Associations Between Metals in Residential Environmental Media and Exposure Biomarkers

Associations between metals in residential environmental media and exposure biomarkers over time in infants living near a mining-impacted site

Ami R. Zota, Anne M. Riederer, Adrienne S. Ettinger, Laurel A. Schaider, James P. Shine, Chitra J. Amarasiriwardena, Robert O. Wright, and John D. Spengler

Supplementary Information

Table S1. Unadjusted and covariate-adjusted β (SE) estimates for environmental predictors of log blood Pb at 12 months – all observations

Table S2. Unadjusted β (SE) estimates for residential environmental predictors of log hair metal concentrations at 12 months

Table S3.Unadjustedβ (SE) estimates for residential environmental predictors of log hair metals concentrations at 12 months– all observations

Table S4.Covariate-adjusted β (SE) estimates for residential environmental predictors of log hair metals concentrations at 12 months– all observations

Table S5. Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Pb

Table S6.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Pb– all observations

Table S7.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Mn

Table S8.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Mn – all observations

Table S9. Unadjusted andcovariate-adjustedβ (SE) estimates for residential environmental predictors of log odds of Cd detection in 12-month blood samples

Table S1.Unadjusted and covariate-adjusted estimates for environmental predictors of log blood Pb at 12 monthsa – all observations

Predictor / Sampl-ing session / 12 month blood Pb / 12 month-cord Pb
N / Unadjusted / Adjusteda / N / Unadjusted / Adjusteda
β (SE) / R2adj / β (SE) / R2adj / β (SE) / R2adj / β (SE) / R2adj
Indoor PM2.5 / S1 / 40b / -0.09 (0.13) / -0.01 / 0.02 (0.12) / 0.23 / 39 / 0.11 (0.20) / -0.02 / 0.19 (0.21) / -0.02
S2 / 39b / 0.19 (0.15) / 0.01 / 0.16 (0.13) / 0.28 / 38 / 0.25 (0.22) / 0.01 / 0.33 (0.23) / 0.02
Mean / 39 / -0.09 (0.16) / -0.02 / 0.02 (0.15) / 0.25 / 38 / 0.15 (0.24) / -0.02 / 0.26 (0.26) / -0.01
House dust (concen-tration) / S1 / 36 / 0.26* (0.13) / 0.07 / 0.34** (0.14) / 0.23 / 34 / 0.28 (0.18) / 0.04 / 0.42** (0.20) / 0.08
S2 / 40 / 0.31** (0.12) / 0.13 / 0.29** (0.11) / 0.24 / 38 / 0.32* (0.16) / 0.07 / 0.32* (0.17) / 0.06
Mean / 35 / 0.38** (0.15) / 0.15 / 0.42** (0.14) / 0.27 / 33 / 0.40* (0.20) / 0.09 / 0.51** (0.21) / 0.12
House dust (loading) / S1 / 35b / 0.22** (0.09) / 0.13 / 0.17** (0.08) / 0.31 / 34 / 0.22 (0.13) / 0.05 / 0.23 (0.14) / 0.02
S2 / 39b / 0.23** (0.09) / 0.15 / 0.16* (0.08) / 0.31 / 38 / 0.18 (0.14) / 0.02 / 0.16 (0.14) / -0.01
Mean / 34b / 0.27** (0.10) / 0.16 / 0.19* (0.31) / 0.25 / 33 / 0.26 (0.15) / 0.06 / 0.26 (0.16) / 0.02
Tap water - standing / S1 / 19 / -0.39* (0.19) / 0.15 / -0.34** (0.15) / 0.52 / 17 / -0.45 (0.26) / 0.11 / -0.44 (0.27) / 0.12
S2 / 41 / -0.27** (0.11) / 0.11 / -0.27** (0.11) / 0.23 / 39 / -0.27* (0.16) / 0.05 / -0.30* (0.16) / 0.08
Mean / 19 / -0.44 (0.27) / 0.08 / -0.50** (0.23) / 0.51 / 17 / -0.41 (0.38) / 0.01 / -0.58 (0.45) / 0.06
Tap water - flushed / S1 / 37 / -0.04 (0.11) / -0.03 / -0.003 (0.11) / 0.19 / 35 / -0.12 (0.14) / -0.01 / -0.01 (0.17) / -0.02
S2 / 41 / -0.21* (0.12) / 0.05 / -0.23* (0.13) / 0.17 / 39 / -0.21 (0.17) / 0.02 / -0.19 (0.20) / -0.02
Mean / 37 / -0.11 (0.13) / -0.01 / -0.07 (0.14) / 0.20 / 35 / -0.15 (0.17) / -0.01 / 0.004 (0.22) / -0.02
Soil (total) / S1 / 41 / 0.15 (0.10) / 0.04 / 0.15 (0.10) / 0.15 / 39 / 0.10 (0.13) / -0.01 / 0.14 (0.14) / -0.02
Soil (bio-accesssible) / S1 / 38 / 0.14* (0.08) / 0.05 / 0.16* (0.08) / 0.19 / 36 / 0.11 (0.11) / 0.001 / 0.17 (0.12) / 0.03

