The idea that specific kind of foods may have an addiction potential and that overeating such as in binge-related eating disorders or obesity may represent a form of addicted behavior has been discussed for decades. The term food addiction
was first introduced in the scientific literature in 1956 by Theron Randolph [1
]. Although comparisons between addiction and eating behavior were sporadically drawn in the following decades [2
], approaches to systematically examine and define food addiction were not pursued until the early 2000s. Particularly, a substantial increase in the number of publications using the term food addiction
can be observed since 2009 [9
This increased scientific interest in this topic was in part driven by the rise of neuroimaging and subsequent findings that obesity and binge eating are associated with alterations in dopaminergic signaling and food-cue elicited hyperactivation of reward-related brain areas which are comparable to processes seen in drug users [10
]. Those findings were further complemented by animal models showing addiction-like behaviors and neuronal changes in rodents after some weeks of intermittent access to sugar [12
]. In the current article, we will not go into further detail about those lines of research and refer the reader to recent works on those topics [13
]. Instead, we will focus on the phenomenological similarities between substance dependence and some forms of overeating in humans.
2. Parallels between Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) Substance Dependence Criteria and Overeating
The diagnostic criteria for substance dependence in the fourth revision of the Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV) included (1) tolerance, defined as consuming increasing amounts of a substance to achieve the same effects or experiencing diminished effects with continued use of the same amounts; (2) withdrawal symptoms when the substance is not consumed or using the substance to avoid withdrawal symptoms; (3) using the substance in larger amounts or over a longer period than intended; (4) a persistent desire or unsuccessful efforts to cut down substance use; (5) increased time effort to obtain or use the substance or recover from its effects; (6) reduction of social, occupational, or recreational activities because of substance use; and (7) use of the substance despite a persistent physical or psychological problem caused or exacerbated by the substance [18
]. Substance dependence could be diagnosed when a clinically significant impairment or distress was present and at least three symptoms were met in the past year.
There are numerous articles in which the applicability of those DSM-IV substance dependence criteria and other features of addicted behavior to bulimia nervosa (BN), binge eating disorder (BED), obesity, or overeating in general are discussed [19
]. However, the translation of substance dependence criteria to eating behavior is not straightforward and, as a result, there is some disagreement among researchers about the precise definitions of food addiction symptoms [31
Although empirical evidence for the applicability of some DSM-IV addiction criteria to eating, such as tolerance and withdrawal, is mostly based on animal studies [12
], all of the seven symptoms conceivably can be found in humans [26
]. Compelling support for this was provided by a study by Cassin and von Ranson [36
], in which almost all participants with BED received a diagnosis of substance dependence when the term substance
was replaced with binge eating
in a diagnostic interview. The authors noted, however, that participants’ responses may have been influenced by demand characteristics and that reliability and validity of their interview assessment was uncertain [36
3. Yale Food Addiction Scale (YFAS)
In an attempt to overcome mixed definitions of food addiction symptoms and to provide a standardized measure for the assessment of food addiction, the YFAS was developed [37
]. This 25-item instrument measures the presence of food addiction symptoms based on the DSM-IV substance dependence criteria (i.e.
, seven symptoms). Additionally, two items assess a clinically significant impairment or distress as a result of overeating. When both a clinically significant impairment or distress is present and
at least three out of the seven symptoms are met, then food addiction can be “diagnosed”. Prevalence rates of these food addiction diagnoses according to the YFAS range between approximately 5%–10% in non-clinical samples [37
], 15%–25% in obese samples [43
], and 30%–50% in morbidly obese bariatric patients or obese individuals with binge eating disorder [48
The most common food addiction symptom as assessed with the YFAS is a persistent desire or unsuccessful efforts to cut down or control eating
]. Among obese individuals, almost all participants fulfill this criterion [46
]. Other commonly endorsed symptoms are continued eating despite physical or psychological problems
, particularly in obese samples (ibid.). The remaining symptoms (consumption of large amounts or over a longer period than intended
, spending much time obtaining food or eating or recover from its effects
, giving up important activities
, and withdrawal symptoms
) are less common, particularly in non-clinical samples [42
], but are nonetheless endorsed by a substantial proportion of obese individuals [48
4. Substance Dependence Criteria in DSM-5
In the newly revised version of the DSM, the diagnostic criteria for substance abuse and—dependence were merged such that criteria for substance use disorders (SUDs) now additionally include (1) failure to fulfill major role obligations at work, school, or home as a result of substance use; (2) continued substance use despite social or interpersonal problems caused or exacerbated by substance use; and (3) recurrent substance use in situations in which it is physically hazardous [54
]. Moreover, the DSM-IV substance abuse criterion of having legal problems was dropped, but a newly created symptom of craving
, or a strong desire or urge to use the substance
was incorporated (Table 1
). Three levels of severity can now be specified ranging from mild
(presence of two to three symptoms) to moderate
(presence of four to five symptoms) to severe
(presence of six or more symptoms).
