Comparison of standardization methods for the harmonization of phenotype data: An application to cognitive measures

Lauren E. Griffith, Edwin Van Den Heuvel, Parminder Raina, Isabel Fortier, Nazmul Sohel, Scott Hofer, Hélène Payette, Christina Wolfson, Sylvie Belleville, Meghan Kenny, Dany Doiron

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Standardization procedures are commonly used to combine phenotype data that were measured using different instruments, but there is little information on how the choice of standardization method influences pooled estimates and heterogeneity. Heterogeneity is of key importance in meta-analyses of observational studies because it affects the statistical models used and the decision of whether or not it is appropriate to calculate a pooled estimate of effect. Using 2-stage individual participant data analyses, we compared 2 common methods of standardization, T-scores and category-centered scores, to create combinable memory scores using crosssectional data from 3 Canadian population-based studies (the Canadian Study on Health and Aging (1991-1992), the Canadian Community Health Survey on Healthy Aging (2008-2009), and the Quebec Longitudinal Study on Nutrition and Aging (2004-2005)). A simulation was then conducted to assess the influence of varying the following items across population-based studies: 1) effect size, 2) distribution of confounders, and 3) the relationship between confounders and the outcome. We found that pooled estimates based on the unadjusted category-centered scores tended to be larger than those based on the T-scores, although the differences were negligible when adjusted scores were used, and that most individual participant data meta-analyses identified significant heterogeneity. The results of the simulation suggested that in terms of heterogeneity, the method of standardization played a smaller role than did different effect sizes across populations and differential confounding of the outcome measure across studies. Although there was general consistency between the 2 types of standardization methods, the simulations identified a number of sources of heterogeneity, some of which are not the usual sources considered by researchers.

Original languageEnglish (US)
Pages (from-to)770-778
Number of pages9
JournalAmerican Journal of Epidemiology
Volume184
Issue number10
DOIs
StatePublished - 2016
Externally publishedYes

Fingerprint

Phenotype
Meta-Analysis
Quebec
Statistical Models
Population Density
Health Surveys
Population
Observational Studies
Longitudinal Studies
Research Personnel
Outcome Assessment (Health Care)
Health

Keywords

  • Cognition
  • Harmonization
  • Individual participant data
  • Meta-analysis
  • Standardization

ASJC Scopus subject areas

  • Epidemiology

Cite this

Comparison of standardization methods for the harmonization of phenotype data : An application to cognitive measures. / Griffith, Lauren E.; Van Den Heuvel, Edwin; Raina, Parminder; Fortier, Isabel; Sohel, Nazmul; Hofer, Scott; Payette, Hélène; Wolfson, Christina; Belleville, Sylvie; Kenny, Meghan; Doiron, Dany.

In: American Journal of Epidemiology, Vol. 184, No. 10, 2016, p. 770-778.

Research output: Contribution to journalArticle

Griffith, LE, Van Den Heuvel, E, Raina, P, Fortier, I, Sohel, N, Hofer, S, Payette, H, Wolfson, C, Belleville, S, Kenny, M & Doiron, D 2016, 'Comparison of standardization methods for the harmonization of phenotype data: An application to cognitive measures', American Journal of Epidemiology, vol. 184, no. 10, pp. 770-778. https://doi.org/10.1093/aje/kww098
Griffith, Lauren E. ; Van Den Heuvel, Edwin ; Raina, Parminder ; Fortier, Isabel ; Sohel, Nazmul ; Hofer, Scott ; Payette, Hélène ; Wolfson, Christina ; Belleville, Sylvie ; Kenny, Meghan ; Doiron, Dany. / Comparison of standardization methods for the harmonization of phenotype data : An application to cognitive measures. In: American Journal of Epidemiology. 2016 ; Vol. 184, No. 10. pp. 770-778.
@article{fc65318ed30e46f0ba99025a58459146,
title = "Comparison of standardization methods for the harmonization of phenotype data: An application to cognitive measures",
abstract = "Standardization procedures are commonly used to combine phenotype data that were measured using different instruments, but there is little information on how the choice of standardization method influences pooled estimates and heterogeneity. Heterogeneity is of key importance in meta-analyses of observational studies because it affects the statistical models used and the decision of whether or not it is appropriate to calculate a pooled estimate of effect. Using 2-stage individual participant data analyses, we compared 2 common methods of standardization, T-scores and category-centered scores, to create combinable memory scores using crosssectional data from 3 Canadian population-based studies (the Canadian Study on Health and Aging (1991-1992), the Canadian Community Health Survey on Healthy Aging (2008-2009), and the Quebec Longitudinal Study on Nutrition and Aging (2004-2005)). A simulation was then conducted to assess the influence of varying the following items across population-based studies: 1) effect size, 2) distribution of confounders, and 3) the relationship between confounders and the outcome. We found that pooled estimates based on the unadjusted category-centered scores tended to be larger than those based on the T-scores, although the differences were negligible when adjusted scores were used, and that most individual participant data meta-analyses identified significant heterogeneity. The results of the simulation suggested that in terms of heterogeneity, the method of standardization played a smaller role than did different effect sizes across populations and differential confounding of the outcome measure across studies. Although there was general consistency between the 2 types of standardization methods, the simulations identified a number of sources of heterogeneity, some of which are not the usual sources considered by researchers.",
keywords = "Cognition, Harmonization, Individual participant data, Meta-analysis, Standardization",
author = "Griffith, {Lauren E.} and {Van Den Heuvel}, Edwin and Parminder Raina and Isabel Fortier and Nazmul Sohel and Scott Hofer and H{\'e}l{\`e}ne Payette and Christina Wolfson and Sylvie Belleville and Meghan Kenny and Dany Doiron",
year = "2016",
doi = "10.1093/aje/kww098",
language = "English (US)",
volume = "184",
pages = "770--778",
journal = "American Journal of Epidemiology",
issn = "0002-9262",
publisher = "Oxford University Press",
number = "10",