R2adj = adjusted R squared. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed blood and environmental concentrations; adjusted analyses control for sex, breastfeeding (0-6 months,6 months), mother’s race/ethnicity (White, Native American, Other), and blood sampling season (winter, summer).

bModels fit excluding one 12-month blood Pb outlier (defined as an observation with an absolute studentized residual >3)

Table S2.Unadjusted β (SE) estimates for residential environmental predictors of log hair metal concentrations at 12 monthsa

Sampling session / N Pb, Mn / Pb / Mn / Cd / As
β (SE) / R2adj / β (SE) / R2adj / N / β (SE) / R2adj / N / β (SE) / R2adj
Indoor PM2.5
S1 / 30 / -0.19 (0.19) / 0.002 / 0.28 (0.20) / 0.03 / Not analyzed
(Cd not detected in most PM2.5 samples) / 25 / -0.26 (0.21) / 0.02
S2 / 30 / -0.42** (0.18) / 0.12 / -0.02 (0.19) / -0.04 / 25 / -0.06 (0.21) / -0.04
Mean / 30 / -0.43** (0.21) / 0.09 / 0.18 (0.21) / -0.01 / 25 / -0.19 (0.23) / -0.02
House dust (concentration)
S1 / 30 / 0.09 (0.15) / -0.02 / 0.31 (0.22) / 0.03 / 27 / -0.02 (0.24) / -0.04 / 26 / -0.41** (0.20) / 0.12
S2 / 30 / 0.20 (0.14) / 0.03 / 0.22 (0.21) / 0.01 / 27 / 0.50** (0.20) / 0.18 / 26 / -0.09 (0.21) / -0.04
Mean / 30 / 0.21 (0.17) / 0.02 / 0.48* (0.28) / 0.06 / 27 / 0.40 (0.27) / 0.05 / 26 / -0.36 (0.23) / 0.05
House dust (loading)
S1 / 30 / 0.38** (0.10) / 0.18 / 0.31** (0.09) / 0.27 / 27 / 0.25 (0.15) / 0.06 / 26 / -0.05 (0.14) / -0.04
S2 / 30 / 0.44** (0.10) / 0.41 / 0.27** (0.10) / 0.18 / 27 / 0.46** (0.12) / 0.34 / 26 / 0.14 (0.12) / 0.01
Mean / 30 / 0.44** (0.10) / 0.37 / 0.36** (0.10) / 0.30 / 27 / 0.42** (0.14) / 0.23 / 26 / 0.04 (0.14) / -0.04
Tap water – standing
S1 / 15 / -0.06 (0.20) / -0.07 / -0.19 (0.17) / 0.02 / Not analyzed
(Cd not detected in most tap water samples) / 14 / 0.14 (0.24) / -0.05
S2 / 17 / -0.69** (0.23) / 0.33 / -0.26 (0.16) / 0.10 / 15 / 0.39 (0.40) / -0.00
Mean / 15 / -0.22 (0.26) / -0.02 / -0.19 (0.17) / 0.02 / 14 / 0.30 (0.39) / -0.03
Tap water - flushed
S1 / 30 / 0.15 (0.13) / 0.01 / 0.16 (0.11) / 0.04 / Not analyzed
(Cd not detected in most tap water samples) / 26 / 0.17 (0.25) / -0.02
S2 / 32 / 0.12 (0.16) / -0.01 / 0.06 (0.11) / -0.02 / 27 / 0.36 (0.26) / 0.03
Mean / 30 / 0.21 (0.16) / 0.03 / 0.16 (0.11) / 0.04 / 26 / 0.29 (0.29) / 0.0004
Soil (total)
S1 / 33 / -0.17 (0.14) / 0.02 / 0.20 (0.38) / -0.02 / 28 / 0.04 (0.19) / -0.04 / 27 / -0.13 (0.32) / -0.03
Soil (bioaccessible)
S1 / 30 / -0.10 (0.12) / -0.01 / 0.25 (0.31) / -0.01 / 27 / 0.03 (0.19) / -0.04 / 26 / -0.47**(0.19) / 0.17

R2adj = adjusted R squared. **p<0.05; *p<0.10.aRegression analyses use ln-transformed hair and environmental concentrations; includes only participants with measurements from both sessions and who did not move between sessions; see Table S2 for analyses using all observations. bPb models fit excluding one 12-month hair Pb outlier (defined as an observation with an absolute studentized residual >3).

