It is well established that although many children with ADHD achieve positive adjustments during adolescence, childhood ADHD is an important risk factor for a variety of negative adolescent outcomes including depression, anxiety, academic failure, and substance use.  An important negative outcome for adolescent females, however, is eating pathology.  This has not been studied in relation to ADHD, perhaps because research on girls with ADHD has been relatively limited, and eating pathology overwhelmingly affects females.

There are several reasons to believe that girls with ADHD may be at risk for disorders relating to eating and body image.  First, the distress associated with ADHD may manifest itself in girls in ways that are more typical of how girls show distress, i.e., through eating and internalizing disorders.

Second, the impulsive behavior seen in many individuals with ADHD is a central element found in individuals with Bulimia Nervosa (BN), an eating disorder characterized by binging and purging.

Third, children with ADHD are at risk for concurrent problems in relationships with parents and peers and problems in these relationship domains have been hypothesized to increase risk for the development of BN.  Although research documenting the link between parent-child relationship problems and BN is not available, results from several longitudinal studies suggest that girls with low self-reported social support and high self-reported unpopularity are at risk for developing BN symptoms.

Despite these reasons to hypothesize that ADHD may predispose girls to the development of eating pathology, studies of this interesting and important issue are lacking.  However, a paper published in a recent issue of the Journal of Abnormal Psychology represents an initial attempt to examine whether this hypothesized linkage actually exists [Mikami, Hinshaw, Patterson, & Lee (2008). Eating pathology among adolescent girls with Attention-Deficit/Hyperactivity Disorder. Journal of Abnormal Psychology, 117, 225-235.]

Participants were 228 girls who were 6-12 years old girls, mean age 9.5, at baseline; 140 were diagnosed with ADHD (93 combined type and 47 inattentive type) and 88 were comparison girls recruited from the same community. Follow up data - the measures collected at baseline and follow-up are described below - were collected an average of 5 years later when girls' average age was 14.2.  Girls with ADHD, Combined Type are those with high levels of both inattentive and hyperactive-impulsive symptoms.  Those with ADHD, Inattentive Type show high levels of inattentive symptoms only, and are not also characterized by high rates of impulsive and hyperactive behavior.

All girls participated together in research summer day camps that provided for extensive interaction with one another.  Nearly all girls with ADHD who were being treated with medication participated in the camp while off medication.  Additional information on the summer program is available in prior issues of Attention Research Update in which studies based on this sample are reviewed - see www.helpforadd.com/2007/february.htm and www.helpforadd.com/2002/october.htm.


- Baseline Measures -

ADHD Status and Symptoms - The primary predictor variable collected at baseline was ADHD diagnosis.  This was assessed using a structured psychiatric interview conducted with girls' parents. Counts of ADHD symptoms were also obtained via parent and teacher rating scales.

In addition to determining ADHD status at baseline, several variables hypothesized to increase the risk of developing eating pathology were also assessed.  These included girls' level of rejection by peers (as determined by information obtained from their camp mates) and parenting practices.  Finally, assessments were made of each girl's IQ, body mass index (BMI), satisfaction with her physical appearance, and other psychiatric disorders.  Information on these variables was obtained so that the researchers could test whether any predictive linkage found between ADHD in childhood and eating pathology in adolescence existed after controlling for these factors.


- Adolescent Outcome Measures -

The key measures collected during adolescence were related to girls eating pathology. Eating pathology was assessed in several ways. 

First, girls completed the Eating Disorders Inventory, a well-validated measure of eating  pathology that assesses Body Dissatisfaction, Bulimia, and Drive for Thinness. 

Participants also completed the Eating Attitudes Test, a well-validated measure of dysfunctional eating behaviors; subscales include dieting and bulimia/food preoccupation. 

Finally, adolescents completed the Body Image Survey, a measure in which they identify figures corresponding both to their current weight and to what they consider their ideal weight. The difference between the two is considered an index of the desire to be thinner.

These different measures were combined using a statistical procedure called factor analysis from which 2 distinct scales emerged.  One contained items that focused on the desire to lose weight and assessed the level of dissatisfaction with one's body. The second was comprised of items that more directly reflected pathological eating behaviors such as bulimic behaviors.  These two broader measures of eating pathology - Desire to lose weight and Pathological Eating - were then examined in relation to childhood ADHD status.


- Results - 

As predicted, childhood ADHD was significantly related to girls' scores on the 2 summary scales of eating pathology collected during adolescence.  Specifically girls with childhood ADHD had higher scores on the Desire to Lose Weight scale and the Eating Pathology Scale than did comparison girls.  This was true even after any differences between the groups in IQ, baseline satisfaction with physical appearance, adolescent BMI, pubertal status, and symptoms of other psychiatric were taken into account.  Differences between girls with and without ADHD were moderate in magnitude.

