Table 3.
Association between convenience stores within the 800 m network buffer near school and students’ obesity in two districts
| Continuous variable (ten additional convenience stores) | Categorical variablea (more convenience stores vs less convenience stores) | |||||
|---|---|---|---|---|---|---|
| ORb | 95% CI | P value | ORc | 95% CI | P value | |
| Model 1 | 1.06 | (0.97, 1.15) | 0.178 | 1.17 | (0.89, 1.54) | 0.259 |
| Model 2 | 1.06 | (0.97, 1.15) | 0.194 | 1.16 | (0.88, 1.54) | 0.291 |
| Model 3 | 1.09 | (1.01, 1.18) | 0.030 | 1.34 | (1.02, 1.76) | 0.035 |
| Model 4 | 1.10 | (1.01, 1.19) | 0.029 | 1.37 | (1.04, 1.81) | 0.026 |
| Model 5 | 1.13 | (1.03, 1.24) | 0.011 | 1.49 | (1.09, 2.03) | 0.013 |
aCategorical variable was divided by 24: the median number of convenience stores
bOR for continuous variable was the odds ratio of ten additional convenience stores
cOR for categorical variable was more than or equal to the median number of 24 versus less than 24 (reference)
Model 1: null model
Model 2: adjusted for districts
Model 3: Model 2 + father’s obesity, mother’s obesity, family type, father’s education, and mother’s education
Model 4: Model 3 + age, gender
Model 5: Model 4 + diet score, meeting the recommendation of MVPA, and meeting the recommendation of screen time