www.ehjournal.net

Table 3
Multiple linear regression analyses for the associations between urine bisphenol A (BPA) concentration (ng/mL) and body mass index (BMI, Kg/m2) among study participants
Group	Crude analysis^a	Adjusted model 1^b	Adjusted model 2^c
	β ^d (95% CI)	β (95% CI)	β (95% CI)
All	0.019 (0.002, 0.035)	0.018 (0.004, 0.032)	0.017 (0.002, 0.032)
	P = 0.027	P = 0.011	P = 0.026
Age (years)
8-11	0.010 (-0.010, 0.030)	0.020 (0.001, 0.038)	0.017 (-0.002, 0.037)
	P = 0.330	P = 0.041	P = 0.081
12-15	0.013 (-0.012, 0.039)	0.020 (-0.003, 0.044)	0.019 (-0.005, 0.044)
	P = 0.306	P = 0.088	P = 0.118
Sex
Female	0.021 (-0.001, 0.043)	0.020 (0.002, 0.038)	0.015 (-0.004, 0.034)
	P = 0.065	P = 0.028	P = 0.115
Male	0.017 (-0.008, 0.042)	0.016 (-0.007, 0.038)	0.018 (-0.005, 0.041)
	P = 0.188	P = 0.164	P = 0.123

Abbreviation: CI (confidence interval).

^aIncluded a variable of urine BPA corrected for specific gravity.

^bNot corrected urine BPA concentrations by specific gravity and included the covariables of age and sex.

^cCorrected urine BPA concentrations by specific gravity and included the covariables of age and sex.

^dRegression coefficient.

Wang et al. Environmental Health 2012 11:79 doi:10.1186/1476-069X-11-79

Table 3