Association between economic status and body mass index among adolescents: a community-based cross-sectional study in Japan

Table 3 Relationship between economic status and BMI Z-Scores of Boys

	Model 1^a			Model 2^b (+ grade, school, family structure)			Model 3 (Model 2 + family members’ smoking status)			Model 4 (Model 2 + breakfast skipping)			Model 5 (Model 2 + activities outside of school)			Model 6 (Model 2 + sleeping patterns)			Model 7 (Model 2 + self-rated health status)			Model 8^c (Model 2 + all mediators)
	B	SE	p	B	SE	p	B	SE	p	B	SE	p	B	SE	p	B	SE	p	B	SE	p	B	SE	p
Family characteristics
Economic status
Low (ref: non-low)	−0.07	0.10	0.460	−0.10	0.10	0.327	−0.10	0.10	0.326	−0.10	0.10	0.312	−0.10	0.10	0.302	−0.10	0.10	0.318	−0.10	0.10	0.291	−0.11	0.10	0.241
Family structure
Single parent or stepfamily (ref: both biological parents)	0.15^**	0.05	0.005	0.15^**	0.05	0.005	0.15^**	0.05	0.005	0.14^**	0.05	0.009	0.15^**	0.05	0.004	0.15^**	0.05	0.004	0.15^**	0.05	0.005	0.14^**	0.05	0.006
Family members’ smoking status
Yes (ref: none)	0.003	0.04	0.944				0.014	0.04	0.725													0.01	0.04	0.798
Child characteristics
Breakfast skipping
Yes (ref: no)	0.13^*	0.06	0.029							0.11	0.06	0.083										0.08	0.06	0.199
Convenience stores or fast-food outlets
<1 time per week (ref: ≥ 1 time per week)	−0.01	0.05	0.802																			−0.05	0.05	0.343
Club activities at school
Non-physical (ref: physical)	0.11	0.08	0.193																			0.02	0.08	0.804
No participation (ref: physical)	0.03	0.05	0.538																			0.03	0.07	0.665
Activities outside of school
No (ref: yes)	−0.06	0.04	0.130										−0.07	0.04	0.078							−0.08	0.04	0.049
Sleeping patterns
<6 h 30 min (ref: ≥ 6 h 30 min)	0.19^***	0.05	<0.001													0.18^***	0.05	<0.001				0.17^**	0.05	0.001
Self-rated health status
Not good (ref: good)	0.09	0.08	0.217																0.11	0.08	0.151	0.08	0.08	0.326
Smoking status
Yes (ref: no)	0.13	0.11	0.226																			0.04	0.11	0.736

B = Partial regression coefficient, ^aunivariate regression; ^bmultiple regression adjusted for confounding variables; ^cmultiple regression adjusted for all mediators
^* p < 0.05; ^** p < 0.01; ^*** p < 0.001

ISSN: 2052-9538