Table S3.Unadjusted β (SE) estimates for residential environmental predictors of log hair metals concentrations at 12 monthsa–all observations

Sampling session / N Pb, Mn / Pb / Mn / Cd / As
β (SE) / R2adj / β (SE) / R2adj / N / β (SE) / R2adj / N / β (SE) / R2adj
Indoor PM2.5
S1 / 35b / -0.18 (0.18) / -0.00 / 0.27 (0.19) / 0.03 / Not analyzed (Cd not detected in most PM2.5 samples) / 29 / -0.21 (0.19) / 0.01
S2 / 35b / -0.41**(0.18) / 0.12 / -0.05 (0.22) / -0.03 / 30 / -0.09 (0.19) / -0.03
Mean / 35b / -0.41**(0.20) / 0.09 / 0.18 (0.20) / -0.01 / 29 / -0.12 (0.20) / -0.02
House dust (concentration)
S1 / 32 / 0.06 (0.17) / -0.03 / 0.30 (0.22) / 0.03 / 29 / -0.01 (0.25) / -0.04 / 28 / -0.38* (0.19) / 0.10
S2 / 35 / 0.20 (0.17) / 0.01 / 0.21 (0.21) / 0.00 / 32 / 0.41**(0.19) / 0.11 / 30 / -0.08 (0.20) / -0.03
Mean / 31 / 0.22 (0.20) / 0.01 / 0.47* (0.27) / 0.06 / 28 / 0.41 (0.27) / 0.05 / 27 / -0.34 (0.23) / 0.05
House dust (loading)
S1 / 32 / 0.31** (0.12) / 0.15 / 0.31**(0.09) / 0.25 / 29 / 0.23 (0.16) / 0.04 / 28 / -0.06 (0.13) / -0.03
S2 / 35 / 0.49** (0.12) / 0.32 / 0.28**(0.10) / 0.19 / 32 / 0.42**(0.12) / 0.27 / 30 / 0.11 (0.11) / -0.00
Mean / 31 / 0.50** (0.13) / 0.34 / 0.36**(0.10) / 0.30 / 28 / 0.42**(0.14) / 0.23 / 27 / 0.03 (0.14) / -0.04
Tap water – standing
S1 / 17 / -0.15 (0.28) / -0.06 / -0.15 (0.15) / 0.01 / Not analyzed (Cd not detected in most tap water samples) / 16 / 0.10 (0.23) / -0.06
S2 / 36 / 0.04 (0.18) / -0.03 / 0.02 (0.11) / -0.03 / 31 / 0.20 (0.27) / -0.02
Mean / 17 / -0.44 (0.37) / 0.03 / -0.19 (0.22) / -0.30 / 16 / 0.25 (0.35) / -0.03
Tap water - flushed
S1 / 33 / 0.14 (0.16) / -0.004 / 0.17 (0.10) / 0.05 / Not analyzed (Cd not detected in most tap water samples) / 29 / 0.12 (0.23) / -0.03
S2 / 36 / 0.15 (0.17) / -0.00 / 0.08 (0.10) / -0.01 / 31 / 0.32 (0.24) / 0.03
Mean / 33 / 0.22 (0.18) / 0.01 / 0.14 (0.10) / 0.02 / 29 / 0.23 (0.26) / -0.01
Soil (total)
S1 / 36 / -0.05 (0.16) / -0.03 / 0.18 (0.37) / -0.02 / 30 / 0.04 (0.20) / -0.04 / 29 / -0.12 (0.31) / -0.03
Soil (bioaccessible)
S1 / 33 / -0.02 (0.14) / -0.03 / 0.22 (0.30) / -0.02 / 30 / 0.04 (0.20) / -0.03 / 29 / -0.46**(0.17) / 0.19

R2adj = adjusted R squared. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed hair and environmental concentrations.

bPb models fit excluding one 12-month hair Pb outlier (defined as an observation with an absolute studentized residual >3.

