Supplement 1.Studies assessing exposure to cadmium and neurodevelopmental/cognitive or behavioral outcomes.

1

Authors & Year / Study design / Location / Sample size / Exposure Measure [Reported level] / Outcome Measure / Confounders / Major Finding
Neurodevelopment / Cognition
Jeonget al. 2015[45] / Prospective cohort / South Korea / 119 / Maternal blood Cd in “early pregnancy” (~20th week)
[Mean: 1.15 μg/L] / FSIQ, VIQ, PIQ (WPPSI) at 5 yrs / Child’s gender, parents’ education, family income, and maternal BMI. / Significant inverse association with PIQ.
Fornset al. 2014 [43] / Prospective cohort / Spain / 385 / Maternal urine Cd
[Median: 9.08 and 11.52μg/L at
1st and 3rd trimester] / MSCA at 4 yrs / Child’s age, quality of test, gender, maternal PIQ and VIQ, SES, country of birth, maternal mental health, maternal age, child’s mood changes, and child’s diagnosis of neuropsychological disorder. / Inverse but not significant association with 1st and 3rd trimester exposure and cognitive score of MSCA.
Kim Y. et al. 2013[46] / Prospective cohort / South Korea / 884 / Prenatal blood Cd (“early” <20th week and “late” near deliverya)
[GMearly: 1.42 μg/L, Range 0.03-9.87 μg/L] / BSID-II at 6 mo / Child’s gender, birth weight, maternal age, maternal education level, family income, breastfeeding status and residential area. / No significant association.
Kippleret al. 2012[47] / Prospective cohort / Bangladesh / 1,305 / Urine Cd (1st trimester and child at 5 yr)
[Mean: 0.63 μg/L at 1st trimester, 0.22μg/L at 5 yrs] / FSIQ, VIQ, PIQ (WPPSI) at 5 yrs / Child’s age, gender, tester (model 1) as well as birth weight, birth order, height-for-age, HOME, maternal BMI, maternal IQ, SES (model 2) and joint maternal and concurrent urine Cd (model 3) / Significant inverse association between maternal and concurrent urines with VIQ, PIQ, and FSIQ.
Tianet al. 2009 [49] / Prospective cohort / China / 106 / Cord blood Cd
[Median: 0.06μg/L, Range 0.02-1.78μg/L] / FSIQ, VIQ, PIQ (WPPSI-R) at 4.5 yrs / Cord blood lead, maternal age, height, and weight, gravidity, gestational age, method of delivery, breastfeeding, SES, environmental tobacco, annual income, and nursery school age. / Significant inverse association with PIQ and FSIQ.
Cao et al. 2009[41] / Nested cohortwithin RCTa / USA / 441 / Blood Cd at 2 yrs,
[GM (95%CI): 0.21 (0.20–0.22)μg/La] / BSID-II at 2 yrs, IQ (WPPSI-R)at 5 yrs and IQ (WISC-III) at 7 yrs / Treatment group, age, caregiver’s IQ, center location, single parent, language, race, gender, parent’s employment status, parents’ education, and concurrent blood lead level. / No significant associations. Suggestive trends of an inverse relationship with IQ at ages 5 and 7.
Al-Saleh et al. 2014 [40] / Cross-sectional / Saudi Arabia / 1,578 / Cord blood Cd, maternal blood (at delivery), and placenta Cd
[Mediancord: 0.70 μg/L] / 1- and 5-minute Apgar score / Maternal age, parity (model 1) as well as gestational age (model 2), and preterm births excluded (model 3) / Significant inverse association between cord blood Cd and 5-min Apgar score. No association with 1-min Apgar score.
Rodriguez-Barrancoet al. 2014 [48] / Cross-sectional / Spain / 261 / Urine and hair Cd
[GMurine: 0.75 μg/g creatinine, GMhair: 0.01 μg/g] / IQ (WISC-IV) at 6-9 yrs / Child’s gender, age, BMI, mother’s age, IQ, and education, monthly family income, family status, gestational age, vegetables and cereal intake, and levels of Mn, As, Pb, Hg in urine. / Significant inverse association between urine Cd and verbal comprehension (boys and girls) and FSIQ in boys only.
Ciesielskiet al.2013 [42] / Cross-sectional / USA
(NHANES) / 2,189-2,196 / Urine Cd
[Median: 0.11μg/L,
Range: 0-14.94 μg/L] / LD and special education assessed by parental report at 6-15 yrs / Urinary creatinine (model 1) as well as child’s age, gender, blood lead, smoker in the home, serum cotinine, prenatal smoke exposure, and poverty income ratio (model 2). / Increasing Cd was significantly associated with increased LD and special education.
Garcia-Esquinaset al. 2013 [44] / Cross-sectional / Spain / 112 / Cord blood Cd
[Median: 0.29μg/L,
GM: 0.27μg/L] / 1- and 5-minute Apgar score / Child’s gender, gestational age, maternal age, cigarette smoking. / Significant inverse association with high (>median) cord blood Cd and lower 1- and 5-min Apgar scores.
Behavior
Fornset al. 2014 [43] / Prospective cohort / Spain / 385 / Urine Cd
[Median: 9.08 and 11.52μg/L at
1st and 3rd trimester] / ADHD-DSM-IV at 4 yrs / Child’s age, maternal social class, and maternal mental health. / No significant association.
Sieon et al. 2013[52] / Prospective cohort / Belgium / 257 / Cord blood Cd
[Mean: 0.22 μg/L] / SDQ (emotion, conduct, hyperactivity, peer, and social problem domains) at 7-8 yrs / Child’s gender, maternal BMI, age, pregnancy weight gain, smoking during pregnancy, paternal BMI, current smoking by either parent, grandmother smoking before mother’s birth, parents’ education level, and serious infections since birth. / Increasing Cd was significantly associated with an increased risk of emotional problems (boys only).
Cao et al. 2009[41] / Nested cohortwithin RCTa / USA / 441 / Blood Cd at 2 yrs,
[GM (95%CI): 0.21 (0.20–0.22)μg/La] / NEPSY and CPRS-R at 5 yrs, NEPSY, CPT, CVLTC, WLPB-R, NEES, BASC-TRS, and BASC-PRS at 7 yrs / Treatment group, age, caregiver’s IQ, center location, single parent, language, race, child’s gender, parent’s employment status, parents’ education, and concurrent blood lead level. / No significant associations. Suggestive trends of an association between increasing Cd and behavioral problems.
Rodriguez-Barrancoet al. 2014 [48] / Cross-sectional / Spain / 261 / Urine and hair Cd [GMurine: 0.75 μg/g creatinine, GMhair: 0.01 μg/g] / Three computerized BARS behavioral tests (RTT, CPT, and SAT) at 6-9 yrs / Child’s gender, age, BMI, mother’s age and education, monthly family income, family status, gestational age, vegetables and cereal intake, and levels of Mn, As, Pb, Hg in urine (SAT model) as well as maternal IQ (RTT model), maternal occupation, residence area, and child’s weight and height at birth (CPT model) / No significant association.
Ciesielskiet al. 2013[42] / Cross-sectional / USA
(NHANES) / 2,189-2,196 / Urine Cd
[Median: 0.11μg/L,
Range: 0-14.94 μg/L] / ADHD assessed by parental report at 6-15 yrs / Urinary creatinine (model 1) as well as child’s age, gender, blood lead, smoker in the home, serum cotinine, prenatal smoke exposure, and poverty income ratio (model 2). / No significant association.
Szkup-Jablonskaet al. 2012[53] / Cross-sectional / Poland / 78 / Blood Cd
[Mean: 0.22 μg/L, Range: 0.05-2.91μg/L] / Parent-ADHD-rating scale-IV at yrs 2-18 / None reported. / No significant association.
Baoet al. 2009[50] / Cross-sectional / China / 549 / Hair Cd
[Median: 7.33 μg/g] / CBCL at 7-16 yrs / Child’s gender, age, income, and parents’ education. / Increasing Cd was significantly associated with increased withdrawn, social & attention problems
Kim S.et al. 2013[51] / Case-control / USA / 129 (71/58) / Blood Cd
[GM: 0.02, Range: 0.0007-8.69μg/L a] / ADHD-DSM-IV at 5-12 yrs / Age group, child’s gender, race, location, maternal smoking and alcohol use, SES, and environmental tobacco smoke. / No significant association.
Yousef et al. 2011 [54] / Matched case-control / UAE / 92 / Blood Cd
[Not reported; Range: ~0 – 0.03μg/L] / ADHD (STI-CRS) and at 5-15 yrs / None reported. / No significant association.