Notably, SUD symptoms also differ across substances (Table 1
). For example, although there is an intoxication and withdrawal syndrome described for caffeine, the other symptoms do not apply for caffeine and, thus, there is no caffeine use disorder. Vice versa
, although all of the eleven symptoms apply to tobacco, there is no intoxication described. Finally, there is no withdrawal syndrome described for hallucinogens, for example phencyclidine, and inhalants.
6. Gambling Disorder and Overeating
Besides the revised SUD criteria, gambling disorder has now been added as a non-substance-related disorder [54
]. Diagnostic criteria include (1) a need to gamble with increasing amounts of money in order to achieve the desired excitement; (2) being restless or irritable when attempting to cut down or stop gambling; (3) repeated unsuccessful efforts to control, cut back, or stop gambling; (4) a preoccupation with gambling; (5) gambling when feeling distressed; (6) after losing money gambling, returning another day to get even; (7) lying to conceal the extent of involvement with gambling; (8) jeopardizing or losing significant relationships, jobs, or educational or career opportunities because of gambling; and (9) relying on others to provide money to relieve desperate financial situations caused by gambling (Table 2
). Gambling disorder can be diagnosed as mild
(four to five criteria met), moderate
(six to seven criteria met), or severe
(eight to nine criteria met), when symptoms were present in the past year.
Some of the gambling disorder criteria can conceivably be applied to eating behavior. For example, repeated unsuccessful efforts to control, cut back, or stop the behavior is a core feature of BN, BED, and food addiction as measured with the YFAS (see above). Moreover, studies using the YFAS consistently show that food addiction is strongly associated with a preoccupation with food and eating and with overeating when feeling distressed [37
]. As with withdrawal syndrome in SUDs, a restlessness or irritability when attempting to cut down or stop overeating seems plausible. Using the YFAS, almost 30% of obese individuals and up to 50% of obese individuals with BED report regular experiences of such withdrawal symptoms when cutting down on certain foods [48
]. However, these subjective reports are potentially biased as it may be hard for respondents to distinguish between symptoms emerging from a general energy deficit (i.e.
, consuming not enough calories) and those that are actually associated with avoiding specific foods.
The criterion of the need to gamble with increasing amounts of money in order to achieve the desired excitement may be translated to a need to eat increasing amounts of food in order to achieve the desired satisfaction. This definition would be, thus, equal to the SUDs’ tolerance criterion, which has been shown to be endorsed by a substantial proportion (approximately 50%–60%) of obese individuals in studies using the YFAS [48
]. However, this criterion may not be applicable to eating when keeping the reference to a feeling of excitement when engaging in the behavior.
Other symptoms seem transferable when substituting the term gambling
). Individuals with BN or BED usually experience marked feelings of shame and, thus, conceal their binge eating and this often involves deceiving others about the extent of involvement with overeating [76
]. Jeopardizing or loss of a significant relationship, job, or educational or career opportunity may most likely occur because of weight gain. For example, there is experimental evidence showing that human resource professionals underestimate the occupational prestige of obese individuals and would less likely hire them [77
]. Regarding the criterion of desperate financial situations caused by gambling, the money spent on binge foods markedly affects quality of life in individuals with BN and BED, the latter of which are in particular bothered by financial problems [78
]. Although binge eating involves spending substantial amounts of money, actually plunging into debt or borrowing money from other people to finance overeating probably only occurs in rare cases. Finally, the symptom of returning another day to get even after losing money gambling does neither seem to be transferable to eating behavior nor to SUDs.
Gambling disorder criteria according to the DSM-5 and possible corresponding food addiction criteria.
Gambling disorder criteria according to the DSM-5 and possible corresponding food addiction criteria.
