Attention-deficit/Hyperactivity Disorder: Associations with Overeating and Obesity
- First Online:
- Cite this article as:
- Davis, C. Curr Psychiatry Rep (2010) 12: 389. doi:10.1007/s11920-010-0133-7
In the past decade, we have become increasingly aware of strong associations between overweight/obesity and symptoms of attention-deficit/hyperactivity disorder (ADHD) in children, adolescents, and adults. This review addresses the prevalence of the comorbidity and discusses some of the mechanisms that could account for their relationship. It is suggested that the inattentive and impulsive behaviors that characterize ADHD could contribute to overeating in our current food environment, with its emphasis on fast food consumption and its many food temptations. It is also proposed—based on the compelling evidence that foods high in fat, sugar, and salt are as addictive as some drugs of abuse—that excessive food consumption could be a form of self-medication. This view conforms with the well-established evidence that drug use and abuse are substantially higher among those with ADHD than among the general population.
Attention-deficit/hyperactivity disorder (ADHD) was originally seen as a neuropsychological disturbance of childhood because the developmentally inappropriate signs of distractibility, impulsiveness, and inattention that characterize this condition typically appear before the age of 7 years. Eventually, however, it became evident that for many individuals, these symptoms did not remit and persisted into adulthood in a significantly debilitating way . This recognition has since prompted the reconceptualization of ADHD as a “life span disorder.” Although its etiology is not fully understood, there is general agreement that a biological component of the disorder relates to (dys)functioning of the prefrontal cortex, processes typically known as “executive functions” because they subserve reasoning, planning, inhibition, and decision making . Considerably less is known about environmental and development factors (including in utero events) that may contribute to ADHD symptoms, or about gene–environment effects that could influence the disorder’s onset.
It is very well-established, however, that ADHD coexists with a host of other clinical disorders. For example, major depression, bipolar disorder, and generalized anxiety are highly prevalent in those with ADHD , and in some cases, the disorder is associated with antisocial behavior and criminality . A wealth of evidence also links ADHD to addiction disorders such as drug abuse and alcoholism—comorbidities that are robust and bidirectional [5, 6]. Indeed, a recent familial risk analysis indicated that ADHD and drug dependence share a common vulnerability profile rather than present with independent modes of transmission . As both disorders have strong links to mesocorticolimbic pathways, Biederman and colleagues  proposed the conjoint involvement of brain dopamine activation and its regulation of arousal, attention, and reinforcement.
In recent years, we have become increasingly aware of strong links between ADHD and obesity [8, 9]. The first studies to draw attention to this association were published in 2002 [9, 10]. Both found a substantially higher than expected prevalence of ADHD in adults receiving treatment for obesity and an even greater occurrence (close to 50% of the sample) in those with class III obesity (body mass index >40 km/m2). The comorbid group also had more clinic visits and longer treatment duration than their non-ADHD counterparts . Recent data from a large, nationally representative sample of adults in the United States also found that adult ADHD was associated with a 1.6 OR of being overweight and a 1.8 OR of being obese [11••].
Since then, other studies have assessed the obesity association in children and adolescents with ADHD. In one study, the prevalence of overweight and obesity was significantly greater in the ADHD sample compared with age-matched population norms . At first glance, these results are rather counterintuitive, as a preeminent characteristic of ADHD in children is physical restlessness and hyperactivity. Moreover, only 14% of the Holtkamp et al.  sample were receiving pharmacologic treatment for ADHD at the time of assessment, and the remainder were medication naive. Medication status is clearly an important consideration, as weight loss and decreased appetite are among the adverse effects of the common psychomotor stimulants (e.g., methylphenidate) used to treat individuals with ADHD. Therefore, when investigating the ADHD–obesity links, it is important to consider possible moderator effects of pharmacologic treatment. Of relevance to this issue, a large US study found that unmedicated children and adolescents with ADHD had about 1.5 times the odds of being overweight, while their medicated counterparts had similar odds of being underweight [13••]. These results were partially confirmed by Ptacek et al. , who found that the nonmedicated children with ADHD had a higher percentage of body fat compared with the methylphenidate-medicated boys and with population norms, although there were no significant height or weight group differences. What is unclear about these correlational findings in children and adolescents is whether treatment status reflects an appetite suppression effect of the medication or an improvement in behavioral regulation and therefore healthier eating habits. Another possible explanation is that the greater cognitive effort required to execute standard mental tasks by those with untreated ADHD may foster hyperphagia. This interpretation is based on evidence that mental work substantially increases ad libitum food intake .
