SUPPLEMENTAL INFORMATION
Table S1. Slopes (95% CI) for sample-weighted, age-stratified multiple regression food models of urinary perchlorate [ng/mL] data in NHANES 2001 – 2008. Estimates corrected for urine volume by including urinary creatinine as a predictor in the food models.
/ Children 6 – 11 YO / Adolescents 12 – 19 YO / Adults 20 – 84 YO /Predictor / Slope (95% CI) / p-Value* / Slope (95% CI) / p-Value* / Slope (95% CI) / p-Value* /
Intercept / 1.93 [0.14:3.72] / <0.0001 / 2.64 [1.38:3.90] / <0.0001 / 0.69 [-1.11:2.48] / 0.0355
Sex
Female / Ref. / Ref. / Ref.
Male / 0.64 [-0.027:1.31] / 0.0021 / 0.39 [-0.095:0.87] / 0.0306 / -0.022 [-0.54:0.49] / 0.3906
Age at Screening
20 – 39 YO / Ref.
40 – 59 YO / 0.44 [-0.063:0.95] / 0.0034
60 – 84 YO / 0.47 [-0.18:1.13] / <0.0001
Race
Non–Hispanic White / Ref. / Ref. / Ref.
Non–Hispanic Black / -0.33 [-1.09:0.43] / 0.1044 / -0.57 [-1.18:0.04] / 0.0002 / -1.79 [-2.44:-1.14] / <0.0001
Mexican American / -0.09 [-1.01:0.83] / 0.1784 / 0.29 [-0.42:1.00] / 0.3000 / 0.32 [-0.50:1.14] / 0.0153
Other Hispanic / -0.48 [-1.90:0.94] / 0.8623 / 0.13 [-0.99:1.25] / 0.4909 / -0.43 [-1.22:0.36] / 0.2102
Other/Multi–Racial / 0.09 [-1.09:1.28] / 0.9792 / -0.24 [-1.50:1.03] / 0.3866 / -0.60 [-1.31:0.10] / 0.3572
Creatinine, urine (g/L) / 3.95 [3.21:4.68] / <0.0001 / 2.16 [1.85:2.46] / <0.0001 / 2.89 [2.46:3.31] / <0.0001
Body Mass Index / 0.01 [-0.09:0.10] / 0.3365 / -0.02 [-0.05:0.01] / 0.6739 / 0.01 [-0.02:0.05] / 0.0058
Fasting Time / -0.12 [-0.18:-0.07] / <0.0001 / -0.10 [-0.15:-0.05] / <0.0001 / -0.06 [-0.10:-0.02] / 0.0001
Education
HS Graduate or Higher / Ref. / Ref. / Ref.
Less Than HS Graduate / 0.35 [-0.73:1.43] / 0.8102 / 0.13 [-0.72:0.97] / 0.8177 / 0.56 [-0.33:1.45] / 0.3512
Poverty Income Ratio
1.00 or Higher / Ref. / Ref. / Ref.
<1.00 (Poor) / 0.02 [-0.80:0.84] / 0.2464 / -0.23 [-0.83:0.37] / 0.2611 / -0.66 [-1.18:-0.14] / 0.0820
Tobacco/Nicotine Last 5 Days
No / Ref. / Ref.
Yes / -0.23 [-0.86:0.41] / 0.0513 / 0.62 [-0.10:1.35] / 0.1555
NHANES Cycle
2001–2002 / 0.55 [-0.76:1.85] / 0.4277 / -0.08 [-0.79:0.63] / 0.8393 / 0.23 [-0.68:1.14] / 0.0933
2003–2004 / Ref. / Ref. / Ref.
2005–2006 / -0.38 [-1.39:0.64] / 0.7658 / 0.21 [-0.62:1.05] / 0.4046 / 0.0007 [-0.78:0.78] / 0.4359
2007–2008 / 0.80 [-0.17:1.77] / 0.3309 / 0.98 [0.10:1.86] / 0.0231 / 0.99 [0.13:1.85] / 0.0070
Diet Intake Day
Weekday / Ref. / Ref. / Ref.
