Table 6.

Multiple linear mixed models with random intercept and slope for the associations between BMI changes and the changes in the biomarker levels over time, adjusted for age and sex.

		BMI
		β coefficient (SE)	P-value	95% CI
Salivary CRP	Main effect	2.40 (0.48)	<0.001*	1.40 , 3.30
Salivary CRP ainteraction with time	CRP#time 2	−0.35 (1.03)	0.7	−2.30, 1.60
(salivary CRP at phase 1 is the reference)	CRP#time 3	3.50 (0.79)	<0.001*	1.90 , 5.07
Salivary Insulin	Main effect	2.10 (0.53)	<0.001*	1.10 , 3.10
Salivary insulin interaction with time	Insulin#time2	1.60 (0.95)	0.08	−0.19 , 3.50
(salivary insulin at phase 1 is the reference)	Insulin#time3	3.20 (0.96)	0.001*	1.30 , 5.10
Salivary Adiponectin	Main effect	−1.60 (0.6)	0.007*	−2.80, −0.40
Salivary Adiponectin interaction with time	Adiponectin#time2	−0.29 (1.15)	0.79	−2.50, 1.90
(salivary adiponectin at phase 1 is the reference)	Adiponectin#time3	−1.19 (1.27)	0.35	−3.60 , 1.30

SE, Standard Error; CI, 95% Confidence Interval; CRP, C-Reactive Protein; #, the estimate for the beta coefficient of the interaction term variable is the multiple of the main effect with the variable ‘time'; the categorical variables are comprised of visit 1, visit 2, and visit 3 that show the rate of BMI change over time for salivary C-Reactive Protein, insulin, and adiponectin.

^*, P-Value <0.05. ^aInteraction term variable is the multiple of the main effect with the variable ‘time'.