In summary, almost a decade of research has established sound links between ADHD and obesity in adults, children, and adolescents. It also appears that this relationship is not attributable to sociodemographic factors that influence people’s dietary patterns and opportunity for physical activity . However, one impediment to our understanding of the ADHD–obesity relationship has been the focus on studies that have included only cases of ADHD that meet full diagnostic criteria for the disorder. Such an approach has the potential for confound due to the higher likelihood of comorbidities in cases compared with controls. There is also the greater possibility of medication effects (as we have seen in the studies described previously). One strategy that avoids the potential confounds of clinical research is to examine associations among ADHD symptoms, aspects of overeating, and body weight in healthy participants from the general population. This approach is predicated on the assumption that personality factors and symptoms of disorder are best conceptualized dimensionally and occur with normal variation in the general population .
Mechanisms of Association
During the past few years, research has moved from an emphasis on comorbidity prevalence rates to attempts at identifying the mechanisms whereby obesity and ADHD present as overlapping conditions. In a current review—and as a way of explaining why the ADHD–obesity connection only appeared in the literature in 2002—Cortese et al. [18••] claim that the association was simply “overlooked” in the past. In my opinion, there is no evidence of neglect or oversight in the obesity or ADHD areas of research. Indeed, a more plausible explanation is that the association only emerged in the past decade with sufficient frequency for it to be recognized as a clinically relevant phenomenon. In other words, current conditions have given rise to a comorbidity that was less visible (or nonexistent) in earlier decades because the environmental risk for its development was largely absent. For example, we have seen an increasingly high prevalence of obesity since the 1980s  that is in stark contrast to earlier time periods. There have also been dramatic changes in our food environment during the past generation that have had a profound effect on population weight gain . An added factor is the groundswell of scientific research and clinical interest in identifying the causes of obesity in recent years. Of particular relevance are the strong links that have been established among impulsivity, the purchase of high-calorie foods, overeating, and obesity . These relationships almost certainly have guided scientific investigations in the direction of syndromes such as ADHD, a disorder in which impulse control is a key diagnostic facet.
For several reasons, I would argue that the ADHD–obesity association is essentially a modern variant of the ADHD–drug abuse comorbidity that has been recognized and documented for many years . In the following sections, I offer support for the premise that ADHD and obesity are interconnected by virtue of their respective associations with addiction disorders, the former displaying a proneness to and the latter occurring as a consequence of addiction. Expressed slightly differently, their comorbidity is the function of a shared diathesis whereby a predisposition to impulsive responding collides with a toxic food environment that exploits this vulnerability. The outcome is an increased probability of overeating and chronic weight gain in those with symptoms of ADHD.
Binge Eating and Food Addiction
Nowhere in nature is there food as high in fat, sugar, and salt (HFSS)—and sometimes caffeine—as we find in most of the processed foods we consume on a daily basis. Such palatable foods have the facility to activate the mesocorticolimbic pathways in the brain in a manner similar to other addictive substances, and much more potently than would occur for natural sources of energy [22••]. Compelling evidence indicates, based largely on well-controlled animal research, that HFSS foods have the potential for abuse and can lead to dependence in the manner of other condensed and concentrated substances such as cocaine and heroin [23, 24]. Like drugs of abuse, they have the ability to alter brain mechanisms in ways that contribute to their increasingly compulsive use [25, 26]. Their excessive ingestion also seems to foster binge eating episodes, symptoms of tolerance and withdrawal, and craving-like behaviors [22••, 27••].
