Clustering of unhealthy food around German schools and its influence on dietary behavior in school children: a pilot study

Table 3 Results of normal and lognormal multilevel regression Models 1 - 6 investigating the effect of the food retail index on food intake and BMI adjusted for sex, age, household income and educational status as well as over- and underreporting (N = 384)

Dependent variable	Model 1 BMI z-score^a		Model 2 Energy (kcal/day)		Model 3 Fat (g/day)		Model 4 Carb. (g/day)		Model 5 Junk Food^b		Model 6 Simple sugar foods^b
	β	p-value	β	p-value	β	p-value	β	p-value	exp(β)	p-value	exp(β)	p-value
Food retail index^c	0.11	0.17	-12.15	0.60	-2.08	0.24	2.43	0.65	1.04	0.57	0.99	0.87
Age	0.08	0.37	71.08	0.01	3.13	0.10	9.46	0.049	1.04	0.63	0.91	0.29
Sex (ref: male)	0.22	0.06	-127.4	0.001	-3.30	0.21	-19.9	0.003	0.98	0.86	0.95	0.68
High income (ref: low)^d	-0.32	0.02	-21.22	0.62	-2.72	0.35	7.40	0.31	0.99	0.93	1.15	0.06
High ISCED (ref: low)^e	-0.36	0.006	-28.75	0.50	-1.18	0.69	1.30	0.86	0.84	0.09	1.08	0.35

^a:According to Cole et al. [33].
^b:Frequencies per week.
^c:Number of stores and restaurants per 1000 residents.
^d:Net household income in categories, low: below €2,250.
^e:Max. ISCED level of the parents, low: level 1 and 2 relates to lower secondary education and less.

ISSN: 1479-5868