Weekend / -0.07 [-0.63:0.49] / 0.3016 / 0.50 [0.03:0.96] / 0.0050 / 0.05 [-0.45:0.56] / 0.0571
Mass Consumed
Milk Products / 2.93 (1.52, 4.34) / <0.0001 / 1.54 (0.87, 2.21) / <0.0001 / 0.69 (0.05, 1.32) / <0.0001
Meat, Poultry, Fish / 0.68 (-2.03, 3.40) / 0.5464 / 0.40 (-0.51, 1.31) / 0.1766 / -0.05 (-1.04, 0.95) / 0.1838
Eggs / 1.67 (-4.68, 8.03) / 0.6784 / 0.46 (-5.22, 6.14) / 0.9163 / 4.39 (-1.04, 9.83) / 0.0031
Legumes, Nuts, Seeds / 1.17 (-3.97, 6.31) / 0.3061 / 8.71 (-2.52, 19.93) / 0.2856 / 0.40 (-2.47, 3.28) / 0.5764
Grain Products / -1.19 (-2.89, 0.52) / 0.0203 / -0.65 (-1.55, 0.26) / 0.3195 / -0.50 (-1.11, 0.11) / 0.3193
Fruits / 0.66 (-0.96, 2.29) / 0.3724 / 0.56 (-0.52, 1.63) / 0.0640 / 1.21 (-0.08, 2.50) / <0.0001
Vegetables / 0.71 (-2.78, 4.19) / 0.8779 / 2.11 (-0.004, 4.22) / 0.0078 / 2.62 (0.69, 4.56) / <0.0001
Dark-Green Leafy Vegetables / 49.75 (-24.45, 123.95) / 0.1688 / 11.47 (-10.70, 33.64) / 0.1411 / 30.83 (5.46, 56.20) / 0.0035
Fats, Oils, Salad Dressings / 19.02 (-6.60, 44.64) / 0.0198 / 5.50 (-4.79, 15.80) / 0.1629 / -2.76 (-14.17, 8.65) / 0.3531
Sugars, Sweets, Beverages / -0.33 (-1.01, 0.34) / 0.1549 / -0.33 (-0.59, -0.07) / 0.0152 / -0.13 (-0.37, 0.12) / 0.0646
Water (Not Bottled) At Home / -0.04 (-0.58, 0.51) / 0.7463 / 0.09 (-0.51, 0.69) / 0.9755 / -0.20 (-0.43, 0.03) / 0.0407
CI: confidence interval
HS: high school
NHANES: National Health and Nutrition Examination Survey
Ref.: reference group
YO: years-old
a: p-Values estimated from identical models where the dependent variable was ln-transformed urinary perchlorate. Hypothesis tests for categorical predictors represent comparisons with reference group, and for continuous predictors represent comparisons with a slope equal to zero.
b: Educational attainment was for the highest among the household reference person and their spouse.
Table S2. Change in urinary perchlorate [ng/mL] attributable to mass consumed estimated from parsimonious, sample-weighted, multiple regression drinking water models of NHANES 2005 – 2006 data. Estimates corrected for urine volume by including urinary creatinine as a predictor in the drinking water models.
(n = 908) / Adults 20 – 84 YO
(n = 1,933) /
Predictor / Slope (95% CI) / p-Valuea / Slope (95% CI) / p-Valuea /
Intercept / 3.46 (1.78, 5.15) / <0.0001 / -1.16 (-6.24, 3.92) / 0.0388
Age at Screening
20 – 39 YO / Ref.
40 – 59 YO / 1.72 (0.31, 3.13) / 0.0266
60 – 84 YO / 1.74 (0.18, 3.30) / 0.0016
Non-Hispanic Black
No / Ref. / Ref.
Yes / -1.30 (-1.98, -0.63) / <0.0001 / -1.27 (-2.72, 0.18) / 0.0045
Urinary creatinine [g/L] / 2.24 (1.54, 2.95) / <0.0001 / 3.58 (2.01, 5.15) / <0.0001
Body Mass Index / -0.02 (-0.07, 0.02) / 0.8349 / 0.01 (-0.09, 0.10) / 0.3517
Fasting Time / -0.12 (-0.22, -0.01) / 0.0074 / 0.01 (-0.08, 0.11) / 0.2679
Poverty Income Ratio
1.00 or Higher / Ref. / Ref.
<1.00 (Poor) / -0.89 (-1.51, -0.26) / 0.0187 / 0.44 (-0.85, 1.72) / 0.0937
Perchlorate from Water (NB) at Home [ng] / -1.38E-5 (-1.32E-4, 1.05E-4) / 0.4222 / 2.11E-4 (-5.56E-6, 4.27E-4) / 0.0035
USDA Food Group [kg]
Milk Products / 1.56 (0.19, 2.93) / 0.0009 / 1.32 (0.19, 2.45) / 0.0001
Dark-Green Leafy Vegetables / 63.73 (3.25, 124.21) / 0.0199 / 19.34 (10.58, 28.11) / <0.0001
Adolescent drinking water model controls for being urinary creatinine, being non-Hispanic black, body mass index, fasting time, and tobacco/nicotine in the last 5 days.
Adult drinking water model controls for urinary creatinine, age (20 – 39 YO, 40 – 59 YO, and 60 – 84 YO), being non-Hispanic black, body mass index, fasting time, and tobacco/nicotine in the last 5 days.
Slopes for USDA food groups are in units of ng perchlorate/mL per kg food reported consumed in the 24 hr preceding the dietary recall interview conducted in-person at the NHANES Mobile Examination Center.
Slope for perchlorate from water (NB) at home is in units of ng perchlorate/mL per ng perchlorate consumed from residential water (NB) at home.
CI: confidence interval
NB: not bottled
NHANES: National Health and Nutrition Examination Survey
Ref.: reference group
YO: years-old
a: p-Values estimated from identical models where the dependent variable was ln-transformed urinary perchlorate. Hypothesis tests represent comparisons with a slope equal to zero.
4
Table S3. Proficiency test results for perchlorate analysis accuracy.
Test 1 / Theoretical Concentration[ng/mL] / Analytic Result [ng/mL] / % Accuracy / % Difference
72 / 73.5 / 102 / 2
0.25 / 0.244 / 98 / -2
12 / 12.1 / 101 / 1
2.4 / 2.3 / 96 / -4
12 / 11.7 / 98 / -3
Test 2 / Theoretical Concentration
[ng/mL] / Analytic Result [ng/mL] / % Accuracy / % Difference
72 / 73.7 / 102 / 2
2.4 / 2.53 / 105 / 5
0.25 / 0.268 / 107 / 7
12 / 12.5 / 104 / 4
72 / 74.9 / 104 / 4
4