Assessing Asthma control in UK primary care: Use of routinely collected prospective observational consultation data to determine appropriateness of a variety of control assessment models

Table 6 Comparison of performance of multiple regression models

Control Model	Pseudo R-squared	Akaike's Information Criterion	Area Under ROC Curve
RCP-3 Q Model	0.1803	46,933	.7924 95% CI 0.7963, 0.7885
Jones Morbidity Index	0.1352	77,811	.7407 95% CI 0.7444, 0.7370
GINA Control	0.0977	61,868	.7135 95% CI 0.7180, 0.7090
GOAL Total Control	0.0843	51,660	.7028 95% CI 0.7079, 0.6969
GOAL Well Controlled	0.0845	65,481	.6970 95% CI 0.7015, 0.6925
PEFR	0.0615	16,602	.7085 95% CI 0.7183, 0.6985
Overuse of SAB	0.0833	82,299	.6942 95% CI 0.6983, 0.6901
BTS Treatment Step	0.0677	78,223	.6711 95% CI 0.6756, 0.6665
Linear Regression	R-squared	AIC
RCP-3 Q Expanded Score Model	0.3258	28,616	N/A

ISSN: 2731-4553