The primary goals of this study were to examine the psychological correlates of food addiction (FA), binge eating (BE), and a combination of both (FA + BE) in the general population. We demonstrated that FA and BE symptoms are present in non-clinical populations at relatively high rates. As expected, compared to individuals without reported disordered eating symptoms, individuals with these symptoms demonstrated greater severity of eating difficulties (i.e., emotional eating and dietary restraint), greater BMI, more internalizing symptoms (depression, anxiety, stress), more PTSD symptoms severity in the clinical range (vs. a non-clinical range for control participants), as well as reported more lifetime traumatic events and adverse childhood events. Within the ED groups, we found that the double diagnosis group presented the most severe pattern of behavioral and emotional symptoms, followed by the FA alone group, while the BE group had the fewest symptoms. Individuals with a double diagnosis, compared to either one of the other groups, reported significantly more food addiction and binge eating symptoms, greater emotional eating, and more adverse childhood events. They also reported higher levels of restrained eating, depression, anxiety, stress, and negative urgency, as well as more severe PTSD symptoms compared to the BE group alone.
We observed that double diagnosis group had more severe symptoms than FA and BE separately. As we previously noticed, FA and BE share many common features23. The studies by Davis et al.51 and Gearhardt et al.52 found that more than half of obese adults diagnosed with food addiction (based on YFAS) also met the criteria for BE. The researchers concluded that the BE-FA group was defined as a more disturbed variant of BE with higher rates of ED psychopathology, negative affect, emotional dysregulation, and low self-esteem52 – which is in line with our results. Our research adds one more stage to this continuum. We observed that people with FA presented more severe impairment in psychological functioning than people with high scores on the BES scale without FA. FA might represent a medium stage on the BE psychopathology continuum. It might be seen as a more disturbed type of BE, showing that binging might evolve into the constant urge to eat. These results are in line with the eating disorders continuum defined by Davis4 and the reward-based process model connecting negative emotional states and food addiction53. We might hypothesize that the double diagnosis group presents addiction to quantity and specific types of food54.
In BE, people experience dietary issues, concerns about weight, body shape, and depression symptoms, as well as brooding rumination. FA can be classified as substance addiction or behavioral addiction, based on criteria such as hunger, taste, pleasure, social connection loss, weight concerns, and awareness about addiction. Research showed that people with FA eat to gain hedonic pleasure, and they tend to look for their desired product, but people with BE might not activate binging without the release54. As such, people with both diagnoses may develop mixed strategies - to gain pleasure and to improve negative mood - and it might not be clear which kind of interventions to implement to decrease ED symptoms.
We also noted that depression, emotional eating, and restrained eating were present in the double diagnosis group - FA + BE group resulted in negative affect and impaired functioning However, they were not specifically in the FA and BE groups separately. Participants with both FA and BE characteristics were more likely to overeat for emotional reasons. It has been reported that negative affect is a key feature across cognitive-behavioural theories of binge eating55. We might say that binge eating serves as an avoidance from awareness56 in which the salience of adverse emotions is limited through cognitive narrowing that occurs during consuming food. There has been empirical evidence that negative mood plays a significant role in binge eating57 – as a consequence but in addition as a trigger for binging. The results are also in agreement with those of Stice et al.58. The dietary/depressed subgroup in his research exhibited mental dietary preoccupation. It also featured increased interpersonal sensitivity, negative urgency, alexithymia, and lower emotional regulation strategies.
Compared to the non-addicted BE participants, the FA group reported more negative consequences due to their eating difficulties and had significantly elevated symptoms of anxiety and stress which is in line with previous research by Parylak et al. 59. According to Burrows et al.26, FA is a result of coping with distress through eating. It is also possible that the mechanism is also connected to compulsive addiction60, in which depriving yourself of a compulsive habit is associated with negative emotionality, such as anxiety, as a consequence of restraining it. We might speculate that people with FA eat to regulate distress emerging from their psychological difficulties – they use eating as an avoidance strategy. The FA group also reported heightened PTSD symptoms, showing potential factors contributing to the diagnosis.
Despite the high anxiety symptoms associated with binge eating overall61, research suggests that binge eating itself is associated with reduced anxiety62. The higher rates of anxiety symptoms, coupled with the reduction in anxiety following binge eating, suggest a process that links these two issues. In our research, we might speculate that while BE may be more connected to depressed mood instead of anxiety, the FA group - being more anxious and stressed - implemented ineffective regulation strategies with eating becoming an urge and conditioned habit with irregular psychological rewards. This reasoning is in line with seeing FA through addiction disorders. As Shapira and Courbasson63 reported, anxiety in ED predicted if patients used substances or not, while anxiety and depression predicted ED severity. FA may be seen as an initial response to high physical discomfort which is conditioned over time. Furthermore, anxiety might represent one of the symptoms of substance or behavioral addiction64.
