With the alarming increase in pediatric obesity1 (Ogden et al., 2006), it is imperative that correlates and predictors of excessive weight gain be identified and targeted for decreasing the vast number of youths struggling with overweight. While genetics plays an enormous role in determining body weight, other physiologic, social, psychological, and behavioral factors have a substantial impact as well.

This section focuses specifically on binge eating, defined as overeating while experiencing a lack of control over what or how much is being eaten (American Psychiatric Association [APA],  2000).  Relatively recent evidence suggests that binge eating may play a significant role in the development and maintenance of obesity.

This section will begin with an overview of binge eating in adults, with a focus on the putative diagnosis of binge eating disorder.

A discussion of the challenge involved in measuring binge eating among youths will follow. Cross-sectional data on binge eating among adolescents will then be reviewed, followed by the corresponding literature including child samples. Since a number of studies have combined participants in middle childhood (6-12 years) with adolescents (13-17 years), for the purposes of this section, all studies including individuals younger than 10 years will be included in sections focusing on children.

Prospective studies analyzing the outcomes and predictors of binge eating will be presented.  The final section will address directions and challenges for future research.

Throughout this section, only studies that defined binge eating as including a sense of “loss of control” while overeating will be addressed. Studies that equate the construct of binge eating with overeating without assessing whether a loss of control is experienced will not be reviewed.

Furthermore, data regarding binge eating should be considered as in the absence of purging or nonpurging compensatory behaviors. A caveat, however, is that a number of epidemiologic studies do not discriminate binge eating in conjunction with compensatory behaviors from binge eating in the absence of such behaviors.

Binge Eating Disorder in Adults

Although binge eating in obese adults was first described in 1959 (Stunkard, 1959), it was not until 35 years later that the proposed criteria for binge eating disorder (BED) were outlined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (APA, 1994).

BED is characterized by recurrent episodes of binge eating (overeating while experiencing a sense of loss of control regarding what or how much one is eating) accompanied by dysfunctional eating behaviors and marked distress regarding binge eating (Table 4.1).

People with BED do not regularly engage in inappropriate compensatory behaviors, such as self-induced vomiting, excessive exercise, fasting, or laxative/diuretic use (APA, 2000). As a result, BED is often associated with excess body weight and obesity (Yanovski, Nelson, Dubbert, & Spitzer, 1993).

As many as 20-30% of adults attending specialized weight loss clinics and approximately 3-5%  of the obese subset of the general population meet criteria for BED (Spitzer et al., 1991; Spitzer, Yanovski, Wadden, et al., 1993). Of note, the estimate is lower when clinical interviews are used in assessment (Kashubeck-West, Mintz, & Saundersam, 2001). BED seems to be equally prevalent in minority populations; a 1998 study found an overall BED prevalence rate of 1.5% in a biracial cohort of young adults (2.9%  in the obese subset),  with similar rates among African American women, white women, and white men and somewhat lower rates among African American men (Smith,  Marcus,  Lewis,  Fitzgibbon,  &  Schreiner, 1998).

By contrast,  a community study found African American women to engage in recurrent binge eating more frequently than white women (Striegel-Moore, Wilfley, Pike, Dohm, & Fairburn, 2000).

BED is associated with a number of health complications and psychosocial problems. Adults with BED suffer from higher levels of disabilities, poorer quality of life, and poorer physical health than obese adults without an eating disorder (de Zwaan et al., 2002; Johnson, Spitzer, & Williams, 2001), and they have eating- and weight-related psychopathology at levels similar to those of people with bulimia nervosa (Masheb & Grilo, 2000; Wilfley et al., 2000).

Women with BED have significantly more physical symptoms (e.g., headaches, menstrual problems) and indicators of poor health (e.g.,  limited ability to conduct usual activities,  difficulty sleeping)  than normal controls after co-occurring psychiatric disorders have been controlled for (Johnson et al.). Moreover, BED has been associated with social impairment and poor interpersonal functioning (Crow, Stewart Agras, Halmi, Mitchell, & Kraemer, 2002; Johnson et al.; Steiger, Gauvin, Jabalpurwala, Seguin, & Stotland, 1999; Wilfley, Wilson, & Agras, 2003).

Finally, recurrent binge eating in obese people has been consistently associated with Axis I psychiatric disorders and Axis II personality disorders (Mussell et al., 1996, Wilfley et al.; Yanovski et al., 1993). Studies have found that as many as 75% of BED patients report a lifetime history of a psychiatric disorder (Marcus, 1995; Yanovski et al.).

Recurrent binge eating is a predictor of inappropriate weight gain and obesity in adults.

In a community sample of individuals with bulimia nervosa and BED, Fairburn and colleagues (Fairburn, Cooper, Doll, Norman, & O’Connor, 2000) found that participants with BED gained an average of 4.2 ± 9.8 kg over five years, while those with bulimia nervosa gained less weight (3.3 ± 10.8 kg) over the same time period. At follow-up, significantly more participants with BED met criteria for obesity (BMI ≤ 30 kg/m2) than those with bulimia nervosa.  Of the BED group,  the proportion of individuals with obesity increased from 22% at baseline to 39% at follow-up (Fairburn et al.).

These findings are notable but should be interpreted cautiously since individuals without an eating disorder were not included in the study as a means of comparison. Nevertheless, among treatment-seeking samples,  some studies have shown that people with BED have poorer responses to obesity treatment or more rapidly regain lost weight following treatment than those without an eating disorder (Sherwood, Jeffery, & Wing, 1999; Yanovski & Sebring, 1994). While some data suggest that BED and subthreshold BED may remit without treatment (Fairburn et al.), results from a longitudinal study of people meeting full criteria for BED found that the disorder tends to persist until treated (Crow, 2002).

A number of retrospective studies of individuals with BED have suggested that the initiation of binge eating episodes occurs well before adulthood. The mean recalled age of first binge episode ranged from 10.6 to 16 years across studies (Abbott et al., 1998; Binford, Pederson-Mussell, Peterson,  Crow,  &  Mitchell,  2004;  Grilo &  Masheb,  2000;  Marcus, Moulton, & Greeno, 1995; Mussell et al., 1995; Spurrell, Wilfley, Tanofsky, & Brownell, 1997), with one study reporting that almost 6% of the sample recalled first binge eating at age 5 or younger (Spurrell et al.).

Such findings have prompted a number of researchers to examine binge eating behaviors in children and adolescents. Given the dramatic rise in pediatric obesity over the past three decades (Ogden et al., 2006), an increase in binge eating among youths is a reasonable hypothesis.

—
1Current Centers for Disease Control and Prevention terminology indicates that children with a Body Mass Index (BMI, kg/m2) greater than or equal to the 85th percentile are “at risk for overweight” and that those with a BMI greater than or equal to the 95th percentile are “overweight” (Ogden et al., 2002). For the purposes of the present section, the terms “overweight” and “obese” are used interchangeably and refer to BMI greater than or equal to the 95th percentile unless otherwise indicated.
—

 

REFERENCES

MARIAN TANOFSKY-KRAFF
Uniformed Services University of the Health Sciences and National Institute of Child Health and Human Development, Bethesda, MD 20892.
This research was supported by the Intramural Research Program of the NIH, grant ZO1-HD-00641 (NICHD, NIH) to Dr. J. Yanovski.