aSee manuscript for detailed information.GM Geometric mean, CI confidence interval.

ADHD attention deficit hyperactivity disorder, ADHD-DSM ADHD Diagnostic and Statistical Manual of Mental Disorders,

BARS Behavioral Assessment Research System, BASC-PRS parent rating scale, BASC-TRS Behavioral Assessment System for Children – Teacher rating scale, BSID Bayley Scales of Infant Development,CBCL Child behavior check-list, CPRS-RConners’ Parent Rating Scale-Revised, CPT Continuous Performance Test, CVLT California Verbal Learning Test, CVLTC California Verbal Learning Test for Children, FSIQ Full scale IQ, HOME Home Observation for Measurement of the Environment, IQ intelligence quotient, LD Learning disability,MSCA McCarthy Scales of Children’s Abilities, NEES Neurologic Examination for Subtle Signs, NEPSY Developmental Neuropsychological Assessment, NHANES National Health and Nutrition Examination Survey, PIQ Performance IQ, RCT Randomized Clinical Trial, RTT Reaction Time Test, SAT Selective Attention Test, SDQ Strengths and Difficulties Questionnaire, SES socioeconomic status, STI-CRS Short Ten-Item Conners’ Rating Scale, VIQ Verbal IQ, WISC Wechsler Intelligence Scale for Children,WLPB-R Woodcock Language Proficiency Battery-Revised, WPPSI-R Wechsler Preschool and Primary Scale of Intelligence Revised Edition.

1