Table S4.Covariate-adjusted β (SE) estimates for residential environmental predictors of log hair metals concentrations at 12 monthsa–all observations

Sampling session / N Pb, Mn / Pb / Mn / Cd / As
β (SE) / R2adj / β (SE) / R2adj / N / β (SE) / R2adj / N / β (SE) / R2adj
Indoor PM2.5
S1 / 35b / -0.29 (0.21) / -0.05 / 0.29 (0.22) / 0.07 / Not analyzed (Cd not detected in most PM2.5 samples) / 29 / 0.29 (0.22) / 0.07
S2 / 35b / -0.45**(0.18) / 0.07 / -0.20 (0.25) / -0.06 / 30 / -0.19 (0.22) / -0.09
Mean / 35b / -0.48**(0.21) / 0.05 / 0.15 (0.24) / 0.02 / 29 / -0.20 (0.22) / -0.07
House dust (concentration)
S1 / 32 / 0.11 (0.20) / -0.09 / 0.35 (0.22) / 0.02 / 29 / -0.00 (0.27) / -0.06 / 28 / -0.36 (0.25) / -0.04
S2 / 35 / 0.20 (0.18) / -0.05 / 0.39* (0.22) / 0.02 / 32 / 0.43**(0.20) / 0.05 / 30 / -0.07 (0.21) / -0.11
Mean / 31 / 0.27 (0.22) / -0.05 / 0.63**(0.28) / 0.10 / 28 / 0.40 (0.30) / -0.02 / 27 / -0.28 (0.28) / -0.10
House dust (loading)
S1 / 32 / 0.40**(0.13) / 0.21 / 0.32**(0.09) / 0.27 / 29 / 0.24 (0.16) / 0.04 / 28 / -0.03 (0.15) / -0.14
S2 / 35 / 0.48**(0.12) / 0.30 / 0.30**(0.10) / 0.19 / 32 / 0.43**(0.13) / 0.22 / 30 / 0.11 (0.12) / -0.08
Mean / 31 / 0.56**(0.13) / 0.39 / 0.37**(0.10) / 0.32 / 28 / 0.41**(0.14) / 0.20 / 27 / 0.06 (0.15) / -0.15
Tap water – standing
S1 / 17 / -0.23 (0.28) / -0.002 / -0.28 (0.20) / -0.09 / Not analyzed (Cd not detected in most tap water samples) / 16 / 0.13 (0.29) / -0.44
S2 / 36 / 0.05 (0.20) / -0.09 / 0.06 (0.12) / -0.07 / 31 / 0.16 (0.29) / -0.09
Mean / 17 / -0.47 (0.44) / 0.04 / -0.28 (0.20) / -0.09 / 16 / 0.25 (0.42) / -0.37
Tap water - flushed
S1 / 33 / 0.09 (0.21) / -0.07 / 0.16 (0.11) / 0.03 / Not analyzed (Cd not detected in most tap water samples) / 29 / 0.00 (0.27) / -0.13
S2 / 36 / 0.08 (0.22) / -0.09 / 0.07 (0.11) / -0.07 / 31 / 0.26 (0.27) / -0.07
Mean / 33 / 0.13 (0.25) / -0.07 / 0.14 (0.12) / -0.03 / 29 / 0.12 (0.30) / -0.11
Soil (total)
S1 / 36 / -0.09 (0.18) / -0.08 / 0.06 (0.39) / -0.08 / 30 / 0.03 (0.22) / -0.12 / 29 / -0.18 (0.36) / -0.11
Soil (bioaccessible)
S1 / 33 / -0.08 (0.16) / -0.11 / 0.05 (0.34) / -0.10 / 30 / 0.05 (0.22) / -0.15 / 29 / -0.49**(0.19) / 0.15

R2adj = adjusted R squared. **p<0.05; *p<0.10.aRegression analyses use ln-transformed hair and environmental concentrations; adjusted analyses control for sex, breastfeeding (0-6 months,6 months) and mother’s race/ethnicity (White, Native American, Other). Pb models also adjusted for hair sampling season (winter, summer). bPb models fit excluding one 12-month hair Pb outlier (defined as an observation with an absolute studentized residual >3.

