Table 4.

Results of linear regression analysis between predicted and observed BMI 1 year after surgery for prediction models in ALL group

Model	B	95% CI	p-value	R	R2	adjusted R2	SE	RMSE
Baltasar_1	1.22	1.05-1.39	<0.0001	0.46	0.21	0.21	5.13	5.13
Baltasar_2	1.01	0.87-1.15	<0.0001	0.46	0.21	0.21	5.13	5.13
Baltasar_3	0.94	0.81-1.07	<0.0001	0.46	0.21	0.21	5.13	5.13
Baltasar_4	0.94	0.81-1.07	<0.0001	0.46	0.21	0.21	5.13	5.13
Wood	0.58	0.50-0.66	<0.0001	0.47	0.22	0.22	5.11	5.10
Wise	0.29	0.25-0.33	<0.0001	0.43	0.18	0.18	5.21	5.20
Goulard	0.62	0.54-0.70	<0.0001	0.47	0.22	0.22	5.09	5.08
Seyssel	0.69	0.60-0.78	<0.0001	0.49	0.24	0.24	5.03	5.03
Cottam 1	0.56	0.49-0.64	<0.0001	0.46	0.21	0.21	5.13	5.12
Cottam 2	0.51	0.44-0.58	<0.0001	0.46	0.21	0.21	5.11	5.11
Janik	0.51	0.45-0.58	<0.0001	0.49	0.24	0.24	5.03	5.02
Velázquez-Fernández	0.67	0.58-0.76	<0.0001	0.47	0.22	0.22	5.10	5.09

p-value refers to linear regression coefficient

Embolden p-values indicate statistically significant result

Abbreviations: B regression coefficient, 95% CI 95% confidence interval, R Pearson’s correlation coefficient, SE standard error of the estimate, RMSE root mean square error