Analysis of nonlinear associations between prenatal methylmercury exposure from fish consumption and neurodevelopmental outcomes in the Seychelles Main Cohort at 17 years

Li-Shan Huang, Deborah A. Cory-Slechta, Christopher Cox, Sally W. Thurston, Conrad F. Shamlaye, Gene E. Watson, Edwin van Wijngaarden, Grazyna Zareba, JJ Strain, Gary J. Myers, Philip W. Davidson

Research output: Contribution to journalArticle

Abstract

Background: The Seychelles Child Development Study has been examining the relationship between prenatal methylmercury (MeHg) exposure from consuming fish during pregnancy and child development. This study re-analyzes seven outcomes in the 17 year Main Cohort data to determine if there are nonlinear or non-homogeneous (subgroup) associations that were not identified in the linear analysis.
Methods: We adopted two statistical approaches. First, we carried out an additive nonlinear analysis assuming homogeneous prenatal MeHg-outcome relationships to explore overall associations. Second, we applied the regression tree to the Woodcock-Johnson Calculation subtest (it was significantly associated in earlier analyses) and identified 4 clusters based on covariates. Then we used additive models to assess the prenatal MeHg association in each of the four clusters for all seven outcomes. This approach assumes nonlinear associations in each cluster and non-homogeneous associations between clusters.
Results: The additive nonlinear analysis yielded prenatal MeHg curves similar to the linear analysis. For the regression tree analysis, the curves relating prenatal MeHg to outcomes between the 4 clusters differed and some crossed at higher prenatal MeHg levels, suggesting non-homogeneity in the upper range of exposure. Additionally, some of the curves suggested a possible non-linear relationship within the range of exposure we studied.
Conclusion: This non-linear analysis supports the findings from the linear analysis. It shows little evidence to support an adverse association of prenatal MeHg exposure through maternal consumption of fish contaminated with natural background levels. However, the tree analysis suggests that the prenatal exposure/outcome relationship may not be homogeneous across all individuals and that some subpopulations may have an adverse association in the upper range of the exposures studied. More robust data in the higher levels of exposure in this cohort are needed to confirm this finding.
LanguageEnglish
Pages893-904
JournalStochastic Environmental Research and Risk Assessment
Volume32
Issue number4
DOIs
Publication statusAccepted/In press - 10 Mar 2018

Fingerprint

Seychelles
Child Development
Fishes
Maternal Exposure
Regression Analysis
Pregnancy

Keywords

  • Child development
  • Methylmercury
  • Prenatal exposure
  • Regression tree

Cite this

Huang, Li-Shan ; Cory-Slechta, Deborah A. ; Cox, Christopher ; Thurston, Sally W. ; Shamlaye, Conrad F. ; Watson, Gene E. ; van Wijngaarden, Edwin ; Zareba, Grazyna ; Strain, JJ ; Myers, Gary J. ; Davidson, Philip W. / Analysis of nonlinear associations between prenatal methylmercury exposure from fish consumption and neurodevelopmental outcomes in the Seychelles Main Cohort at 17 years. In: Stochastic Environmental Research and Risk Assessment. 2018 ; Vol. 32, No. 4. pp. 893-904.
@article{ac42b8e4b8614e2cb8ddfb2d57697fd2,
title = "Analysis of nonlinear associations between prenatal methylmercury exposure from fish consumption and neurodevelopmental outcomes in the Seychelles Main Cohort at 17 years",
abstract = "Background: The Seychelles Child Development Study has been examining the relationship between prenatal methylmercury (MeHg) exposure from consuming fish during pregnancy and child development. This study re-analyzes seven outcomes in the 17 year Main Cohort data to determine if there are nonlinear or non-homogeneous (subgroup) associations that were not identified in the linear analysis. Methods: We adopted two statistical approaches. First, we carried out an additive nonlinear analysis assuming homogeneous prenatal MeHg-outcome relationships to explore overall associations. Second, we applied the regression tree to the Woodcock-Johnson Calculation subtest (it was significantly associated in earlier analyses) and identified 4 clusters based on covariates. Then we used additive models to assess the prenatal MeHg association in each of the four clusters for all seven outcomes. This approach assumes nonlinear associations in each cluster and non-homogeneous associations between clusters. Results: The additive nonlinear analysis yielded prenatal MeHg curves similar to the linear analysis. For the regression tree analysis, the curves relating prenatal MeHg to outcomes between the 4 clusters differed and some crossed at higher prenatal MeHg levels, suggesting non-homogeneity in the upper range of exposure. Additionally, some of the curves suggested a possible non-linear relationship within the range of exposure we studied. Conclusion: This non-linear analysis supports the findings from the linear analysis. It shows little evidence to support an adverse association of prenatal MeHg exposure through maternal consumption of fish contaminated with natural background levels. However, the tree analysis suggests that the prenatal exposure/outcome relationship may not be homogeneous across all individuals and that some subpopulations may have an adverse association in the upper range of the exposures studied. More robust data in the higher levels of exposure in this cohort are needed to confirm this finding.",
keywords = "Child development, Methylmercury, Prenatal exposure, Regression tree",
author = "Li-Shan Huang and Cory-Slechta, {Deborah A.} and Christopher Cox and Thurston, {Sally W.} and Shamlaye, {Conrad F.} and Watson, {Gene E.} and {van Wijngaarden}, Edwin and Grazyna Zareba and JJ Strain and Myers, {Gary J.} and Davidson, {Philip W.}",
year = "2018",
month = "3",
day = "10",
doi = "10.1007/s00477-017-1451-7",
language = "English",
volume = "32",
pages = "893--904",
journal = "Stochastic Environmental Research and Risk Assessment",
issn = "1436-3240",
number = "4",

