Assessment of sugar-sweetened beverage consumption and weight change: a prospective cohort study

BMC Nutrition

Table 4 Linear regression analysis of weight change over 10 years according to baseline values of SSB, energy intake and physical activity

Independent variables	β-coefficients	Standard error	95% CI		P-value
Independent variables	β-coefficients	Standard error	Lower bound	Upper bound	P-value
Model 1
Constant	9.187	1.698	5.851	12.523	<0.001
Weight 2002 (kg/m²)	0.961	0.022	0.918	1.003	<0.001
SSB never	ref
SSB low vs never	0.152	0.600	−1.027	1.330	0.800
SSB medium vs never	−0.173	0.592	−1.337	0.990	0.770
SSB high vs never	4.349	1.210	1.971	6.726	<0.001
Model 2
Constant	9.728	2.618	4.585	14.871	<0.001
Weight 2002 (kg/m²)	0.960	0.022	0.918	1.003	<0.001
Physical activity level 2002^a	−0.297	1.023	−2.306	1.713	0.772
SSB never	ref
SSB low vs never	0.157	0.602	−1.025	1.340	0.794
SSB medium vs never	−0.161	0.597	−1.333	1.012	0.788
SSB high vs never	4.338	1.246	1.891	6.786	0.001
Energy 2002 (unit 40 kcal)	0.001	0.014	−0.026	0.028	0.960
Model 3
Constant	10.196	2.638	5.014	15.337	<0.001
Weight 2002 (kg/m²)	0.957	0.022	0.915	1.000	<0.001
Physical activity level 2002^a	−0.486	1.030	−2.510	1.538	0.637
Log SSB 2002	0.190	0.198	−0.198	0.578	0.335
Energy 2002 (unit 40 kcal)	0.010	0.014	−0.017	0.037	0.485

SSB sugar sweetened beverages
Each model is adjusted for all model-specific variables displayed in the Table plus age and gender
Categorization for SSB consumption in 2002: never SSB; low SSB consumption (less than 200 ml/d); moderate SSB consumption (between 200 and less than 700 ml/d); high SSB consumption (700 ml/d and more)
^aPhysical activity level was estimated using the Flemish Physical Activity Questionnaire [9]

ISSN: 2055-0928