Table 3.
Association between sugar-sweetened beverages consumption and the school environment among Brazilian adolescents included in the ERICA study, Brazil (2013–2014; n = 71,549).
| Variables | Sugar-sweetened beverages consumption (mL) | |||
|---|---|---|---|---|
| Public school | Private school | |||
| β (95% CI)1 | β (95% CI)2, a | β (95% CI)1 | β (95% CI)2, b | |
| Drinking fountains per 100 students | 1.5 (−20.2; 23) | −6.2 (−26.2; 14) | 2 (−7.3; 11) | 1.5 (−7.4; 10.3) |
| Soft drinks sale | 16 (−2; 33.4) | 17.5 (2; 33)* | 26.2 (−3.2; 56) | 55 (24; 86)** |
| Soft drink advertising | −9 (−55.3; 37) | 5.2 (−35.4; 46) | −16 (−51.4; 19.4) | −10 (−44.3; 21) |
| Street vendor of soft drink | −5 (−24.3; 14.2) | 8.8 (−8; 26) | −15.5 (−48; 17) | −11 (−42; 21) |
| Variance | 13,147 (11,756.2; 14,702) | 8,919 (7,889; 10,083) | 6,072 (4,661; 7,909.2) | 5,205 (3,916.2; 6,918) |
| Intraclass correlation coefficient | 0.08 (0.07; 0.09) | 0.06 (0.05; 0.07) | 0.4 (0.3; 0.6) | 0.04 (0.03; 0.05) |
| Akaike criteria | 765,332.5 | 700,726 | 207,450.5 | 190,126.2 |
CI, Confidence Interval; mL, milliliters.
1
Unadjusted multilevel linear regression model.
2
Adjusted multilevel linear regression model.
a
Adjusted for gender, race/ethnicity, age, socioeconomic score, school region, offer school meals, and capital.
b
Adjusted for gender, race/ethnicity, age, school region, and capital.
*
p < 0.05.
**
p = 0.001.
Values in bold are statistically significant (p < 0.05).