Table S5.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Pba

Predictor / Sampl-ing session / Blood Pb (cord, 12 month, 24 month)
Nobs / Unadjusted / Unadjusted (age)a / Adjustedb
β (SE) / AIC / Variance expl.c / β (SE) / AIC / Variance expl.c / β (SE) / AIC / Var. expl.c
Indoor PM2.5 / S1 / 95 / 0.20** (0.07) / 271 / -0.003 / 0.25** (0.07) / 259 / 0.31 / 0.19** (0.06) / 250 / 0.32
S2 / 93 / 0.11 (0.09) / 266 / -0.01 / 0.17* (0.09) / 255 / 0.33 / 0.11 (0.07) / 249 / 0.35
Mean / 93 / 0.17* (0.09) / 265 / -0.01 / 0.23** (0.09) / 253 / 0.33 / 0.16** (0.07) / 247 / 0.34
House dust (concentr.) / S1 / 85 / 0.26** (0.08) / 236 / 0.03 / 0.29** (0.08) / 227 / 0.30 / 0.34** (0.05) / 210 / 0.29
S2 / 94 / 0.20** (0.07) / 266 / 0.02 / 0.23** (0.07) / 255 / 0.36 / 0.21** (0.03) / 233 / 0.35
Mean / 84 / 0.31** (0.08) / 230 / 0.04 / 0.35** (0.08) / 220 / 0.32 / 0.37** (0.04) / 199 / 0.29
House dust (loading) / S1 / 85 / 0.16** (0.06) / 238 / 0.05 / 0.17** (0.06) / 229 / 0.39 / 0.15** (0.05) / 222 / 0.38
S2 / 94 / 0.14** (0.06) / 267 / 0.03 / 0.16** (0.06) / 257 / 0.38 / 0.13** (0.03) / 244 / 0.37
Mean / 84 / 0.17** (0.06) / 235 / 0.04 / 0.19** (0.06) / 226 / 0.38 / 0.14** (0.04) / 220 / 0.37
Tap water - standing / S1 / 40 / -0.20** (0.07) / 97 / 0.04 / -0.20** (0.07) / 95 / 0.29 / d
S2 / 94 / -0.15* (0.08) / 270 / 0.001 / -0.15* (0.08) / 262 / 0.35 / -0.08 (0.08) / 254 / 0.35
Mean / 39 / -0.11 (0.12) / 98 / -0.002 / -0.14 (0.12) / 96 / 0.28 / d
d
d
Tap water - flushed / S1 / 82 / 0.04 (0.07) / 233 / -0.01 / 0.06 (0.08) / 226 / 0.35 / 0.11 (0.07) / 221 / 0.26
S2 / 95 / -0.09 (0.08) / 276 / -0.01 / -0.07 (0.07) / 257 / 0.34 / -0.03 (0.08) / 249 / 0.36
Mean / 82 / 0.07 (0.08) / 232 / -0.01 / 0.09 (0.09) / 226 / 0.35 / 0.12 (0.08) / 221 / 0.25
Soil (total) / S1 / 95 / 0.11* (0.07) / 274 / 0.01 / 0.14** (0.07) / 264 / 0.34 / 0.11** (0.05) / 254 / 0.36
Soil (bioacc.) / S1 / 88 / 0.09* (0.05) / 250 / 0.02 / 0.13** (0.06) / 239 / 0.36 / 0.08** (0.04) / 224 / 0.35

Nobs = number of observations used in mixed regression, up to 3 per child. AIC = Akaike’s Information Criterion. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed blood and environmental concentrations; unadjusted (age) analyses control for child’s age (days) at blood sample collection. Models exclude data from families who moved between residential sampling sessions and families who moved between the 12- and 24-month blood sample collection.

bAdjusted analyses control for child’s age (days) at blood sample collection, maternal smoking (yes/no), maternal age, child’s sex, breastfeeding (0-6 months,6 months), mother’s race/ethnicity (White, Native American, Other), and blood sampling season (winter, summer, not sampled).

cVariance explained calculated as the R-squared from a regression of predicted versus observed values.

dModel not fit; too many likelihood evaluations.

Table S6.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Pba– all observations

