. 2015 Sep 15;124(4):468–476. doi: 10.1289/ehp.1409551

Table 4.

Creatinine-corrected models of 24-hr metal excretions using same-day spot or first morning urine samples as predictors.^a

Type of sample	Mixed regression model			Pearson correlation
Type of sample	β (95% CI)	Intercept	Estimated R^2c	r^d	p-Value^d
Spot sample (n = 88)^b
As	0.66 (0.55, 0.76)*	2.38	0.67	0.82	< 0.001
Cd	0.63 (0.52, 0.75)*	1.66	0.76	0.87	< 0.001
Co	0.63 (0.52, 0.74)*	1.45	0.71	0.84	< 0.001
Cu	0.55 (0.47, 0.63)*	2.92	0.69	0.83	< 0.001
Pb	0.46 (0.31, 0.61)*	3.10	0.29	0.54	< 0.001
Mo	0.51 (0.39, 0.64)*	3.78	0.44	0.66	< 0.001
Ni	0.47 (0.36, 0.57)*	2.90	0.48	0.69	< 0.001
First morning sample (n = 88)
As	0.79 (0.70, 0.88)*	1.44	0.81	0.90	< 0.001
Cd	0.79 (0.70, 0.88)*	0.96	0.85	0.92	< 0.001
Co	0.69 (0.60, 0.78)*	1.26	0.76	0.87	< 0.001
Cu	0.70 (0.54, 0.86)*	2.02	0.45	0.67	< 0.001
Pb	0.55 (0.36, 0.74)*	2.71	0.28	0.53	< 0.001
Mo	0.50 (0.39, 0.62)*	3.91	0.49	0.70	< 0.001
Ni	0.57 (0.44, 0.70)*	2.35	0.48	0.69	< 0.001
^aSpot and first morning urine samples were analyzed on the log₁₀ scale. ^bSpot urine samples were randomly collected from 11 persons on each day. ^cEstimated from single variable linear regression model. ^dEstimated from Pearson’s correlation analyses. *p < 0.001.