Anthropometric Characteristics of Underprivileged Adolescents: A Study from Urban Slums of India

Table 5

Thinness-regression coefficients (β) and 90% confidence intervals under the mixed effects logistic regression model for thinness by sex. Log odds of thinness versus normal BMI were modeled. Each model included age and an additional covariate.

		Boys		Girls
		β (CI)	exp(β) (CI)
Age		−0.10 (−0.26, 0.06)	0.91 (0.77, 1.06)	−0.18 (−0.35, −0.02)	0.83 (0.70, 0.98)
Household size	4 or less	Reference		Reference
	5	0.17 (−0.89, 1.23)	1.18	−0.88 (−1.87, 0.11)	0.41 (0.15, 3.42)
	More than 5	0.36 (−0.68, 1.41)	1.44 (0.51, 4.10)	−0.76 (−1.56, 0.03)	0.47 (0.21, 1.03)
Mother’s education	Primary or no	Reference		Reference
	Secondary or higher	0.12 (−0.71, 0.96)	1.13 (0.49, 2.61)	−1.42 (−2.32, −0.52)	0.24 (0.10, 0.59)
Income	Standardised	0.04 (−0.34, 0.43)	1.04 (0.71, 1.54)	−1.02 (−1.64, −0.41)	0.34 (0.19, 0.66)
Consumption of	Oil	1.13 (−2.62, 4.87)	3.08 (0.07, 130.32)	1.40 (−1.88, 4.69)	4.08 (0.15, 108.85)
	Dal/pulses	−7.44 (−19.26, 4.38)	0.0006 (0.43 × 10^−8, 79.83)	3.84 (−2.78, 10.46)	46.36 (0.06, 34891.55)
	Meat/egg/fish	−2.06 (−6.47, 2.35)	0.13 (0.002, 10.49)	0.49 (−3.53, 4.52)^1	1.64 (0.03, 91.84)

Age effect was not significant.
0.01 ≤ -value ≤ 0.10, and -value < 0.01.