Table 3.

Associations of neighbourhood disadvantage and the food environment with overweight among children in the sub-district that the school is located in.

	Association with overweight among children in the sub-district
	β (95% confidence interval)
	Univariable	Multivariablea
Relative FFD400 b
All fast-food outlets	2.74*** (1.73, 3.76)	1.16** (0.46, 1.86)
Fast-food restaurants	7.71***c (4.82, 10.61)	1.54*** (0.76, 2.33)
Grillrooms & kebab shops	5.11*** (4.01, 6.22)	2.15*** (1.15, 3.15)
Take-away restaurants	10.57***c (6.21, 14.94)	3.69*c (0.67, 6.71)
Relative FFD1000 d
All fast-food outlets	1.09***c (0.55, 1.64)	0.46*c (0.10, 0.82)
Fast-food restaurants	1.15***c (0.66, 1.64)	0.45**c (0.11, 0.78)
Grillrooms & kebab shops	3.39***c (2.34, 4.45)	2.01***c (1.13, 2.89)
Take-away restaurants	−1.07***c (−1.67, −0.47)	−0.18° (−0.36, 0.005)
FFP e
All fast-food outlets	25.33**c (7.70, 42.96)	11.54°c (−0.17, 23.25)
Fast-food restaurants	33.38**c (12.58, 54.18)	13.91°c (−0.04, 27.86)
Grillrooms & kebab shops	−4.48*** (−5.60, −3.35)	−1.60** (−2.56, −0.63)
Take-away restaurants	20.64*c (3.11, 38.17)	−0.26 (−1.34, 0.82)

^***p < 0.01,

^**p < 0.05,

^*p < 0.1, ° p < 0.001.

^aThe multivariable regressions were adjusted for the neighbourhood disadvantage level around schools (based on the schools' disadvantage scores).

^bThe variable relative FFD₄₀₀ was transformed using the expression ln(X + 1). Relative FFD₄₀₀ refers to the number of fast-food outlets within 400 m (Euclidean distance) from a school as a proportion of the total amount of food retailers in this area.

^cThis effect was better estimated by including a quadratic term in the regression.

^dRelative FFD₁₀₀₀ refers the number of fast-food outlets within 1000 m (Euclidean distance) from a school as a proportion of the total amount of food retailers in this area.

^eThe variable FFP was transformed using the expression ln(X). FFP is the shortest Euclidean distance from the school to the nearest fast-food outlet.