An Examination of the Representativeness Assumption for Twin Studies of Eating Pathology and Internalizing Symptoms

Abstract

Little research has investigated whether the twin representativeness assumption (that results from twin research generalize to singletons) holds for eating pathology and internalizing symptoms. This study compared disordered eating, depression, and anxiety among young adult female twins versus singletons. Participants included 292 twins and 997 singletons in three samples. Questionnaires included the Minnesota Eating Behavior Survey, Eating Disorder Examination Questionnaire, Beck Depression Inventory, and State-Trait Anxiety Inventory. We examined mean differences between twins’ and singletons’ scores, after adjusting for age, body mass index, and ethnicity. We found statistically significant mean differences on psychopathology, with twins reporting less disordered eating and internalizing symptoms compared with singletons. Effect sizes of these mean differences were small to moderate. Our results suggest that twins report less disordered eating and internalizing symptoms than singletons, which, combined with the generally small effect sizes, indicate that results from twin samples generalize to singletons.

Clinical features associated with an increased risk for alcohol use disorders among family members.

“…Previous research on the family aggregation of AUDs, including studies that sought to identify clinical features associated with an increased familial risk, has been primarily limited by five factors: (a) the frequent use of nonrepresentative samples (i.e., clinic, referred, at-risk, twin, medical or forensic national registries) that, in turn, creates biases toward greater disorder severity or otherwise limits generalizability (Berkson, 1946; Low, Cui, & Merikangas, 2008; Munn-Chernoff et al, 2013; Røysamb & Tambs, 2016); (b) reliance on cross-sectional samples of probands rather than age-based cohorts followed longitudinally, which can be particularly problematic when evaluating clinical features related to etiology, onset, course, symptoms, and comorbidity given recall biases associated with retrospective reports (Haeny, Littlefield, & Sher, 2014; Moffitt et al, 2010); (c) the usual reliance on a single informant (often the proband) for ascertaining AUD diagnostic histories among first-degree relatives; (d) insufficient evaluations as to whether clinical features denoting family-based AUD risk are accounted for by other forms of family-based psychopathology and, consequently, nonspecific indicators of family-based AUD risk; and (e) questionable operational definitions of clinical features. As examples of this latter limitation, Milne (Milne et al, 2009) and Kendler (Kendler et al, 2016) defined “episode recurrence” as the diagnosis of AUD on two separate occasions without any demonstration of first-episode remission or recovery.…”

Phenotypic factor analysis of psychopathology reveals a new body-related transdiagnostic factor

“…Our sample comprised Finns only, hence we cannot conclude that our results would generalize to other genetic and cultural groups. Participants were twins and siblings of twins, a non-random sample of the general population, but research on external validity of twin and family studies overall support their representativeness [ 91 , 92 ]. Although we have been able to correct for the nested nature of our data, we had to assume homogeneity of family dependency across twins and their siblings to perform factorial analyses.…”

Genetic and environmental components to self‐induced vomiting

“…However, initial reports suggest no differences in genetic risk for bulimic behaviors between African Americans and European Americans 34 or mean differences between twins and singletons for disordered eating behavior. 36 Notwithstanding these limitations, this study has three important strengths including the use of a large, population-based sample and use of the CCC model. The CCC model is an improvement on previous examinations of the heritability of SIV because it takes into account the contingent nature between having ever engaged in the behavior and regularly engaging in the behavior.…”