Determinants of Binge Eating Symptoms in Children with Overweight/Obesity

Introduction

The prevalence of childhood obesity remains high despite public health and clinical efforts to prevent and treat obesity. ¹ There is also an increasing prevalence of eating disorders, which share common risk factors with obesity such as body dissatisfaction, unhealthy weight control behaviors/dieting, and weight teasing. ² Owing to these shared risk factors, there are recommendations that interventions to prevent and manage obesity should address both. ² Furthermore, children who eat in the absence of hunger, who have guilty feelings about the behavior, and try to hide their overeating may develop a binge eating disorder (BED), a condition commonly associated with obesity in adults.^3,4 Identifying the prevalence and examining the determinants of binge eating symptoms in children with obesity may have important influences on reductions in eating disorders and obesity incidence and prevalence. ⁵

The Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria for diagnosis of BED requires episodes of binge eating that are defined by at least three of the following symptoms: (1) eating more rapidly than normal, (2) eating until feeling uncomfortably full, (3) eating large amounts of food when not feeling physically hungry, (4) eating alone because of embarrassment by the amount of food consumed, and (5) feeling disgusted with oneself, depressed, or guilty after overeating. ⁶ These symptoms alone in childhood may precede the development of BED. Among individuals who do not meet diagnostic criteria for an eating disorder, the presence of binge eating and loss of control (LOC) eating still yields risk for excess body weight, elevated eating disorder psychopathology, psychosocial distress, and clinical impairment. ⁷ Thus, detection of these symptoms before the development and diagnosis of eating disorders may provide an avenue for prevention.

The prevalence of binge eating among adults with obesity is nearly twice as high as the general population ⁸ and evidence suggests that childhood obesity is associated with a history of eating disorders, in particular BED, at a later age. ⁹ For example, a recent meta-analysis of 36 articles indicated that binge eating was prevalent among more than one quarter of children and adolescents (age range 5–21 years) with overweight and obesity. ¹⁰ Although adults with obesity show a high prevalence of binge episodes, in the case of children, LOC episodes are prevalent. ¹¹ In addition, problematic binge eating symptoms in childhood seem to be related to the development of obesity in adolescence.^9,12

Recent studies suggest that shame feelings, particularly those related to body image, play an important role in the development of binge eating. These shame feelings may have their roots in early life experiences such as being criticized by parents or bullied by peers. ¹³ In a study examining the origins of shame experience from childhood through adolescence, physical appearance-related shame experiences were the most frequently recalled. ¹³ Body dissatisfaction, a known risk factor for disordered eating, longitudinally predicted excessive weight gain in a population-based sample of overweight adolescent girls. ¹⁴ Binge eating symptoms may emerge as a way to temporarily avoid or decrease the intensity of these negative feelings. Perhaps modifiable behaviors such as sleep and physical activity could alleviate these feelings, whereas other behaviors such as screen time may exacerbate them. Physical activity has been shown to reduce binge eating episodes in adults with BED. ¹⁵ A review by Sanchez-Carracedo et al. suggested working with the media to decrease the objectification of women and extreme thinness in girls or extremely muscular bodies in men as a way to prevent disordered eating and obesity, suggesting that screen time may be contributing to negative body image and binge eating symptoms. ¹⁶

Although studies support the high prevalence of binge eating among individuals with overweight and obesity and demonstrate the close relationship between disordered eating behaviors and obesity, few studies have examined binge eating symptoms and correlates including modifiable behaviors (sleep, physical activity, and screen time) among a diverse sample of children with overweight or obesity, which could elucidate avenues for prevention of obesity and disordered eating. ¹⁷

The objective of this study was to examine the prevalence and multilevel determinants of three binge eating symptoms: (1) sneaking, hiding, or hoarding food; (2) eating in the absence of hunger; and (3) inhibition or embarrassment when eating in front of others in children 5–12 years of age with overweight or obesity. We chose to focus on these three symptoms because of the participants' developmental stage, ease of identifying these symptoms through parental report, and a high prevalence of these symptoms in a previous study conducted by our team. ¹⁸ We hypothesized that modifiable behaviors such as sleep, screen time, and physical activity would be associated with these symptoms.

Methods

Results

Table 1 gives characteristics of the study participants. At baseline, the mean (SD) age was 9.7 years (1.9). Approximately 53% were non-Hispanic white, 20% were black, and 15% were Hispanic; 22% had severe obesity. The mean (SD) parent age was 41 years (6.7). The mean (SD) parental BMI was 29.6 (6.6) kg/m² and 31% of children lived in a family with an income ≤$50,000 per year.