These findings, in addition to a host of testimonials from those who suffer from compulsive overeating, have prompted the claims that some cases of obesity may be the consequence of a “food addiction” [27••, 28]. For some individuals, overeating is mostly a passive event that occurs without much awareness and takes place in the form of frequent snacking, eating calorie-dense meals, and selecting large portion sizes. For others, however, it can be an excessively driven activity. We have recently argued that binge eating disorder (BED) is a particular pattern of overeating with the strongest parallels to other addiction disorders and therefore is the quintessential food addiction phenotype [27••]. We also believe that sound clinical and scientific evidence supports this viewpoint. For example, Cassin and von Ranson  found that 94% of their adult BED sample described themselves as “food addicts” or “compulsive overeaters” and met criteria for substance-dependence disorder when the term substance referred to “binge eating.”
In the general population, binge eating has strong connections with symptoms of ADHD, similar to the longstanding associations of ADHD with addictive drug use/abuse . Clinical research also has demonstrated that binge eating behaviors, such as those seen in BED or bulimia nervosa, occur in patients with ADHD . In addition, BED seems to mediate the relationship between ADHD and overweight/obesity [11••]. Various causal mechanisms could link binge eating and ADHD. The symptoms and hypodopaminergic functioning that characterize ADHD seem to be ameliorated by dopamine-elevating behaviors (hence the treatment success of psychomotor stimulants) and may explain why some individuals succumb to illicit and/or recreational drug taking. It is therefore plausible that “comfort foods” could serve as a form of self-medication in those with ADHD given their dopamine-activating properties. If we accept that BED is appropriately conceptualized as a food addiction, then excessive HFSS food consumption can be seen as just another form of drug abuse. In other words, ADHD is a disorder defined by various maladaptive traits, and drug use is just one of the behavioral manifestations of the deficits associated with this condition.
Another possible causal mechanism is based on the well-established evidence that cross-sensitization occurs from one drug of abuse to another—and that cross-sensitization also occurs between intermittent sugar intake and addictive drugs [22••]. It may therefore be that stimulant treatment (eg, Ritalin; Novartis, Basel, Switzerland) in early life predisposes these individuals to the reinforcing properties of sugar as they mature, thereby increasing their proneness to obesity. Currently, I am unaware of any research that has prospectively assessed whether those with childhood ADHD and previously treated with stimulants are more likely to be obese as adults compared with their nontreated counterparts. Such a possibility provides an intriguing area for future research.
Among the impairing traits associated with obesity and ADHD—and another unifying thread in their comorbidity—is the centrality of high impulsivity in both conditions. This term is used to describe a multidimensional and relatively stable human personality trait that is characterized by 1) the diminished ability to exhibit restraint when this is the most advantageous and socially appropriate response in a particular situation, 2) an enhanced tendency to approach and engage in rewarding and pleasurable stimuli, and 3) impetuous responding without an appropriately adaptive concern for the future consequences of one’s actions.
We have learned that brain dopamine pathways play an important regulatory role in the expression of this endophenotype, which varies greatly in the general population [32, 33]. A high level of impulsivity has been associated with a wide range of psychological problems and disturbances. For instance, it correlates positively and consistently with drug use and abuse , binge eating [35••], and ADHD, for which it serves as one of the diagnostic criteria for this disorder. In a recent and relevant prospective study, it was also found that low impulsivity predicted decreases in body mass index percentile rank in children from fifth to eighth grade. In other words, the self-controlled (nonimpulsive) children seemed to be protected from weight gain in their transition to adolescence .
The impulsivity that characterizes ADHD is almost certainly a key factor mediating the relationship between this disorder and obesity. However, the mechanisms by which impulsivity contributes to weight gain are less clear. Nowadays, in the face of an overabundance of high-calorie snacks and meals, effortful control of food intake—what is colloquially called willpower—is essential for the maintenance of a healthy body weight. Historically, however, such self-regulated restraint would have been rather antithetical to the strong appetitive drives that fostered optimal fitness during most of our evolution. In the world of ready-made meals and cafeteria dining, healthy food choices are clearly not the easiest option. They typically require forethought and planning, extended time for preparation, and the ability to show steadfast control when faced with tempting and quicker alternatives. A better understanding of the natural human variation in self-regulated behaviors is clearly a way forward in explaining why some individuals are prone to chronic overeating while others, living in the same environment, can eat moderately and maintain a healthy body weight [37•].
