ABSTRACT
This chapter discusses general analytic issues which arise in studying twins and then summarizes the major findings of twin studies of obesity. Methods for assessing twin similarity and comparing the similarity across zygosity groups vary in complexity and informativeness. Several methods have been proposed to estimate the heritability of traits. The methods for analyzing twin data described above have important limitations. There exist a number of possible genotypic and environmental effects which the classical twin design cannot detect, but which nonetheless may result in biasing heritability estimates derived from such a design. Maximum likelihood model-fitting methods were applied to the twin covariance matrices of Bodurtha and colleagues’ anthropometric data. Evidence supporting the relatively high heritability estimates obtained from the male twins in the NAS-NRC registry comes from several sources. The many achievements of twin research in obesity during the recent past provide a foretaste of future achievements.
