Table 2.

Assessing model fit using the Akaike information criteria and the relative likelihood that progressive models are similar to or worse than the previous model using estimates linking nasoalveolar molding status to surgical outcomes

		Logistic Regression		Multilevel regression adjusting for doctor-level grouping		Cross-classified multi-level logistic models shown
		Est.	p-value	Est.	p-value	Est.	p-value
UCLP Minimum severity cleft	AIC RL	1034	<0.001	891 <0.001	<0.001	873 <0.001	<0.001
Best anticipated surgical outcome	AIC RL	1246	<0.001	1101 <0.001	<0.001	1088 0.002	<0.001
No revision necessary	AIC RL	1459	<0.001	886 <0.001	<0.001	878 0.011	<0.001
BCLP Minimum severity cleft	AIC RL	1040	<0.001	990 <0.001	<0.001	806 <0.001	<0.001
Best anticipated surgical outcome	AIC RL	1178	<0.001	1112 <0.001	<0.001	971 <0.001	<0.001
No revision necessary	AIC RL	1861	<0.001	1348 <0.001	<0.001	1255 <0.001	<0.001

Note: Est. is the estimated value; UCLP is unilateral cleft lip and palate. BCLP is bilateral cleft lip and palate. AIC is the Akaike’s information criterion