827
p-value	–	0.734	0.945	0.751	0.723

ISSN: 1477-7525