Supplement 3. Studies assessing exposure to metal mixtures and neurodevelopmental/cognitive or behavioral outcomes.

Authors & Year / Study design / Location / Sample size / Exposure Measure / Outcome Measure / Confounders / Major Finding
Neurodevelopment/Cognition
Bobbet al. 2014[16] / Prospective cohort / Bangladesh / 375 / As, Mn, Pb in umbilical cord blood / BSID-III (motor development) at approx. 2 years of age / Gender, age at time of assessment, maternal education, maternal IQ, clinic, HOME score / Inverted U-shaped association between Mn and PDI, but only at middle levels of As exposure
Yang et al. 2014 [79] / Prospective cohort / China / 933 / Umbilical cord serum Mn, Se / NBNA in 3-day olds / Gender, maternal age, maternal and paternal education, maternal and paternal occupation, family income, gestational age, birth weight / High Se is protective against adverse effects of Mn
Kim et al. 2013[46] / Prospective cohort / Korea / 884 / Maternal blood Cd, Pb during early and/or late pregnancy / BSID-II in 6-month olds / Gender, birth weight, maternal age at delivery, maternal education, family income, breastfeeding status, residential area / Interaction during early pregnancy, and effect modification in late pregnancy whereby Pb inversely associated with neurodevelopment at high Cd levels
Lin et al. 2013 [61] / Prospective cohort / Taiwan / 230 / Umbilical cord Mn, Pb (and As, Hg) / CDIIT in 2-year olds / Gender, maternal age, maternal education, environmental tobacco smoke, fish intake, HOME score / Lowest neurodevelopment scores observed with high Mn-Pb co-exposure
Claus Hennet al. 2012 [74] / Prospective cohort / Mexico / 455 / Blood Mn, Pb at 1 and 2 years of age / BSID-II from 1 to 3 years of age / Gender, maternal education, maternal IQ, gestational age, hemoglobin / Increased Pb toxicity among children with high Mn co-exposure
Yorifujiet al. 2011 [80] / Prospective cohort / Denmark / 896 / Hg in umbilical cord blood/maternal hair, Pb in cord blood / WISC-R, Boston Naming Test, CVLT-C at 7 and/or 14 years of age / Outcomes at 7 years: gender, age, maternal IQ, maternal and paternal employment, maternal and paternal education, daycare at age 7, medical risk, maternal smoking and alcohol use during pregnancy.
Outcomes at 14 years: gender, age, maternal IQ, maternal and paternal employment, residence, language, school grade, smoking and alcohol use / Less than additive combined effect of Hg and Pb
McDermott et al. 2011 [77] / Retrospective cohort / USA / 3988 / Soil As, Pb, Ba, Cr, Cu, Hg, Mn, Ni at maternal residence during pregnancy / ID in ~8- to 12-year olds / Gender, gestational age, birth weight, maternal age, maternal race, parity, tobacco and alcohol use, housing density, age of housing / High As-Pb co-exposure associated with higher probability of ID
Kordaset al. 2015 [76] / Cross-sectional / Uruguay / 92 / Pb in blood; As, Cd, Mn in hair / BSID-III (mental development) in 1- to 3.5-year olds / Age, hemoglobin, maternal IQ, maternal depressive symptoms, household density, HOME score, SES / No associations between clusters of metal exposure and neurodevelopment scores
Ciesielskiet al. 2012 [42] / Cross-sectional / USA (NHANES) / 2189-2196a / Urine Cd, blood Pb / Learning disability, special education utilization in 6- to 15-year olds / Urinary creatinine, age, gender, smoker in home, serum cotinine, prenatal smoke exposure, poverty income ratio / No statistical interaction between Cd and Pb observed
Khan et al. 2012 [75] / Cross-sectional / Bangladesh / 840 / Drinking water As, Mn / Academic achievement in mathematics and languages in 8- to 11-year olds / School grade, maternal and paternal education, head circumference / No statistical interaction between Mn and As observed