Participants with more food addiction symptoms had higher PTSD symptoms than the BE group. There might be an overlap in mechanisms between eating disorders and PTSD, as concluded in Meule and Gearhardt's review65. People with PTSD have a higher prevalence of all EDs66 especially those connected to binging67. Trottier and MacDonald68 conclude that the potential mechanisms of the trauma-ED relationship may lie in emotional dysregulation, anger, and impulsivity/compulsivity. In Brewerton’s review48, it was stated that experiencing ACEs may lead to self-destructive behaviors, including ED behaviors. It is probable that participants with FA symptoms exhibited a maladaptive regulation due to higher arousal or avoidance mechanisms than BE.
According to the MATEO study, FA is also associated with higher psychological distress and maladaptive eating patterns69. Higher levels of uncontrolled and emotional eating, stress, as well as lower psychological well-being and cognitive restraint, predicted food addiction in individuals with obesity. It is important to note that diagnosing FA among binge eating patients may direct professionals towards more effective treatment – focusing on anxiety for FA, on mood for BED, or both in case of double diagnosis.
Negative urgency was pronounced in both FA groups (FA group and double diagnosis group) compared to BE alone. A few studies have also indicated a positive relationship between negative urgency and food addiction symptoms70, 36. Negative urgency, defined as the disposition to act rashly when in distress, might make people more likely to seek out stress regulation strategies like eating behaviors. Most research shows a positive association between impulsivity and ED symptoms, especially those with binging episodes71. Taking into account our results, negative urgency in FA may act like an urge to consume certain foods or to the process of eating as a whole – and it is not characterized as an objective binge (i.e., having a lot of food at a certain time, often in loneliness, being planned and connected with negative mood and guilt). As Ratkovic et al.54 concluded that FA is driven mainly by impulsivity.
The study by Wolz et al.36 showed that patients who, in addition to emotional difficulties, tend to act impulsively when in negative mood states (especially anxiety), are at risk of developing addictive eating patterns. In other words, ED patients who are more prone to act impulsively in reaction to anxiety or stress might be at higher risk of developing eating patterns connected with addictive behavior.
Concluding, the anxious and stressed FA group may represent a different pattern of eating behaviors than the depressed BE group. Emotional eating, which includes snacking, taking a regular meal, or overeating, might be specific for BE due to the consumption of an unusually large amount of food but in a short period of time with a sense of loss of control. The FA construct, also connected with binge eating symptoms, might be an indicator of higher ED severity4.
Finally, we noted that BE group had a higher BMI than FA group. The correlation analyses (reported in the Supplementary Materials) indicated that the relationship between FA symptoms, binge eating severity, and BMI was very weak. Cernelic-Bijzak and Guine72 proposed that uncontrolled eating and BMI might be separate phenomena that affect one another only in certain cases. The results of this study demonstrated that adults with FA or double diagnosis may not have obesity and still report symptoms of uncontrolled eating and their consequences. We must note that compensatory behaviors were not controlled and may influence the overall weight of the participants. However, other factors contributing to uncontrolled eating and BMI should also be considered. It is possible that potential BMI-influencing factors like level of everyday spontaneous activity, type of work, dietary habits, medication use, and biological predispositions affected the relationship between disordered eating and BMI. In the group comparison, people with FA and BE in the ED group had a slightly higher average BMI than those in the control group, indicating weight gain. Due to the non-clinical character of the research sample, they may represent a risk group for future weight gain.
The FA/BE symptoms require further investigation to check the possible compensation strategies and the temporal aspect of the FA/BE symptoms (short time to gain weight due to binging or eating too much of the desired product). Excessive eating, like in BE, might be consistent with cultural and social background. People with normal weight might be less distressed and impaired by food addiction symptoms as they have personal and social acceptance of eating, as proposed by Carr et al.73. Moreover, FA eating patterns are slightly different from BE eating patterns. As such there might be less high-calorie food eaten in a large amount and in a short period of time, like binging (resulting in this group with a slightly higher BMI).
On the one hand, binging episodes are the answer to dieting, eating restriction, and weight reduction. However, on the other hand, binging might have an onset before dieting48. It may also explain our results – our participants may experience excessive food consumption as a result of emotional dysregulation or restriction of their favorite food. In spite of this, low calorie intake is not responsible for this. Another explanation might be the participants’ high weight before starting binging episodes (which is in line with the dietary restriction hypothesis as a background for binging).
LIMITATIONS
Some limitations of the present study should be mentioned. Due to the cross-sectional study design, it was impossible to investigate causal relationships between variables. Also, the online version of the research might have had an impact on the participants' responses, as external stimuli might have influenced their ability to focus on the questionnaire. Furthermore, as we only used self-reports, we did not have an objective measure of participants’ height and weight to generate measured BMI as well as access to the history of weight along with the dietary history. Moreover, not all participants filled out BES as it was an optional questionnaire for individuals who endorsed loss of control over eating. It required personal insight into the participant’s behaviors and experienced subjective distress that may not be present in all people suffering from BED. Furthermore, caloric intake and activity levels were not controlled. Those aspects are relevant in analyzing the binge behavior structure, as it is possible that some percentage of binges might be the result of excessive activity or low caloric intake during the day. Further studies should consider compensatory behaviors. Lastly, our study included a high number of women, which further limits the generalizability of the population.