Quantifying the potential impact of measurement error in an investigation of autism spectrum disorder (ASD)

Karyn Heavner, Craig Newschaffer, Irva Hertz-Picciotto, Deborah H Bennett, Igor Burstyn

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The Early Autism Risk Longitudinal Investigation (EARLI),an ongoing study of a risk-enriched pregnancy cohort, examines genetic and environmental risk factors for autism spectrum disorders (ASDs). We simulated the potential effects of both measurement error (ME) in exposures and misclassification of ASD-related phenotype (assessed as Autism Observation Scale for Infants (AOSI)scores) on measures of association generated under this study design. We investigated the impact on the power to detect true associations with exposure and the false positive rate (FPR) for a non-causal correlate of exposure (X2, r=0.7) for continuous AOSI score (linear model) versus dichotomised AOSI (logistic regression) when the sample size (n), degree of ME in exposure, and strength of the expected (true) OR (eOR)) between exposure and AOSI varied. Exposure was a continuous variable in all linear models and dichotomised at one SD above the mean in logistic models. Simulations reveal complex patterns and suggest that: (1) There was attenuation of associations that increased with eOR and ME; (2) The FPR was considerable under many scenarios; and (3) The FPR has a complex dependence on the eOR, ME and model choice, but was greater for logistic models. The findings will stimulate work examining cost-effective strategies to reduce the impact of ME in realistic sample sizes and affirm the importance for EARLI of investment in biological samples that help precisely quantify a wide range of environmental exposures.

Original languageEnglish (US)
Pages (from-to)438-445
Number of pages8
JournalJournal of Epidemiology and Community Health
Volume68
Issue number5
DOIs
StatePublished - 2014

Fingerprint

Autistic Disorder
Observation
Logistic Models
Sample Size
Linear Models
Environmental Exposure
Autism Spectrum Disorder
Phenotype
Costs and Cost Analysis
Pregnancy

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health
  • Epidemiology

Cite this

Quantifying the potential impact of measurement error in an investigation of autism spectrum disorder (ASD). / Heavner, Karyn; Newschaffer, Craig; Hertz-Picciotto, Irva; Bennett, Deborah H; Burstyn, Igor.

In: Journal of Epidemiology and Community Health, Vol. 68, No. 5, 2014, p. 438-445.

Research output: Contribution to journalArticle

@article{5ab0eff42ccb427199a1407b85e1ff48,
title = "Quantifying the potential impact of measurement error in an investigation of autism spectrum disorder (ASD)",
abstract = "The Early Autism Risk Longitudinal Investigation (EARLI),an ongoing study of a risk-enriched pregnancy cohort, examines genetic and environmental risk factors for autism spectrum disorders (ASDs). We simulated the potential effects of both measurement error (ME) in exposures and misclassification of ASD-related phenotype (assessed as Autism Observation Scale for Infants (AOSI)scores) on measures of association generated under this study design. We investigated the impact on the power to detect true associations with exposure and the false positive rate (FPR) for a non-causal correlate of exposure (X2, r=0.7) for continuous AOSI score (linear model) versus dichotomised AOSI (logistic regression) when the sample size (n), degree of ME in exposure, and strength of the expected (true) OR (eOR)) between exposure and AOSI varied. Exposure was a continuous variable in all linear models and dichotomised at one SD above the mean in logistic models. Simulations reveal complex patterns and suggest that: (1) There was attenuation of associations that increased with eOR and ME; (2) The FPR was considerable under many scenarios; and (3) The FPR has a complex dependence on the eOR, ME and model choice, but was greater for logistic models. The findings will stimulate work examining cost-effective strategies to reduce the impact of ME in realistic sample sizes and affirm the importance for EARLI of investment in biological samples that help precisely quantify a wide range of environmental exposures.",
author = "Karyn Heavner and Craig Newschaffer and Irva Hertz-Picciotto and Bennett, {Deborah H} and Igor Burstyn",
year = "2014",
doi = "10.1136/jech-2013-202982",
language = "English (US)",
volume = "68",
pages = "438--445",
journal = "Journal of Epidemiology and Community Health",
issn = "0143-005X",
publisher = "BMJ Publishing Group",
number = "5",

}

TY - JOUR

T1 - Quantifying the potential impact of measurement error in an investigation of autism spectrum disorder (ASD)

AU - Heavner, Karyn

AU - Newschaffer, Craig

AU - Hertz-Picciotto, Irva

AU - Bennett, Deborah H

AU - Burstyn, Igor

PY - 2014

Y1 - 2014

N2 - The Early Autism Risk Longitudinal Investigation (EARLI),an ongoing study of a risk-enriched pregnancy cohort, examines genetic and environmental risk factors for autism spectrum disorders (ASDs). We simulated the potential effects of both measurement error (ME) in exposures and misclassification of ASD-related phenotype (assessed as Autism Observation Scale for Infants (AOSI)scores) on measures of association generated under this study design. We investigated the impact on the power to detect true associations with exposure and the false positive rate (FPR) for a non-causal correlate of exposure (X2, r=0.7) for continuous AOSI score (linear model) versus dichotomised AOSI (logistic regression) when the sample size (n), degree of ME in exposure, and strength of the expected (true) OR (eOR)) between exposure and AOSI varied. Exposure was a continuous variable in all linear models and dichotomised at one SD above the mean in logistic models. Simulations reveal complex patterns and suggest that: (1) There was attenuation of associations that increased with eOR and ME; (2) The FPR was considerable under many scenarios; and (3) The FPR has a complex dependence on the eOR, ME and model choice, but was greater for logistic models. The findings will stimulate work examining cost-effective strategies to reduce the impact of ME in realistic sample sizes and affirm the importance for EARLI of investment in biological samples that help precisely quantify a wide range of environmental exposures.

AB - The Early Autism Risk Longitudinal Investigation (EARLI),an ongoing study of a risk-enriched pregnancy cohort, examines genetic and environmental risk factors for autism spectrum disorders (ASDs). We simulated the potential effects of both measurement error (ME) in exposures and misclassification of ASD-related phenotype (assessed as Autism Observation Scale for Infants (AOSI)scores) on measures of association generated under this study design. We investigated the impact on the power to detect true associations with exposure and the false positive rate (FPR) for a non-causal correlate of exposure (X2, r=0.7) for continuous AOSI score (linear model) versus dichotomised AOSI (logistic regression) when the sample size (n), degree of ME in exposure, and strength of the expected (true) OR (eOR)) between exposure and AOSI varied. Exposure was a continuous variable in all linear models and dichotomised at one SD above the mean in logistic models. Simulations reveal complex patterns and suggest that: (1) There was attenuation of associations that increased with eOR and ME; (2) The FPR was considerable under many scenarios; and (3) The FPR has a complex dependence on the eOR, ME and model choice, but was greater for logistic models. The findings will stimulate work examining cost-effective strategies to reduce the impact of ME in realistic sample sizes and affirm the importance for EARLI of investment in biological samples that help precisely quantify a wide range of environmental exposures.

UR - http://www.scopus.com/inward/record.url?scp=84899490776&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899490776&partnerID=8YFLogxK

U2 - 10.1136/jech-2013-202982

DO - 10.1136/jech-2013-202982

M3 - Article

C2 - 24470431

AN - SCOPUS:84899490776

VL - 68

SP - 438

EP - 445

JO - Journal of Epidemiology and Community Health

JF - Journal of Epidemiology and Community Health

SN - 0143-005X

IS - 5

ER -