Predictor / Sampl-ing session / Blood Pb (cord, 12 month, 24 month)
Nobs / Unadjusted / Unadjusted (age)a / Adjustedb
β (SE) / AIC / Variance expl.c / β (SE) / AIC / Variance expl.c / β (SE) / AIC / Var. expl.c
Indoor PM2.5 / S1 / 109 / 0.16** (0.08) / 302 / -0.01 / 0.18** (0.08) / 288 / 0.29 / 0.19** (0.07) / 283 / 0.27
S2 / 107 / 0.14 (0.09) / 295 / -0.01 / 0.18* (0.10) / 279 / 0.32 / 0.14* (0.08) / 278 / 0.32
Mean / 107 / 0.14 (0.10) / 295 / -0.01 / 0.18* (0.10) / 280 / 0.31 / 0.20** (0.08) / 276 / 0.29
House dust (concentr.) / S1 / 96 / 0.27** (0.08) / 255 / 0.02 / 0.29** (0.08) / 241 / 0.30 / 0.29** (0.08) / 245 / 0.31
S2 / 106 / 0.20** (0.07) / 290 / 0.02 / 0.22** (0.07) / 277 / 0.33 / 0.17** (0.06) / 269 / 0.29
Mean / 93 / 0.31** (0.08) / 243 / 0.04 / 0.33** (0.08) / 231 / 0.31 / 0.33** (0.07) / 223 / 0.29
House dust (loading) / S1 / 96 / 0.13** (0.06) / 259 / 0.04 / 0.15** (0.06) / 246 / 0.35 / 0.14** (0.06) / 251 / 0.34
S2 / 106 / 0.14** (0.06) / 291 / 0.03 / 0.16** (0.06) / 278 / 0.35 / 0.13** (0.04) / 268 / 0.29
Mean / 93 / 0.14** (0.06) / 250 / 0.04 / 0.16** (0.06) / 239 / 0.35 / 0.14** (0.05) / 233 / 0.29
Tap water - standing / S1 / 52 / -0.25** (0.06) / 123 / 0.06 / -0.25** (0.07) / 118 / 0.31 / -0.31** (0.06) / 124 / 0.28
S2 / 108 / -0.13 (0.08) / 230 / -0.001 / -0.13 (0.08) / 287 / 0.33 / -0.04 (0.07) / 285 / 0.30
Mean / 51 / -0.18 (0.13) / 127 / 0.02 / -0.21 (0.13) / 22 / 0.28 / -0.50** (0.15) / 126 / 0.27
Tap water - flushed / S1 / 100 / 0.07 (0.07) / 276 / -0.01 / 0.10 (0.07) / 264 / 0.33 / 0.11 (0.07) / 264 / 0.29
S2 / 109 / 0.004 (0.08) / 306 / -0.01 / 0.004 (0.08) / 292 / 0.34 / -0.01 (0.07) / 289 / 0.31
Mean / 100 / 0.06 (0.08) / 277 / -0.01 / 0.08 (0.08) / 265 / 0.33 / 0.09 (0.09) / 265 / 0.29
Soil (total) / S1 / 109 / 0.11* (0.06) / 303 / 0.02 / 0.11* (0.07) / 289 / 0.35 / 0.08 (0.06) / 287 / 0.32
Soil (bioacc.) / S1 / 102 / 0.11** (0.05) / 280 / 0.03 / 0.12** (0.06) / 265 / 0.35 / 0.09* (0.05) / 260 / 0.34

Nobs = number of observations used in mixed regression, up to 3 per child. AIC = Akaike’s Information Criterion. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed blood and environmental concentrations; unadjusted (age) analyses control for child’s age (days) at blood sample collection.

bAdjusted analyses control for child’s age (days) at blood sample collection, maternal smoking (yes/no), maternal age, child’s sex, breastfeeding (0-6 months,6 months), mother’s race/ethnicity (White, Native American, Other), and blood sampling season (winter, summer).

cVariance explained calculated as the R-squared from a regression of predicted versus observed values.

