A Critical Review of Statistical Methods for Twin Studies Relating Exposure to Early Life Health Conditions

Abstract

When investigating disease etiology, twin data provide a unique opportunity to control for confounding and disentangling the role of the human genome and exposome. However, using appropriate statistical methods is fundamental for exploiting such potential. We aimed to critically review the statistical approaches used in twin studies relating exposure to early life health conditions. We searched PubMed, Scopus, Web of Science, and Embase (2011–2021). We identified 32 studies and nine classes of methods. Five were conditional approaches (within-pair analyses): additive-common-erratic (ACE) models (11 studies), generalized linear mixed models (GLMMs, five studies), generalized linear models (GLMs) with fixed pair effects (four studies), within-pair difference analyses (three studies), and paired-sample tests (two studies). Four were marginal approaches (unpaired analyses): generalized estimating equations (GEE) models (five studies), GLMs with cluster-robust standard errors (six studies), GLMs (one study), and independent-sample tests (one study). ACE models are suitable for assessing heritability but require adaptations for binary outcomes and repeated measurements. Conditional models can adjust by design for shared confounders, and GLMMs are suitable for repeated measurements. Marginal models may lead to invalid inference. By highlighting the strengths and limitations of commonly applied statistical methods, this review may be helpful for researchers using twin designs.

Genetic and environmental influences on infant anthropometry at birth and four months of life: evidence from singleton and twin data in the HEALS and earlyFOOD projects

“…Another strength is to have considered both singleton and twin data, using traditional and modern approaches at the same time. Indeed, whereas singleton studies are suitable for studying the effects of individual environmental exposures (using conventional regression approaches) or individual genes (using GWAS) on developmental outcomes separately, twin studies allow studying the relative contributions of the whole genome and exposome (7).…”

“…Twins share several genes, as well as the prenatal and postnatal environment, and provide a unique opportunity to disentangle the relative contributions of the genome and the exposome (i.e., the totality of the exposures experienced by an individual throughout life) (6) in explaining inter-individual differences in developmental outcomes (7). Indeed, previous studies have reported considerable geographical differences in the relative contribution of environmental factors, mainly due to marked geographical differences in the inter-individual variation of maternal dietary habits and other family factors (8).…”

Genetic and Environmental Effects on the Development of White Matter Hyperintensities in a Middle Age Twin Population

“…These variables include additive genetic (A), common (or shared) environmental (C) and unshared (or unique) environmental (E) components. The random variables are expected to be mutually independent within a single twin and to follow a conventional normal distribution within the same twin [ 24 ]. For the ACE analysis, version 2.19.5 of the OpenMx package for structural equations and other statistical modeling has been used under version 3.6.3 of the R programming language [ 25 , 26 ].…”

Exposome and pharmacogenomics: towards precision medicine in childhood asthma

“…In 2014, the "Health and Environment-wide Associations based on Large population Surveys" (HEALS) project (FP7-ENV-2013-603946 http://www.heals-eu. eu/) was funded by the European Commission with the aim to describe in a series of about 335,000 individuals the internal exposome by integrating omics and biomonitoring data through population studies taking into account different levels of environmental exposure, age-specific windows of exposure, and genetic variability (17)(18)(19). Notably, the HELIX project (https://www.…”