|Gambling Disorder *|
|Plausible, but not applicable when referring to excitement|
Restlessness or irritability when attempting to cut down or stop gambling
Restlessness or irritability when attempting to cut down or stop overeating
|Plausible, but hard to distinguish from energy deficit|
Repeated unsuccessful efforts to control, cut back, or stop gambling
Repeated unsuccessful efforts to control, cut back, or stop overeating
Preoccupation with gambling
Preoccupation with food and eating
Gambling when feeling distressed
(Over-)eating when feeling distressed
After losing money gambling, often return another day to get even
Lying to conceal the extent of involvement with gambling
Lying to conceal the extent of involvement with overeating
Jeopardizing or loss of a significant relationship, job, or educational or career opportunity because of gambling
Jeopardizing or loss of a significant relationship, job, or educational or career opportunity because of overeating
Relying on others to provide money to relieve desperate financial situations caused by gambling
Relying on others to provide money to relieve desperate financial situations caused by overeating
|Plausible, but unusual|
7. Implications of the Research Domain Criteria for Food Addiction Research
Recently, the Research Domain Criteria
(RDoC) have been introduced as a new approach to classifying mental illnesses, although it is important to note that the RDoC is designed as a research framework rather than an alternative diagnostic framework [80
]. The RDoC approach is designed to focus on domains that reflect neurobiological, physiological, genetic and behavioral underpinnings. The current domains focus on positive valence, negative valence, cognitive functioning, social processes, and arousal/regulation [80
]. Critics of the DSM suggest that the focus on “theory free” assessment has limited the incorporation of scientific advances into the diagnostic framework [82
]. Thus, in its current form, the DSM may not adequately reflect knowledge gained in the areas of genetic, physiological, and neurobiological research. Although the RDoC system is not designed to be implemented as a diagnostic method in clinical settings, it is likely to be a major guiding factor in scientific evaluations of psychopathology and will hopefully improve treatment effectiveness [80
The RDoC approach to diagnosis will also likely guide research on whether an addictive process contributes to certain types of overeating. Binge eating disorder appears to be related to many of the mechanisms implicated in addictive disorders, including elevated motivation to seek out palatable foods, greater neural activation in reward-related circuitry to high-calorie food cues, and limitations in cognitive control [23
]. However, individuals with a BED diagnosis are not homogenous, with a subtype that is indicated by high-levels of dietary restraint and another subtype that exhibits greater negative affect, impulsivity, and overall pathology [84
]. These two subtypes of BED could potentially be driven by different mechanisms with an addictive process possibly contributing to the latter subtype (but not the former). Thus, some (but not all individuals) with a BED diagnosis may experience an addictive response to certain foods.
Finally, one of the major proposed mechanisms underlying addiction is the ability of an addictive substance/behavior to alter underlying systems in a manner that drives problematic behavior [86
]. In other words, individual risk factors (e.g., impulsivity, reward sensitivity, negative affect) interact with the addictive potential of a substance/behavior to result in pathology. As the RDoC approach highlights the importance of identifying mechanisms, examining whether certain foods or ingredients in foods are capable of altering the system in a manner that is akin to addictive substances/behaviors will be an essential line of research. There has been significant progress in this area using animal models of eating behavior [87
], but research in humans is limited. Addressing this gap in the literature is extremely important for evaluating the validity of the food addiction concept. In sum, the RDoC system will be important for the evaluation of the concept of food addiction as it highlights moving beyond shared signs and symptoms and instead focuses on evaluating whether the etiology and underpinnings of addictions are contributing to compulsive food consumption.
Research on the DSM-IV diagnostic criteria for substance dependence shows that they can be translated to eating behavior and that many individuals with obesity and/or BED fulfill those criteria based on self-report measures such as the YFAS. With regard to the newly added criteria in DSM-5, one study shows that three out of four symptoms may be less relevant in the context of food and eating [90
]. However, this was a small-sized qualitative study based on the themes that participants spontaneously mentioned during a semi-structured interview. As we have outlined above, all of the new symptoms can conceivably be applied to eating. Thus, future studies using standardized measures such as a revised YFAS are necessary for appropriately evaluating the relevance of the new SUD criteria for food addiction.
Even if it turns out that the new symptoms, except craving, do not occur in the context of food and eating, it may still be questioned if this would disprove the existence of food addiction. As can be seen in Table 1
, the diagnostic criteria as outlined in the DSM-5 do not apply to each substance to the same extent. Specifically, there are SUDs that do not cover the full range of symptoms (caffeine, hallucinogens, inhalants) or do not include intoxication (tobacco). In addition to this, the DSM criteria in general have been criticized for being rather inappropriate for tobacco [94
]. Also, the DSM is criticized for its lack of focus on underlying mechanisms, which is a central component of the newly proposed RDoC system. Thus, a major test of the food addiction hypothesis will be to not only focus on the signs and symptoms linking addiction and problematic eating behavior, but also to examine the similarities and differences in the underpinnings of these conditions.
To conclude, we think that the DSM-5 criteria may be valuable for food addiction research, even if some of those symptoms may rarely be endorsed by participants exhibiting addiction-like eating. On the other hand, using those criteria for diagnosing food addiction entails the risk to overestimate the occurrence of food addiction. Thus, future investigations need to take great care that the new SUD criteria are properly translated to food and eating and that reasonable diagnostic thresholds are applied when diagnosing food addiction. Finally, we emphasize the need to think more mechanistically in the evaluation of food addiction by examining the contribution of biological, psychological, and behavioral circuits implicated in addiction to problematic eating behaviors.