Lucchiniet al. 2012 [62] / Cross-sectional / Italy / 299 / Blood/hair/air/soil Mn, blood Pb / WISC-III in 11- to 14-year olds / Gender, age at testing, maternal and paternal education, SES, family size, parity, BMI, area of residence, hemoglobin, smoking status, alcohol consumption / No statistical interaction between Mn and Pb observed
Wasserman et al. 2011 [78] / Cross-sectional / Bangladesh / 299 / As in drinking water/blood/urine, Mn in drinking water/blood / WISC-IV in 8- to 11-year olds / Head circumference, maternal age, maternal intelligence, school attendance, ferritin / No interaction observed between As and Mn
Kim et al.2009[60] / Cross-sectional / Korea / 261 / Blood Mn, Pb / IQ in 8- to 11-year olds / Age, gender, maternal and paternal education, family income, maternal smoking during pregnancy, birth weight, maternal age at delivery / Additive interaction and effect modification, whereby Pb inversely associated with IQ only at high Mn levels
Behavior
Roberts et al. 2013 [86] / Nested case-control / USA / 325 cases; 22,101 controls / Modeled air concentrations of As, Cd, Cr, Hg, Mn, Ni, Pb, Sb during perinatal period / ASD, based on parent report of physician-diagnosis / Maternal grandparents’ education (SES proxy), tract median income, percent college educated, maternal age, year of birth, gender / An overall measure of metals concentrations and pooled OR were associated with increased risk of ASD
Boucher et al. 2012 [85] / Prospective cohort / Canada / 196 / Hg, Pb (and PCBs) in umbilical cord blood and blood at 9-13 years of age / Response inhibition on go/no-go task and ERPs in 9- to 13-year olds / Gender, age, whether child was adopted, whether child was transported by plane to assessment, time (AM/PM) at assessment, maternal age, SES, maternal reasoning abilities, breastfeeding, DHA fatty acid, and maternal smoking, alcohol, and marijuana use / Prenatal Pb associated with increased impulsivity that is intensified by higher Hg exposure
Ciesielskiet al. 2012 [42] / Cross-sectional / USA (NHANES) / 2,189-2,196a / Urine Cd, blood Pb / ADHD in 6- to 15-year olds / Urinary creatinine, age, gender, smoker in home, serum cotinine, prenatal smoke exposure, poverty income ratio / Suggestive, though non-significant, evidence that Cd attenuates Pb-mediated increase in ADHD
Lucchiniet al. 2012 [62] / Cross-sectional / Italy / 299 / Blood Pb, blood/hair/air/soil Mn / CASS in 11- to 14-year olds / Gender, age at testing, maternal and paternal education, SES, family size, parity, BMI, area of residence, hemoglobin, smoking status, alcohol consumption / No statistical interaction between Mn and Pb observed
Khan et al. 2011 [71] / Cross-sectional / Bangladesh / 201 / Drinking water Mn, As / CBCL-TRF in 8- to 11-year olds / Gender, maternal education, arm circumference, BMI / No statistical interaction between Mn and As observed

a Sample size differed by outcome due to missing outcome data.

ASD Autism spectrum disorders, BSID Bayley Scales of Infant Development, CASS Conners-Wells’ Adolescent Self-Report Scale

CBCL-TRF Child Behavior Check List – Teacher’s Report Form, CDIIT Comprehensive Developmental Inventory for Infants and Toddlers, CVLT-C California Verbal Learning Test for Children, DSM-IV-TR Diagnostic and Statistical Manual of the American Psychiatric Association, ERPs Event-related potentials, ID intellectual disability, IQ Intelligence quotient, LD Learning disability, MDI Mental Development Index (of BSID), NBNA Neonatal Behavioral Neurological Assessment, NHANES National Health and Nutrition Examination Survey, PCB Polychlorinated biphenyl, PDI Psychomotor Development Index (of BSID), SES socioeconomic status, WISC Wechsler Intelligence Scale for Children