Influences on Child Eating and Weight Development From a Behavioral Genetics Perspective

Tanja V. E. Kral, PhD; Myles S. Faith, PhD


J Pediatr Psychol. 2009;34(6):596-605. 

In This Article

EAH: A Behavioral Eating Trait

Perhaps the best example of how genetic and environmental factors act in concert to influence a specific eating trait linked to obesity is the study of "eating in the absence of hunger" (EAH). EAH refers to children's tendency to eat in response to the presence of palatable foods in the absence of hunger. The trait has been extensively studied both from a genetic and environmental perspective and is associated both with hyperphagia (overeating) and increased weight gain in children.

EAH in children is operationally assessed by the "free access procedure" which was developed by Birch and colleagues (Birch et al., 2003; Fisher & Birch, 1999ab). Following the consumption of a self-selected standard meal, each child rates his/her perceived hunger and fullness to ensure they are fully satiated. Each child is then asked to participate in a rank-order preference assessment of a variety of snacks with different sensory properties (e.g., potato chips, chocolate bars, and frozen yogurt) to provide him/her with the opportunity to taste each experimental snack food. Following the preference assessment each child is given toys and asked to play by him/herself. During this "play session" the child is given access to an array of sweet and savory snack foods and told that they could eat any of these foods and as much as they want. EAH refers to the number of calories the child consumes from the snack foods despite being fully satiated.

EAH in children shares behavioral characteristics with the trait of disinhibited eating (or "disinhibition") in adults. Disinhibition is the loss of self-imposed cognitive control over eating in response to external or emotional stimuli, and is the eating pattern that most consistently differentiates obese and nonobese adults (Lindroos et al., 1997). There is evidence that obese individuals show greater disinhibition than do nonobese individuals (Carmody, Brunner, & St Jeor, 1995; Westenhoefer, Stunkard, & Pudel, 1999) and that the degree of disinhibition is strongly associated with energy intake (Lawson et al., 1995; Lindroos et al., 1997), weight status and weight gain (McGuire, Wing, Klem, Lang, & Hill, 1999; Williamson et al., 1995), weight fluctuations (Carmody et al., 1995), binge eating (Howard & Porzelius, 1999), and body fat (Provencher, Drapeau, Tremblay, Despres, & Lemieux, 2003). Interestingly, maternal disinhibition has been shown to predict daughters' overweight and, when both maternal disinhibition and daughters' EAH were used to predict daughters' overweight, mothers' disinhibition showed an independent prediction (Cutting, Fisher, Grimm-Thomas, & Birch, 1999).

Thus, to the extent that EAH in childhood is a behavioral precursor to disinhibited eating in adulthood, these findings further support the importance of EAH as an early behavioral marker or risk factor for obesity onset. Indeed, there is evidence that EAH is relatively stable overtime in childhood (Birch et al., 2003), which suggests that the trait could extend into adulthood.

Genetic Influences on EAH

"Heritability" (h 2) refers to the extent to which variability in a trait is influenced by genetic variations within a population (Maes, Neale, & Eaves, 1997), such that a trait with h 2 = 0% is completely influenced by environmental factors while a trait with h 2 = 100% is entirely genetically influenced. Most traits are at least partially genetically influenced, and so heritability is useful for understanding the relative influence of genes, or the overall "genetic loading" of a trait. Heritability is most commonly estimated from twin studies that compare the phenotypic similarity of a trait in MZ twins compared to DZ twins, using "biometrical" statistical methods for heritability estimation (for details, see Neale & Cardon, 1992). There are a number of methodological issues, caveats, and limitations to twin designs. One example is the "equal environments assumption" (i.e., the assumption that the extent to which MZ twin pairs are exposed to similar environments is not different from the extent to which DZ twin pairs are exposed to similar environments). These methodological issues have been reviewed extensively by Maes and colleagues (Maes et al., 1997) and Guo (2001).

Numerous studies show that child weight status is highly genetically influenced. In a sample of 66 pairs of 3- to 17-year-old twins residing in the New York metropolitan area, the heritability of BMI and percent body fat was estimated to be 86% and 76%, respectively (Faith et al., 1999). Analyses of a population-based sample of 608 MZ and 1210 DZ twin pairs from the UK estimated the heritability of weight (corrected for height) to be 64% for boys and 61% for girls (Koeppen-Schomerus, Spinath, & Plomin, 2003). In a large national sample of adolescents from the United States, Jacobson and Rowe found that total genetic influences on BMI were 67%, 45%, and 81% for Black females, White females, and all males (i.e., all ethnic groups combined), respectively (Jacobson & Rowe, 1998). High heritability estimates for child overweight from twin studies have been shown in the literature, even in recent publications from relatively current cohorts and despite the increasing prevalence of pediatric obesity (Musani, Erickson, & Allison, 2008). Thus, despite secular changes in childhood obesity prevalence in recent decades, the prominent role of genes cannot be dismissed.

By contrast, the genetics of child eating patterns is a relatively understudied field (Faith & Keller, 2004). Studies which have examined EAH point to the importance of genetic factors on the trait. Faith and colleagues (Faith et al., 2006) found that EAH in boys who were born at high risk for obesity on the basis of maternal prepregnancy body weight was more than twice that of boys who were born at low risk for obesity. This investigation, however, could not tease apart genetic from environmental influences on the trait. Findings from a study with 801 children from 300 Hispanic families who were part of the Viva la Familia Study (Fisher, Cai et al., 2007) indicated that EAH was a highly heritable eating trait (51%).