Impulsivity has been a difficult construct to study because it involves at least two identifiable cognitive/emotional processes: reward-driven behaviors on the one hand and those characterized by poor inhibition on the other. It has also generated many measures and assessment devices that are not always highly correlated . Inhibitory control deficits, seen as impulsive responding and inattention, are largely regulated by the prefrontal cortex, which undergoes major developmental changes from childhood to adolescence . Impairment in these processes is typical of those with ADHD. Verbeken et al.  also found that overweight children showed less efficient executive inhibitory control and more reward-directed behavior as assessed by performance-based tasks. In addition, in a recent study, we demonstrated poor decision making and diminished ability to delay gratification in two groups of obese adults—those with and those without BED [37•]. Although the two obese groups did not differ from each other, they were both significantly impaired compared with an age- and gender-matched group of normal weight adults.
To our knowledge, only two studies have assessed whether impulsivity predicts success in a weight reduction program [41, 42]. In the more recent and larger study, high impulsivity predicted greater weight loss in adolescents, but not in children. Not only is this finding rather counterintuitive, but it is in direct contrast with the former study, which showed, as one would expect, that low-impulsive children lost more weight—findings that also mesh with adult research obtained from patients at a bariatric treatment center . At this point, however, we must conclude that the role of impulsivity in weight loss and weight loss maintenance is still inconclusive given the very few systematic treatment studies that have been published. Future replication studies and more detailed investigations with stratified analyses based on age, gender, and other relevant covariates are needed before firm conclusions can be drawn.
Twin and adoption studies indicate that genetic factors make a large contribution to the etiology of ADHD, with estimates of heritability ranging from 60% to 91% . It is also generally agreed that complex mental disorders such as ADHD are the result of a reciprocity between genetic factors and environmental influences (G × E). To date, however, the manner in which susceptibility genes interact with environmental risk factors is not clearly understood. Essentially, there are two broad possibilities to describe how a G × E interplay could come about. On the one hand, environmental factors only lead to an adverse outcome in the presence of a specific genetic makeup; on the other hand, an individual with a susceptible genetic makeup will only develop the adverse outcome if additional environmental pathogens are present . In the case of the ADHD–obesity links, the latter seems the more probable causal route in that the association only emerged when the food environment changed dramatically during the past generation or so. That the exponential increase in the prevalence of obesity has also occurred in a relatively short period of time highlights the importance of the interaction between the ADHD genotype and the toxic food culture in which it began to find itself. Those with the greatest genetic vulnerability are those who are likely to show the greatest weight gain, whereas those with genetic resistance tend to remain relatively weight stable .
Prenatal Drug Use
ADHD and its subsequent developmental course, which shows both continuity and change, are influenced by various pre- and perinatal, biological, and psychosocial environmental risk factors. Some that have stood the test of replication include the deleterious influence of maternal smoking and alcohol consumption during pregnancy, as well as low infant birth weight/prematurity . Fetal alcohol syndrome was first identified in 1973 as a condition with severe birth defects, including facial disfiguration and mental retardation, and was associated with heavy maternal alcohol use during pregnancy. However, it was not until the 1980s that the US Surgeon General issued an advisory recommending that women abstain from drinking alcohol during pregnancy, and almost a decade later before health warning labels were added to alcoholic beverage containers . During the late-1980s, it was estimated that about 30% of American women were still drinking alcohol during pregnancy despite these warnings, so one can only assume the rates were considerably higher before that time, especially with the growing visibility of female alcoholism in the decades following World War II .
The neurobehavioral deficits found in children with prenatal exposure to alcohol—what are now called fetal alcohol spectrum disorders (FASDs) or alcohol-related neurodevelopmental disorder—include symptoms of inattention, restlessness, impulsivity, antisocial behavior, and a diminished ability to anticipate future consequences. Clearly, these deficits vary in severity, and in cases of mild symptomatology, there may not be any formal diagnosis at all. To some extent, the severity of symptoms may be a dose-dependent consequence of the extent and timing of the mother’s alcohol intake. It is now generally agreed that even relatively small amounts of alcohol can cause fetal impairments in some cases.
