Table 3.

Multilevel regression of obesity among Brazilian adolescents and individual- and school-level variables, National School Health Survey, 2015.

Variables	Null Model	Model 1		Model 2
		PR (CI 95%)	p-Value	PR (CI 95%)	p-Value
School Level
Administrative dependence
Private				1.16 (1.04–1.30)	0.010
Geographic macro-region
Northeast				0.92 (0.78–1.10)	0.369
Southeast				1.10 (0.93–1.30)	0.265
South				1.22 (1.04–1.44)	0.014
Central-West				0.99 (0.84–1.16)	0.869
Individual Level
Gender
Female		0.66 (0.60–0.73)	<0.001	0.67 (0.61–0.73)	<0.001
Age
15–20 years		0.61 (0.55–0.68)	<0.001	0.61 (0.55–0.68)	<0.001
Dietary pattern
Higher nutritional risk		0.72 (0.65–0.80)	<0.001	0.72 (0.64–0.79)	<0.001
Food consumption while watching TV or studying
Regular (≥5 days)		0.88 (0.80–0.97)	0.009	0.89 (0.81–0.99)	0.024
Breakfast consumption
Irregular (<5 days)		1.32 (1.20–1.45)	<0.001	1.31 (1.19–1.44)	<0.001
Dining at fast-food restaurants
Yes		0.83 (0.75–0.91)	<0.001	0.82 (0.74–0.90)	<0.001
Body satisfaction
Indifferent		2.51 (2.20–2.86)	<0.001	2.47(2.17–2.82)	<0.001
Dissatisfied		3.42 (3.08–3.80)	<0.001	3.36 (3.03–3.75)	0.010
Number of residents in the household
<5 residents		1.26 (1.14–1.39)	<0.001	1.23 (1.11–1.36)	<0.001
Fixed effects
Intercept (CI 95%)	−2.244 (−2.304–(−2.185))	0.099 (0.087–0.112)		0.093(0.079–0.110)
Random Effects	Variance (SE)	Variance (SE)		Variance (SE)
Education level	0.092 (0.057–0.148)	0.042 (0.019–0.098)		0.033 (0.012–0.092)
Variation (%)		−54.3		−64.1
LR Test (Chi2; p-value)	30.32 (<0.001)	7.75 (p = 0.003)		5.84 (p = 0.007)