European Journal of Public Health · 2012 · 11 citations · 33 references
Life SciencesGeneticsBehavioral GeneticsGenetic EpidemiologyHuman GeneticsComplex Trait GeneticsGenetic CorrelationGenetic RiskGenetic Risk FactorsHeritabilityTwin StudiesBivariate Twin AnalysisBlood PressureCholesky DecompositionPrevalence EstimationStructural Equation Modeling In Twin ResearchSystolic Blood PressureTwin Study HeritabilityYoung Adulthood
BACKGROUND: Lower educated people have a higher prevalence of metabolic risk factors (MRF), that is, high waist circumference (WC), high systolic blood pressure, low high-density lipoprotein cholesterol level, high triglycerides and high fasting glucose levels. Behavioural and psychosocial factors cannot fully explain this educational gradient. We aim to examine the possible role of genetic factors by estimating the extent to which education and MRF share a genetic basis and the extent to which the heritability of MRF varies across educational levels. METHODS: We examined 388 twin pairs, aged 18-34 years, from the Belgian East Flanders Prospective Twin Survey. Using structural equation modelling, a Cholesky bivariate model was applied to assess the shared genetic basis between education and MRF. The heritability of MRF across education levels was estimated using a non-linear multivariate Gaussian regression. RESULTS: Fifteen percent (P < 0.01) of the negative relation between education and WC was because of genes shared between these two traits. Furthermore, the heritability of WC was lower in the lowest educated group (65%) compared with the highest educated group (78%, P = 0.04). The lower heritabilities among the lower educated twins for the other MRF were not significant. The heritability of glucose was higher in the lowest education (80%) group compared with the high education group (67%, P = 0.01). CONCLUSION: Our findings suggest that genetic factors partly explain educational differences in WC. Furthermore, the lower heritability estimates in WC in the lower educated young adults suggest opportunities for environmental interventions to prevent the development of full-blown metabolic syndrome in middle and older age.
33
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10
The metabolic syndrome—a new worldwide definition
K. G. M. M. Alberti, Paul Zimmet, Jonathan E. Shaw · The Lancet · 2005 · 9.3K citations
Chronic stress at work and the metabolic syndrome: prospective study
Tarani Chandola, Eric J. Brunner, Michael Marmot · BMJ · 2006 · 1.1K citations · Full text