What is instantly evident is the remarkable similarity of FASD symptoms to those of ADHD. Indeed, a substantially high proportion of children with prenatal alcohol exposure—as many as 70%—are also diagnosed with ADHD . Some recent studies have compared FASD and ADHD and reported certain qualitative differences in their respective types of attention problems . However, such studies are problematic and inconclusive because of the high dual diagnosis between FASD and ADHD and therefore the inability to compare “pure groups” of each condition. Moreover, few if any studies of ADHD take account of those with and without prenatal alcohol exposure. It is entirely possible that maternal alcohol use, which was not significantly curtailed until a decade or so ago, is one environmental causal contributor to this disorder and not merely a condition with pronounced symptom similarities. Prenatal exposure may have a sensitizing drug influence on the developing fetus and may also account for the higher than expected drug use and abuse in those with ADHD.
Prenatal Dietary Factors
Using data collected from three affluent Scandinavian societies, a prospective study showed for the first time that maternal overweight and obesity increased the risk of having a child with ADHD symptoms compared with mothers who were normal weight at the time they became pregnant . Interestingly, these findings persisted after controlling for baby’s birth weight, maternal age, and mother’s smoking status in the statistical analyses. However, the authors speculated that perhaps the relationship occurred because a genetic predisposition accounted for both the mother’s weight status and the subsequent ADHD symptoms in the child.
To test this prediction and to ascertain whether the previous associations would exist independent of other potential causal risk factors (eg, parental ADHD symptoms), a large replication and extension study was carried out of prepregnancy obesity and offspring symptoms of ADHD. Again it was found that children of obese mothers had a twofold increase in teacher-rated inattention scores compared with children of normal weight mothers [51•]. These inattentive children also displayed significantly greater negative emotionality. Arguments against the possibility that a common genetic predisposition linked maternal weight and child ADHD symptoms rest on the fact that the associations remained statistically significant after controlling for ADHD symptoms in both parents. The author concluded that maternal prepregnancy obesity is not merely an artifact of other influences that may coexist with mother’s weight status.
While Rodriguez [51•] proposed certain mechanisms that could account for the associations observed in her study, in my view, one possibility of considerable relevance to the ADHD–obesity link was overlooked. Excess body weight is in large part the result of high fat consumption, whereas high sugar intake seems to be responsible for producing addictive-like behaviors such as cravings, withdrawal, and compulsive intake . In our current food environment, with its superfluity of highly palatable foods, mothers who are overweight and obese are likely to consume larger and more frequent quantities of HFSS foods than normal weight mothers. As noted previously, good evidence indicates that these foods can cause brain neurochemical responses similar to those caused by other drugs of abuse. Therefore, highly processed and calorie-dense food consumed in abundance during pregnancy could produce deleterious outcomes in the unborn child in the same way that we see in the offspring of mothers who smoke and/or drink alcohol during pregnancy. In other words, the excessive ingestion of HFSS foods could produce what I will loosely call a fetal sugar spectrum disorder, with consequences that are not dissimilar to those seen in the offspring of mothers who drank alcohol and smoked nicotine during pregnancy.
I am aware of only one study that has examined the effects of treating ADHD symptoms in obese individuals who are attempting to lose weight. Levy and colleagues [52••] recently reported the results of a longitudinal investigation in which they observed obese patients with newly diagnosed ADHD for more than a year following pharmacologic treatment for the disorder. Significant weight loss was noted in these patients compared with an untreated control group, who actually gained weight during the same time period. The authors also found that treating symptoms of ADHD with medication improved the mental and physical health and the quality of life of those who were obese. Although the appetite-suppressing effects of the medication appeared to diminish after a few months, patients continued to lose weight and reported improvements in self-directedness and an increased capacity for persistence. They also reported that they no longer used food for self-medication purposes—that is, to restore energy or to focus attention. Importantly, binge eating diminished very substantially or stopped altogether in these patients.
This study was the first to target ADHD in the treatment protocol and therefore has important implications for the management of obesity and its adverse health consequences. Although surgical procedures such as laparoscopic banding and gastric bypass are increasingly popular options for handling refractory obesity, their success depends on adhering to strict postoperative dietary instructions following the treatment.
Based on clinical observation, Levy and colleagues [52••] claim that obese patients with comorbid ADHD are less able to comply with postsurgical recommendations than others; therefore, careful screening of obese patients seeking surgical treatment is strongly advised.
No potential conflict of interest relevant to this article was reported.