Table 4.

Results for regression and Bland Altman analysis for comparison of the criterion method and the fat mass estimation equations at birth and 3 months

Comparison	Regression analysis			Bland and Altman
	Slope	R2	p-value	SEE	Mean bias ± SD	95% limits of agreement	Pearson Correlation (r)	p-value*
Birth (n = 95)
Deierlein vs Criterion	−0.87	0.61	<0.0001	0.108	0.114 ± 0.109	−0.010 - 0.328	−0.17	0.099
Catalano vs Criterion	0.92	0.55	<0.0001	0.116	−0.012 ± 0.116	−0.240 - 0.215	−0.31	0.002
Lingwood vs Criterion	0.93	0.55	<0.0001	0.116	−0.045 ± 0.116	−0.272 - 0.183	−0.33	0.001
Aris vs Criterion	0.87	0.62	<0.0001	0.106	−0.034 ± 0.107	−0.245 - 0.176	−0.15	0.140
3 months (n = 63)
Deierlein vs Criterion	0.29	0.42	<0.0001	0.333	3.325 ± 0.784	1.789 - 4.862	0.77	<0.0001
Catalano vs Criterion	1.02	0.50	<0.0001	0.308	−0.271 ± 0.306	−0.871 - 0.328	−0.47	<0.0001
Lingwood vs Criterion	1.24	0.55	<0.0001	0.294	−0.286 ± 0.298	−0.871 - 0.299	−0.63	<0.0001
Aris vs Criterion	1.15	0.52	<0.0001	0.303	−0.230 ± 0.303	−0.824 - 0.363	−0.57	<0.0001

* Significance for the correlation of the strength for the relationship between the mean of the criterion and each equation correlated to the difference between the equation estimated infant fat mass and the criterion measured fat mass. A non-significant correlation suggests no bias in the technique across the range of fatness