Table S7.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Mna

Predictor / Sampl-ing session / Blood Mn (cord, 12 month, 24 month)
Nobs / Unadjusted / Unadjusted (age)a / Adjustedb
β (SE) / AIC / Variance expl.c / β (SE) / AIC / Variance expl.c / β (SE) / AIC / Var. expl.c
Indoor PM2.5 / S1 / 85 / -0.05 (0.09) / 136.2 / -0.01 / -0.05 (0.06) / 101 / 0.52 / -0.06 (0.05) / 88 / 0.58
S2 / 93 / -0.03 (0.08) / 154 / -0.01 / -0.05 (0.06) / 109 / 0.53 / -0.03 (0.06) / 107 / 0.58
Mean / 83 / -0.05 (0.09) / 133 / -0.01 / -0.08 (0.07) / 99 / 0.52 / -0.08 (0.06) / 86 / 0.58
House dust (concentr.) / S1 / 85 / -0.02 (0.09) / 142 / -0.01 / -0.05 (0.08) / 102 / 0.51 / -0.09 (0.07) / 98 / 0.57
S2 / 94 / 0.05 (0.08) / 154 / -0.01 / -0.03 (0.06) / 110 / 0.52 / -0.01 (0.05) / 107 / 0.57
Mean / 84 / 0.04 (0.11) / 139 / -0.01 / -0.04 (0.08) / 101 / 0.51 / -0.09 (0.07) / 96 / 0.56
House dust (loading) / S1 / 85 / -0.04 (0.04) / 141 / -0.01 / -0.02 (0.03) / 102 / 0.52 / -0.04 (0.03) / 98 / 0.58
S2 / 94 / 0.05 (0.04) / 153 / -0.01 / 0.02 (0.03) / 110 / 0.53 / 0.001 (0.03) / 107 / 0.57
Mean / 84 / 0.002 (0.05) / 139 / -0.01 / 0.0002 (0.03) / 101 / 0.51 / -0.03 (0.03) / 96 / 0.57
Tap water - standing / S1 / 40 / -0.10 (0.05) / 66 / -0.02 / -0.09 (0.05) / 52 / 0.54 / d
S2 / 94 / -0.08** (0.04) / 153 / -0.006 / -0.04 (0.03) / 109 / 0.53 / -0.03 (0.03) / 108 / 0.58
Mean / 40 / -0.10* (0.05) / 66 / -0.02 / -0.09 (0.05) / 52 / 0.54 / d
d
d
Tap water - flushed / S1 / 82 / -0.11** (0.04) / 130 / 0.01 / -0.11** (0.04) / 96 / 0.53 / -0.09** (0.03) / 93 / 0.60
S2 / 95 / -0.04 (0.03) / 156 / -0.01 / -0.03 (0.03) / 110 / 0.53 / -0.04* (0.02) / 107 / 0.59
Mean / 82 / -0.11** (0.04) / 130 / 0.01 / -0.06** (0.03) / 96 / 0.53 / -0.09** (0.03) / 93 / 0.60
Soil (total) / S1 / 95 / 0.05 (0.15) / 157 / -0.01 / 0.05 (0.10) / 111 / 0.53 / 0.01 (0.09) / 109 / 0.57
Soil (bioacc.) / S1 / 88 / 0.02 (0.10) / 150 / -0.01 / -0.01 (0.08) / 109 / 0.52 / -0.06(0.07) / 107 / 0.56

Nobs = number of observations used in mixed regression, up to 3 per child. AIC = Akaike’s Information Criterion. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed blood and environmental concentrations; unadjusted (age) analyses control for child’s age (days) at blood sample collection. Models exclude data from families who moved between residential sampling sessions or between the 12- and 24-month blood sample collection.

bAdjusted analyses control for child’s age (days) at blood sample collection, maternal smoking (yes/no), maternal age, child’s sex, breastfeeding (0 - 6 months, 6 months), and child’s race/ethnicity (White, Native American, Other).

cVariance explained calculated as the R-squared from a regression of predicted versus observed values.

dModel did not converge.

Table S8.Unadjusted and covariate-adjusted β (SE) mixed model estimates for environmental predictors of log blood Mna – all observations

Predictor / Sampl-ing session / Blood Mn (cord, 12 month, 24 month)
Nobs / Unadjusted / Unadjusted (age)a / Adjustedb
β (SE) / AIC / Variance expl.c / β (SE) / AIC / Variance expl.c / β (SE) / AIC / Var. expl.c
Indoor PM2.5 / S1 / 99 / 0.01 (0.07) / 144.2 / -0.01 / 0.03 (0.06) / 109 / 0.54 / 0.04 (0.05) / 101 / 0.60
S2 / 107 / -0.01 (0.07) / 163 / -0.01 / -0.01 (0.06) / 115 / 0.56 / 0.01 (0.06) / 109 / 0.61
Mean / 97 / 0.01 (0.08) / 141 / -0.01 / 0.01 (0.07) / 107 / 0.54 / 0.07 (0.06) / 98 / 0.61
House dust (concentr.) / S1 / 96 / 0.01 (0.08) / 151 / -0.01 / 0.02 (0.07) / 109 / 0.54 / 0.02 (0.06) / 103 / 0.60
S2 / 106 / 0.04 (0.07) / 159 / -0.01 / -0.01 (0.06) / 112 / 0.55 / 0.04 (0.05) / 107 / 0.61
Mean / 93 / 0.04 (0.10) / 145 / -0.01 / 0.03 (0.08) / 104 / 0.53 / 0.02 (0.08) / 100 / 0.60
House dust (loading) / S1 / 96 / -0.03 (0.04) / 151 / -0.01 / 0.01 (0.03) / 109 / 0.53 / -0.02 (0.03) / 103 / 0.60
S2 / 106 / 0.02 (0.04) / 159 / -0.01 / 0.02 (0.03) / 112 / 0.56 / 0.02 (0.03) / 107 / 0.61
Mean / 93 / -0.02 (0.04) / 145 / -0.01 / 0.01 (0.04) / 104 / 0.53 / -0.01 (0.03) / 100 / 0.60
Tap water - standing / S1 / 52 / -0.07 (0.05) / 80 / -0.02 / -0.06 (0.05) / 61 / 0.54 / -0.08 (0.05) / 61 / 0.60
S2 / 108 / -0.09** (0.03) / 160 / -0.004 / -0.05* (0.03) / 114 / 0.56 / -0.05* (0.03) / 114 / 0.61
Mean / 52 / -0.07 (0.05) / 80 / -0.02 / -0.06 (0.05) / 61 / 0.54 / -0.08 (0.05) / 64 / 0.61
Tap water - flushed / S1 / 100 / -0.11** (0.03) / 146 / 0.01 / -0.07** (0.03) / 104 / 0.56 / -0.09** (0.02) / 94 / 0.62
S2 / 109 / -0.04 (0.03) / 165 / -0.01 / -0.02 (0.03) / 116 / 0.56 / -0.02 (0.03) / 110 / 0.61
Mean / 100 / -0.11** (0.03) / 146 / 0.01 / -0.07** (0.03) / 104 / 0.56 / -0.09** (0.02) / 94 / 0.62
Soil (total) / S1 / 109 / 0.03 (0.10) / 166 / -0.01 / 0.04 (0.09) / 116 / 0.56 / 0.05 (0.08) / 111 / 0.60
Soil (bioacc.) / S1 / 102 / 0.02 (0.09) / 160 / -0.01 / -0.002 (0.07) / 114 / 0.55 / 0.01 (0.06) / 108 / 0.60