In bivariate analyses (Table 2), children who sneak, hide, or hoard food were more likely to have a US-born parent (81.5%) than children without this symptom (72.7%, p = 0.01) (Table 2). Children who had this symptom were also more likely to have higher average screen time per day vs. those without (3.7 hours vs. 3.4 hours, p = 0.02). Those with the symptom of eating in the absence of hunger reported getting less sleep (9.2 hours vs. 9.3 hours, p = 0.03), more screen time (3.8 hours vs. 3.3 hours, p < 0.01), and less physical activity (1.4 hours vs. 1.6 hours p = 0.01) than those without this symptom. Children with the symptom of feeling embarrassed/inhibited when eating in front of others vs. those without this symptom were more likely to have severe obesity (36.2 vs. 20.6, p = 0.01) and more likely to live in a household where the annual income was ≤$50,000 (46.0% vs. 29.7%, p = 0.01).

In fully adjusted models, each 1 hour increase in screen time was associated with a 1.06 (95% CI: 1.01–1.11) times higher likelihood of having symptoms of sneaking, hoarding, or hiding food (Table 3). No other characteristics were significantly associated with sneaking, hiding, or hoarding food. Children with severe obesity had a higher likelihood of eating in the absence of hunger (PR: 1.50, 95% CI: 1.24–1.81) vs. those without severe obesity. Those children with a parent born outside of the United States were less likely to report this binge eating symptom (PR: 0.75, 95% CI: 0.58–0.96). In fully adjusted models, with each 1-hour increase in sleep, children were less likely to eat in the absence of hunger (PR: 0.90, 95% CI: 0.82–0.99). In addition, each 1-hour increase in screen time was associated with a higher likelihood of children eating in the absence of hunger (PR: 1.07, 95% CI: 1.03–1.11). Each additional year of age was associated with a higher likelihood of feeling inhibited/embarrassed when eating in front of others (PR: 1.33, 95% CI: 1.13–1.56).

Discussion

In this study of 817 children, we found that binge eating symptoms were prevalent and associated with several child and parent characteristics as well as modifiable child behaviors. Of the three binge eating symptoms of interest, we observed that parental report of the child's eating in the absence of hunger had the highest prevalence and was associated with severe obesity in children, parental US born status, and screen time and sleep.

Our results are consistent with previous reports that individuals with shorter sleep duration and insufficient sleep demonstrate overeating, and that shorter sleep duration may be associated with a higher risk of childhood obesity.^39–42 Our data suggest that shorter sleep duration may be associated with eating behavior in children with obesity. In particular, children getting less sleep were more likely to eat large amounts of food in absence of hunger. A potential mechanism for this observed relationship is supported by previous research that demonstrated that sleep restriction impairs glucose metabolism and alters central nervous control of hunger and appetite, leading to excess food intake. ⁴³

In our sample, children exposed to more screen time were more likely to have symptoms of eating in the absence of hunger and sneaking, hiding and hoarding food. We speculate that a potential mechanism could be that children are more likely to eat larger amounts of food in front of the TV because satiety cues are often disregarded.^44,45 Furthermore, these findings could be explained by previous reports of associations of TV content and its impact on food preference.^46,47 As noted in the introduction, binge eating symptoms may develop from shame feelings surrounding body image, which are often propogated by the media.

Physical activity was not found to be associated with any of the binge eating symptoms, which is inconsistent with the adult literature. ¹⁵ However, our measure of physical activity was through parental report and not by objective measurement such as by accelerometer. It is also possible that there is a different relationship between physical activity and binge eating in adults vs. children given the different recommendations for physical activity across the lifespan.

Strengths of our study include its large sample size, diversity in sociocultural and economic characteristics, and the measurement of modifiable health behaviors such as screen time and sleep as potential mediators of these binge eating symptoms. In addition, we were able to adjust for multiple covariates. This study also has several limitations. Because the binge eating symptoms were a report of parent's response of “yes” to three questions, it is possible there is a parental reporting bias of children's behavior. In addition, these items came from larger scales that we did not administer in their entirety due to participant burden from the larger randomized controlled trial survey questions. Consequentially, a binge eating diagnosis and associated psychiatric comorbidities were not clinically confirmed. There are also potential biases that are associated with parent-reported measures of both the binge eating symptoms and screen time, physical activity, and sleep, including both overestimation and underestimation. In the future, actual measurements of physical activity and sleep could be obtained by accelerometers. Other limitations to our study include its cross-sectional design preventing a causal relationship to be deduced.

Conclusions

These novel findings have important implications for public health and clinical practice. First, the results emphasize the need to improve awareness and recognition of binge eating symptoms among children with overweight and obesity. The literature has suggested that specialized mental health and earlier group-based interventions may be effective in preventing disordered eating. ⁴⁸ The findings also motivate the need for monitoring of these symptoms in child weight management interventions to ensure no adverse consequences occur and that the interventions seek to improve these symptoms. Our findings further suggest that recognition of binge eating symptoms in concurrence with health behaviors such as excess screen time or short sleep duration may yield a novel opportunity for a provider to address modifiable health behaviors and possibly reduce binge eating symptoms.