The notion that EAH is heritable is consistent with findings from prior studies establishing the heritability of disinhibition in adults. For example, Steinle and colleagues (Steinle et al., 2002) showed that among the three eating behaviors (i.e., cognitive restraint, disinhibition, and hunger) assessed with the Eating Inventory (Stunkard & Messick, 1985), disinhibition showed the highest heritability (h 2 = 40%) and the strongest association with obesity traits. Findings from the Quebec Family Study indicated a heritability estimate for disinhibition of 17.5% as well as significant parent-offspring and spousal correlations for disinhibited eating (Provencher et al., 2005).

Environmental Influences on EAH

Parental feeding practices have been identified as one environmental factor which can considerably influence children's susceptibility to EAH. Specifically, dietary restriction, or the tendency of parents to restrict children's access to desired, palatable foods (such as snack foods), has been found to be a significant predictor of EAH. EAH was predicted by mothers' reports of using restrictive feeding practices at age 5 (Birch et al., 2003; Fisher & Birch, 2002; Francis & Birch, 2005), and was associated with increased risk of overweight in children (Fisher & Birch, 2002). Interestingly, the association between restrictive feeding patterns and child EAH depended upon maternal overweight status. That is, among daughters of overweight mothers only, greater use of restrictive feeding patterns when the children were 5 years of age predicted greater EAH by daughters when they were 7 (Birch et al., 2003; Fisher & Birch, 2002) and 9 years of age (Birch et al., 2003; Francis & Birch, 2005). However, daughters of normal-weight mothers showed significantly greater EAH only at age 5, but not at ages 7 and 9 years (Francis & Birch, 2005). In addition, those girls who were overweight at age 5 and subject to higher levels of restriction showed the greatest EAH at age 9 (Birch et al., 2003; Fisher & Birch, 2002).

In another study, Faith et al. (Faith et al., 2004) showed that maternal use of restrictive feeding patterns was associated with excess weight gain in children, but only among children who were born at high risk for obesity based on elevated maternal prepregnancy body weight. There was no association between maternal use of restrictive feeding and excess weight gain among children born at low risk for obesity. These findings suggest that the use of restrictive feeding practices by parents may interact with a child's genetic predisposition for obesity to promote excess weight gain, which may occur through greater EAH. Although Faith and colleagues did not measure EAH per se, the findings are consistent with those generated by Francis and Birch (Francis & Birch, 2005).

That being said, maternal obesity as a risk factor for overweight in their offspring may not necessarily be conferred by ways of maternal feeding practices. A study by Wardle and colleagues (Wardle, Sanderson, Guthrie, Rapoport, & Plomin, 2002) found that obese mothers did not differ from normal-weight mothers in the extent to which they offered food to deal with emotional distress, used food as a form of reward, or encouraged children to eat more than they wanted. With respect to exerting control over their child's eating, the results of this study showed that obese mothers reported significantly less control over children's intake than did normal-weight mothers (Wardle et al., 2002).

Table II depicts key findings from select investigations which further illustrate EAH's role as a behavioral eating trait in the development of childhood obesity.

Competing Conceptual Models of Factors Influencing EAH

An important area for future research will be to determine the extent to which the association between EAH and excess weight gain is mediated by genetic factors, environmental factors, or both. That is, research studies should attempt to discern if EAH and weight gain have genetic factors in common, environmental factors in common, or both genetic and environmental factors in common. Figure 1 illustrates three competing models that could be tested. Model 1 posits that the relationship between EAH and excess weight gain is entirely due to genes that influence both traits, while Model 2 posits that the association is entirely environmentally mediated. Model 3 posits that both genetic and environmental factors are influential. A properly designed study could test these competing models, as illustrated in a recent study by Fisher and colleagues (Fisher, Cai et al., 2007).

Figure 1.

Model 1: There are genetic and environmental influences on EAH and weight gain, respectively. The correlation between EAH and weight gain is due entirely to genetic pathways that impact on both traits. Model 2: There are genetic and environmental influences on EAH and weight gain, respectively. The correlation between EAH and weight gain is due entirely to environmental pathways that impact on both traits. Model 3: There are genetic and environmental influences on EAH and weight gain, respectively. The correlation between EAH and weight gain is due both to genetic and environmental pathways that impact on both traits.

It will be important to know the extent to which response to childhood obesity treatment and prevention initiatives is determined by genetic and environmental factors. Behavioral genetics research is expected to contribute to future research endeavors by, first, continuing to identify the heritability of child eating phenotypes that are associated with excess body weight, as well as by estimating genetic and environmental influences on treatment response. A range of behavioral genetics studies including twin designs, adoption designs, and (extended) family designs can address these questions (see Neale & Cardon, 1992, for details). However, it is implausible that, for example, the classic twin design, by itself, will be able to address the full range of research questions concerning the familial transmission of child eating traits. The various behavioral genetics designs have their own respective strengths and limitations that make certain designs more desirable than others for certain research questions. A more detailed discussion of these designs is provided elsewhere (Plomin, DeFries, McClearn, & McGuffin, 2000).

The value of such studies rests in their potential to resolve the extent to which the association between EAH and weight gain is due to genetic and environmental factors, respectively. If the association is primarily environmentally mediated, this would provide greater support for the potential of environmental interventions that could foster healthier eating patterns and weight trajectories among children. However, because child weight status is highly heritable, it is conceivable that obesity-promoting genes also drive increased EAH. Genetically informative designs are very powerful for resolving these questions.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.