Consumers are increasingly aware of their daily meals nutritional properties and they are also more and more interested to obtain foods which are generally recognized as “safe”(1). This awareness is triggering, however, not only benefits for personal health (or for their children’s health, in a broader context), but also certain tendencies toward eating disorders, like orthorexia nervosa (ON). Making a distinction between lifestyle peculiarities and eating behavior-related pathology is not at all simple, if social, psychological and cultural moderating factors are taken into consideration. Although the line between eating healthy and eating “pathologically healthy” is very thin(2), consistent efforts have been made to define certain diagnostic criteria for ON. These criteria should be used for helping high-risk individuals and patients with functional impairments due to their eating behaviors to make changes in their diet.
Orthorexia nervosa is a recently-cornered diagnosis that may be grouped together with other less well-defined eating disorders such as food/eating addiction, pregorexia or bigorexia, because it implies the use of only a restricted variety of foods, with possible long-term psychological and biological negative consequences(3). Dietary restrictions based on the perceived need to consume only the “healthiest” aliments lead to the elimination of multiple foods from these individuals’ meals, and are accompanied by stress, anxiety and self-aversion(1,4). Research conducted in the last decade shows that the risk of orthorexic behaviors and ON in the general population is increasing(1). Because a very high prevalence of ON was reported in nonclinical populations, there is a continuous debate about how to consider this phenomenon: is it a disorder in its own right, or does it become a problem only when occurring in the same time with another psychiatric disorder?(5) In a sample of 329 adults, that included four different groups – anorexia/bulimia nervosa group, obesity/binge eating disorder group, diet group, and the control group (represented by people not following a diet) –, higher rates of ON were reported in all three study groups when compared to controls(6). Individuals who pursue a diet share some common characteristics with those diagnosed with an eating disorder regarding their orthorexic tendencies at emotional, behavioral and personality levels(6).
Although not formally recognized by the current international psychiatric disorders classifications, orthorexia nervosa is considered an atypical eating disorder defined by an obsession for “healthy” eating(7). S. Bratman was the first physician who wrote about orthorexia in an article published in Yoga Journal (1997) and subsequently in his book, Health food junkies (2000)(8). Contrary to ON, the “healthy orthorexia” is a nonpathological type of orthorexia, defined by rational importance attributed to a healthy diet and lifestyle(9). Whereas ON is positively associated with a negative affect, disordered eating, and obsessive-compulsive symptoms, healthy orthorexia presents an opposite pattern of associations(10,11).
Several authors interpret orthorexia nervosa in the broader movement of “healthism”, a concept defined by the emphasis placed on personal responsibility in achieving and maintaining health, avoiding risk factors and preventing ill health(12,13). Sports and exercise, as well as awareness of the foods that are eaten are all components of this trend, but their interrelationship remain unclear in ON patients(13). The contradictory concept of “healthy anorexia” has been detected among individuals with orthorexic behavior who were apparently healthy, although they maintained a disordered eating behavior(14). Such people may be undetected by their general practitioners, because all their usual blood tests are within normal range and regular somatic exams did not offer enough evidence to support a suspicion of eating disorder.
According to the DSM-5, ON may be classified within the category of Avoidant/restrictive food intake disorder (ARFID)(15,16). However, unlike ARFID, orthorexia nervosa is not involving only restrictions of patients’ intake based on disinterest in food, sensorial properties of foods, or previous aversive experiences with food, but due to a pathological tendency to be as healthy as possible(17).
Another important differential diagnosis is represented by anorexia nervosa, but in orthorexia nervosa there is no significant influence of personal body weight over self-esteem, no lack of insight about the personal weight, and no prevalent ideas about the need to lose weight. Nonetheless, increased evidence for overlapping symptoms of anorexic and orthorexic eating behavior exist(18). Several authors have suggested that orthorexic eating behavior might serve as a coping strategy in anorexic patients, because it may enhance self-perception of eating behavior as autonomous and competent(19). Patients with anorexia nervosa have higher fulfilment of basic psychological needs in the domain of autonomy and competence if they presents also significant orthorexic behaviors, compared to patients with low level of orthorexic tendencies(19). Other authors consider ON to be just a new cultural manifestation of anorexia nervosa, that arose in the context of healthism and awareness of the foods nutritional properties(20). Therefore, these authors conclude that ON should be classified under the diagnostic umbrella of anorexia nervosa, and resembling more closely to the restrictive subtype(20). Another aspect of the complicated relationship between ON and anorexia nervosa is represented by the risk of developing anorexia nervosa in patients with severe orthorexic attitude toward food(21).
