Saunders Catherine L, Gulliford Martin C
Department of Public Health Sciences, King's College London, Capital House, 42 Weston Street, London SE1 3QD, UK.
J Clin Epidemiol. 2006 Nov;59(11):1191-8. doi: 10.1016/j.jclinepi.2006.02.015. Epub 2006 Jul 31.
The relative contributions of genetic and environmental variables to within-household clustering of quantitative traits in household surveys are poorly characterized. We estimated shared genetic and shared environmental contributions to within-household correlation for anthropometric variables and cardiovascular disease risk factors.
Data were analyzed for the Health Survey for England 1998, a representative national household survey. Two-generation pedigrees were defined using information for relationships within households. After standardizing for age and sex, data were analyzed for 11 quantitative traits. Variance components models were fitted to estimate the proportion of variance due to additive genetic variance or shared environmental effects.
Within-household correlation coefficients for all related and unrelated subjects ranged from 0.10 for C-reactive protein to 0.31 for height. Pairwise correlations between related individuals within households were consistently higher than those between unrelated individuals. Estimated heritability ranged from 6% for diastolic blood pressure to 40% for serum cholesterol. The proportion of variance attributable to shared environmental effects ranged from 8% for cholesterol to 24% for height.
In this large, representative national sample of generally small families, estimates for heritability were generally lower than previously reported, whereas the contribution of shared environment and individual-level variation were greater.