Nobs = number of observations used in mixed regression, up to 3 per child. AIC = Akaike’s Information Criterion. **p<0.05; *p<0.10.

aRegression analyses use ln-transformed blood and environmental concentrations; unadjusted (age) analyses control for child’s age (days) at blood sample collection.

bAdjusted analyses control for child’s age (days) at blood sample collection, maternal smoking (yes/no), maternal age, child’s sex, breastfeeding (0 - 6 months, 6 months), and child’s race/ethnicity (White, Native American, Other).

cVariance explained calculated as the R-squared from a regression of predicted versus observed values.

Table S9.Unadjusted and covariate-adjusted β (SE) estimates for residential environmental predictors of log odds of Cd detection in 12-month blood samplesa

Predictor / Sampl-ing session / 12 month blood samples – Families who did not move, with data from both sampling sessions / 12 month blood samples – All observations
N / Unadjusted / Adjusteda / N / Unadjusted / Adjusteda
β (SE) / R2b / β (SE) / R2b / β (SE) / R2b / β (SE) / R2b
Indoor PM2.5 / c
House dust (concentr.) / S1 / 31 / -0.31 (0.42) / 0.02 / d / 36 / -0.40 (0.41) / 0.03 / -0.35 (0.45) / 0.12
S2 / 36 / -0.28 (0.36) / 0.02 / -0.22 (0.37) / 0.07 / 41 / -0.16 (0.35) / 0.004 / -0.10 (0.36) / 0.05
Mean / 31 / -0.26 (0.47) / 0.01 / d
d / 35 / -0.21 (0.45) / 0.01 / -0.18 (0.48) / 0.11
House dust (loading) / S1 / 31 / -0.13 (0.26) / 0.01 / d / 36 / -0.06 (0.24) / 0.002 / -0.03 (0.27) / 0.10
S2 / 36 / -0.27 (0.26) / 0.03 / -0.30 (0.27) / 0.10 / 41 / -0.19 (0.24) / 0.01 / -0.21 (0.25) / 0.07
Mean / 31 / -0.17 (0.28) / 0.01 / d
0.10 / 35 / -0.08 (0.26) / 0.003 / -0.06 (0.29) / 0.10
Tap water - standing / e
Tap water - flushed / e
Soil (total) / S1 / 34 / 0.55* (0.31) / 0.10 / 0.68* (0.36) / 0.20 / 39 / 0.48* (0.29) / 0.08 / 0.61* (0.33) / 0.17
Soil (bioacc.) / S1 / 32 / 0.48 (0.31) / 0.08 / 0.59* (0.34) / 0.17 / 38 / 0.40 (28) / 0.06 / 0.57* (0.33) / 0.15

*p<0.10.

aRegression analyses use ln-transformed environmental concentrations; 12-month adjusted analyses control for child’s sex, breastfeeding (0-6 months, 6 months),maternal smoking (yes/no), and mother’s race/ethnicity (White, Native American, Other).

bGeneralized R2 from SAS PROC LOGISTIC.

cNot modeled; Cd detected in <10% of indoor PM2.5 samples.

dNot modeled; validity of model fit questionable.

eNot modeled; Cd detected in <50% of standing and <20% of flushed tap water samples.