}

Analysis of nonlinear associations between prenatal methylmercury exposure from fish consumption and neurodevelopmental outcomes in the Seychelles Main Cohort at 17 years. / Huang, Li-Shan; Cory-Slechta, Deborah A.; Cox, Christopher; Thurston, Sally W.; Shamlaye, Conrad F.; Watson, Gene E.; van Wijngaarden, Edwin; Zareba, Grazyna; Strain, JJ; Myers, Gary J.; Davidson, Philip W.

In: Stochastic Environmental Research and Risk Assessment, Vol. 32, No. 4, 10.03.2018, p. 893-904.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analysis of nonlinear associations between prenatal methylmercury exposure from fish consumption and neurodevelopmental outcomes in the Seychelles Main Cohort at 17 years

AU - Huang, Li-Shan

AU - Cory-Slechta, Deborah A.

AU - Cox, Christopher

AU - Thurston, Sally W.

AU - Shamlaye, Conrad F.

AU - Watson, Gene E.

AU - van Wijngaarden, Edwin

AU - Zareba, Grazyna

AU - Strain, JJ

AU - Myers, Gary J.

AU - Davidson, Philip W.

PY - 2018/3/10

Y1 - 2018/3/10

N2 - Background: The Seychelles Child Development Study has been examining the relationship between prenatal methylmercury (MeHg) exposure from consuming fish during pregnancy and child development. This study re-analyzes seven outcomes in the 17 year Main Cohort data to determine if there are nonlinear or non-homogeneous (subgroup) associations that were not identified in the linear analysis. Methods: We adopted two statistical approaches. First, we carried out an additive nonlinear analysis assuming homogeneous prenatal MeHg-outcome relationships to explore overall associations. Second, we applied the regression tree to the Woodcock-Johnson Calculation subtest (it was significantly associated in earlier analyses) and identified 4 clusters based on covariates. Then we used additive models to assess the prenatal MeHg association in each of the four clusters for all seven outcomes. This approach assumes nonlinear associations in each cluster and non-homogeneous associations between clusters. Results: The additive nonlinear analysis yielded prenatal MeHg curves similar to the linear analysis. For the regression tree analysis, the curves relating prenatal MeHg to outcomes between the 4 clusters differed and some crossed at higher prenatal MeHg levels, suggesting non-homogeneity in the upper range of exposure. Additionally, some of the curves suggested a possible non-linear relationship within the range of exposure we studied. Conclusion: This non-linear analysis supports the findings from the linear analysis. It shows little evidence to support an adverse association of prenatal MeHg exposure through maternal consumption of fish contaminated with natural background levels. However, the tree analysis suggests that the prenatal exposure/outcome relationship may not be homogeneous across all individuals and that some subpopulations may have an adverse association in the upper range of the exposures studied. More robust data in the higher levels of exposure in this cohort are needed to confirm this finding.

AB - Background: The Seychelles Child Development Study has been examining the relationship between prenatal methylmercury (MeHg) exposure from consuming fish during pregnancy and child development. This study re-analyzes seven outcomes in the 17 year Main Cohort data to determine if there are nonlinear or non-homogeneous (subgroup) associations that were not identified in the linear analysis. Methods: We adopted two statistical approaches. First, we carried out an additive nonlinear analysis assuming homogeneous prenatal MeHg-outcome relationships to explore overall associations. Second, we applied the regression tree to the Woodcock-Johnson Calculation subtest (it was significantly associated in earlier analyses) and identified 4 clusters based on covariates. Then we used additive models to assess the prenatal MeHg association in each of the four clusters for all seven outcomes. This approach assumes nonlinear associations in each cluster and non-homogeneous associations between clusters. Results: The additive nonlinear analysis yielded prenatal MeHg curves similar to the linear analysis. For the regression tree analysis, the curves relating prenatal MeHg to outcomes between the 4 clusters differed and some crossed at higher prenatal MeHg levels, suggesting non-homogeneity in the upper range of exposure. Additionally, some of the curves suggested a possible non-linear relationship within the range of exposure we studied. Conclusion: This non-linear analysis supports the findings from the linear analysis. It shows little evidence to support an adverse association of prenatal MeHg exposure through maternal consumption of fish contaminated with natural background levels. However, the tree analysis suggests that the prenatal exposure/outcome relationship may not be homogeneous across all individuals and that some subpopulations may have an adverse association in the upper range of the exposures studied. More robust data in the higher levels of exposure in this cohort are needed to confirm this finding.

KW - Child development

KW - Methylmercury

KW - Prenatal exposure

KW - Regression tree

U2 - 10.1007/s00477-017-1451-7

DO - 10.1007/s00477-017-1451-7

M3 - Article

VL - 32

SP - 893

EP - 904

JO - Stochastic Environmental Research and Risk Assessment

T2 - Stochastic Environmental Research and Risk Assessment

JF - Stochastic Environmental Research and Risk Assessment

SN - 1436-3240

IS - 4

ER -