}

TY - JOUR

T1 - Comparison of standardization methods for the harmonization of phenotype data

T2 - An application to cognitive measures

AU - Griffith, Lauren E.

AU - Van Den Heuvel, Edwin

AU - Raina, Parminder

AU - Fortier, Isabel

AU - Sohel, Nazmul

AU - Hofer, Scott

AU - Payette, Hélène

AU - Wolfson, Christina

AU - Belleville, Sylvie

AU - Kenny, Meghan

AU - Doiron, Dany

PY - 2016

Y1 - 2016

N2 - Standardization procedures are commonly used to combine phenotype data that were measured using different instruments, but there is little information on how the choice of standardization method influences pooled estimates and heterogeneity. Heterogeneity is of key importance in meta-analyses of observational studies because it affects the statistical models used and the decision of whether or not it is appropriate to calculate a pooled estimate of effect. Using 2-stage individual participant data analyses, we compared 2 common methods of standardization, T-scores and category-centered scores, to create combinable memory scores using crosssectional data from 3 Canadian population-based studies (the Canadian Study on Health and Aging (1991-1992), the Canadian Community Health Survey on Healthy Aging (2008-2009), and the Quebec Longitudinal Study on Nutrition and Aging (2004-2005)). A simulation was then conducted to assess the influence of varying the following items across population-based studies: 1) effect size, 2) distribution of confounders, and 3) the relationship between confounders and the outcome. We found that pooled estimates based on the unadjusted category-centered scores tended to be larger than those based on the T-scores, although the differences were negligible when adjusted scores were used, and that most individual participant data meta-analyses identified significant heterogeneity. The results of the simulation suggested that in terms of heterogeneity, the method of standardization played a smaller role than did different effect sizes across populations and differential confounding of the outcome measure across studies. Although there was general consistency between the 2 types of standardization methods, the simulations identified a number of sources of heterogeneity, some of which are not the usual sources considered by researchers.

AB - Standardization procedures are commonly used to combine phenotype data that were measured using different instruments, but there is little information on how the choice of standardization method influences pooled estimates and heterogeneity. Heterogeneity is of key importance in meta-analyses of observational studies because it affects the statistical models used and the decision of whether or not it is appropriate to calculate a pooled estimate of effect. Using 2-stage individual participant data analyses, we compared 2 common methods of standardization, T-scores and category-centered scores, to create combinable memory scores using crosssectional data from 3 Canadian population-based studies (the Canadian Study on Health and Aging (1991-1992), the Canadian Community Health Survey on Healthy Aging (2008-2009), and the Quebec Longitudinal Study on Nutrition and Aging (2004-2005)). A simulation was then conducted to assess the influence of varying the following items across population-based studies: 1) effect size, 2) distribution of confounders, and 3) the relationship between confounders and the outcome. We found that pooled estimates based on the unadjusted category-centered scores tended to be larger than those based on the T-scores, although the differences were negligible when adjusted scores were used, and that most individual participant data meta-analyses identified significant heterogeneity. The results of the simulation suggested that in terms of heterogeneity, the method of standardization played a smaller role than did different effect sizes across populations and differential confounding of the outcome measure across studies. Although there was general consistency between the 2 types of standardization methods, the simulations identified a number of sources of heterogeneity, some of which are not the usual sources considered by researchers.

KW - Cognition

KW - Harmonization

KW - Individual participant data

KW - Meta-analysis

KW - Standardization

UR - http://www.scopus.com/inward/record.url?scp=85014828430&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85014828430&partnerID=8YFLogxK

U2 - 10.1093/aje/kww098

DO - 10.1093/aje/kww098

M3 - Article

C2 - 27769990

AN - SCOPUS:85014828430

VL - 184

SP - 770

EP - 778

JO - American Journal of Epidemiology

JF - American Journal of Epidemiology

SN - 0002-9262

IS - 10

ER -