The authors next examined how girls with the different types of ADHD - combined type and inattentive type - compared.  Interestingly, girls with ADHD, Combined Type For Girls with ADHD, Combined Type showed higher levels of desire to lose weight than did inattentive girls; they also showed significantly higher levels of pathological eating than.  Among the 3 types of ADHD symptoms - inattention, hyperactivity, and impulsivity - impulsivity was found to be the strongest predictor.

Girls with inattentive ADHD had scores on these scales that were in between those of comparison girls and girls with ADHD, Combined Type.  However, differences between girls with inattentive ADHD and comparison girls were not statistically significant.  Thus, higher levels of eating pathology and desire to lose weight were only found in girls who had both inattentive and hyperactive-impulsive ADHD symptoms.

Interestingly, girls in both ADHD groups had higher BMIs during adolescence than comparison girls.  Because adolescent BMI was controlled for in the analyses, however, it does not explain the higher level of eating pathology and body image dissatisfaction found among girls with ADHD, Combined Type.


Did Girls with ADHD, Combined Type have Diagnosable Eating Disorders?

It is important to note that higher scores on the scales of eating pathology used in this study does not necessarily mean that girls with ADHD, Combined Type were struggling with diagnosable eating disorders.  In fact, the authors indicate that none of the girls met full diagnostic criteria for Bulimia Nervosa. 

However, between 5 and 10 percent of girls in the ADHD, Combined group showed levels of eating pathology that were regarded as "clinically concerning" compared to fewer than 1% of girls with inattentive ADHD or comparison girls.  In addition, parents reported that 8% of girls with ADHD-C had engaged in at least 1 binge-eating episode in the prior year compared to none of the girls in either of the other groups.


Parenting Practices and Peer Rejection as Predictors of Adolescent Eating Pathology

As noted above, measures of peer rejection and parenting practices had also been collected at baseline.  These factors were also found to predict eating pathology in girls during adolescence.  Specifically, after controlling for childhood ADHD status, girls with more punitive parents and girls whose parents were more openly critical of them, had higher levels of pathological eating.  Interestingly, punitive parenting did not predict eating pathology in comparison girls but only among girls with ADHD.

Higher levels of childhood peer rejection (as discussed above, peer rejection was assessed based on feedback provided by girls in the summer camp program) predicted higher levels of desire to lose weight.  This relationship held for both comparison girls and girls with ADHD.


- Summary and Implications -

Results of this study indicate that girls with ADHD, Combined Type in childhood are at risk for higher rates of eating pathology symptoms and body image dissatisfaction during adolescence.  Eating pathology appeared to be most closely linked to childhood impulsivity symptoms.  According to the authors, this suggests that the "...lack of self-regulation in ADHD-C and BN - bulimia nervosa - may link these two types of psychopathology."

The authors also suggest that another potential link between childhood ADHD and adolescent eating pathology "...may involve a shared dysregulated temperament present at birth before symptoms of either ADHD or eating pathology develop."  In other words, it is not ADHD per se that increases the risk for disordered eating during adolescence; instead, difficulty with emotional and behavioral regulation that is characteristic of children with difficult temperaments is what predisposes girls to develop both ADHD and BN symptoms.

As noted above, although girls with ADHD, Combined Type had higher rates of eating disorder symptoms, none qualified for an actual diagnosis.  Thus, the authors are appropriately cautious about drawing conclusions that childhood ADHD in girls predisposes them to develop eating disorders.  They also note, however, that the mean age of girls at follow-up - 14.2 years - is below the peak incidence of BN diagnoses.  This may explain why none of the girls in the sample had yet to meet full diagnostic criteria.

An important clinical implication of these findings is the need to be especially vigilant to the presence of disordered eating in adolescent girls with ADHD.  Because the adolescents in this study all had sub-clinical levels of eating pathology, it is possible that these difficulties would typically go undetected, particularly when other problems are more prominent.  However, the presence of such symptoms in early adolescence is likely to be an important risk factor for the development of more serious eating pathology as girls move into late adolescence and early adulthood.  Identifying eating difficulties early in their development could possibly trigger interventions to prevent the progression to more serious eating difficulties.

In summary, this is an interesting and important study that represents an initial attempt to examine the consequences of ADHD that may be more typical for girls than for boys.  The authors conclude that "...BN symptoms and body dissatisfaction should be incorporated into conceptions of risk and impairment among youths with ADHD."  They also note - correctly in my view - that the preponderance of males in studies of ADHD conducted to date has restricted our understanding of the developmental course of ADHD in females.

With additional work that builds on the foundation provided by these researchers, important questions about ADHD in girls and women will begin to be more completely addressed.