Validity and predictors of BMI derived from self-reported height and weight among 11- to 17-year-old German adolescents from the KiGGS study

Table 5 Summary of multivariate logistic regression models

	Model 2a (n = 615)			Model 2b (n = 2 593)			Model 2c (n = 2 593)
	Odds ratio	95% CI	p	Odds ratio	95% CI	p	Odds ratio	95% CI	p
Age (per year)	1.04	0.95-1.15	0.387	0.92	0.81-1.06	0.249	0.97	0.86-1.09	0.588
Gender
Boys	0.78	0.53-1.15	0.210	2.68	1.50-4.80	0.001	0.69	0.48-0.99	0.045
Girls	1.00	reference	reference	1.00	reference	reference	1.00	reference	reference
Body perception
Too thin	-	-	-	0.63	0.21-1.85	0.399	2.97	1.99-4.43	0.000
Too fat	0.28	0.14-0.54	0.000	2.20	1.25-3.87	0.006	0.28	0.16-0.48	0.000
Right weight	1.00	reference	reference	1.00	reference	reference	1.00	reference	reference
Symptoms of an eating disorder
Conspicuous	0.78	0.52-1.18	0.241	not significant in the univariate model			not significant in the univariate model
Inconspicuous	1.00	reference	reference
Sexual maturation
Tanner 1							2.48	1.20-5.13	0.014
Tanner 2-3	not significant in the univariate model			not significant in the univariate model			2.40	1.46-3.94	0.001
Tanner 4-6							1.00	reference	reference
SES
Per index point	1.06	1.01-1.11	0.021	not significant in the univariate model			not significant in the univariate model
Migrant background
Migrant	not significant in the univariate model			2.36	1.28-4.34	0.006	not significant in the univariate model
Non-migrant				1.00	reference	reference
Parental overweight
Both	0.47	0.28-0.82	0.007	4.16	1.75-9.88	0.001	0.49	0.28-0.86	0.013
One	0.64	0.38-1.06	0.085	2.71	1.18-6.24	0.019	0.84	0.58-1.21	0.343
None	1.00	reference	reference	1.00	reference	reference	1.00	reference	reference
Pseudo R ²	0.077			0.083			0.122

Model 2a includes all overweight individuals, with the target variable 'overweight misclassified by self-reports as normal weight or underweight' (n = 77 individuals misclassified).
Model 2b includes all normal weight individuals, with the target variable 'normal weight misclassified by self-reports as overweight' (n = 42 individuals misclassified).
Model 2c includes all normal weight individuals, with the target variable 'normal weight misclassified by self-reports as underweight' (n = 81 individuals misclassified).

ISSN: 1756-0500