Also, a degree of overlap between orthorexia nervosa and obsessive-compulsive disorder has been suggested: obsessive thoughts related to foods, compulsive and ritualized activities about food preparation, etiquettes checking etc., functional impairment secondary to the previously mentioned symptoms, and lower quality of life(15,22). Also, high anxiety traits, a need to exert control, perfectionism, rigid thinking, excessive devotion, hypermorality, preoccupation with details and concerns about contamination have been detected in both ON and obsessive-compulsive personality disorder(11,23). According to a multicentric, controlled trial, 328 subjects were evaluated using a structured questionnaires for orthorexia nervosa and obsessive-compulsive disorder(24). ON symptoms were more prevalent among patients presenting obsessive-compulsive disorder than in the two control groups (individuals with anxiety-depressive disorders and general population), but the type of ON scoring tools may decisively influence this observation(24). The presence of obsessive-compulsive personality disorder had positively predicted the development of pathological eating habits, although not specifically orthorexia nervosa(23,25).
Other possible contexts of interpreting orthorexia nervosa that have been mentioned in the literature are the somatoform disoders, namely the somatic symptom disorder and the illness anxiety disorder, but also the category of psychotic disorders(23,26). Health anxiety may be a common feature between food-related worries and orthorexic behaviors, on one hand, and somatoform disorders, on the other hand(23). Psychotic disorders should also be included in the differential diagnosis of ON, especially in young patients(27). In one case study, ON has been described as a prodromal phase syndrome of schizophrenia(26). Also, the food-related magical thinking, and erroneous beliefs based on laws of contagion may be features of ON with relevance for schizophrenia spectrum disorders(3). Magical beliefs about food are of medical concern if excessive dietary restrictions are self-imposed and lead to nutritional deficiencies(23).
Risk factors and pathogenesis of orthorexia nervosa
Several risk factors for this pathology, which have been identified in the literature, are obsessive-compulsive features, eating-related disturbances and higher socioeconomic status(28). “The pursuit of an extreme dietary purity” tends to constitute a social and cultural fashion, which represents another obstacle in establishing a clear definition for orthorexia nervosa (29). Women, adolescents and those who practice certain sports (i.e., bodybuilding or athletics) are considered categories with a high risk for developing ON(30). Higher orthorexic tendencies correlated with higher score for perfectionism, either self-oriented, others-oriented or socially prescribed, but also with appearance orientation, overweight preoccupation, self-attributed weight, and fearful/dismissing attachment(31). Perfectionism, narcissism, rigidity, need for control and self-discipline are considered important variables in the creation of a vulnerability profile for orthorexia nervosa(11). Higher ON scores correlated with lower body areas satisfaction and less secure attachment style(31). A history of an eating disorder was the strongest predictor for the ON onset(31). Other reviews have reported obsessive-compulsive traits, psychopathology, disordered eating, history of an eating disorder, dieting, poor body image, and drive for thinness as being positively associated with greater risk for ON(32). According to other research, the most significant risk factors for orthorexia nervosa were education, the choice of profession, the socioeconomic status, and the internalization of the ideals of society are(33).
Factors like age, Body Mass Index, health-related profession, exercise engagement, vegetarianism/veganism, body dissatisfaction and alcohol/tobacco/drug use were not constantly related to a greater risk for ON(32).
Although the most accepted perspective about ON patients describe them as being concerned with food quality and healthy eating behavior, according to an online survey, they are also preoccupied with food quantity and physical appearance(34). The main predictors of orthorexia nervosa in a group of 469 yoga practitioners were the drive for thinness and a healthy orthorexia(34). As a consequence, recommending yoga for the prevention or treatment of eating disorders should be carefully considered in vulnerable individuals. Yoga practitioners are considered such a high risk population, along with dieticians, medical students, or gym-goers(34).
A neuropsychological evaluation of orthorexia nervosa (N=100 adults, 18-22 years old) detected a negative correlation between the severity of central symptoms and executive functioning(22). Results of the neuropsychological evaluations supported an independent association between low scores in “set-shifting”, “emotional control”, “self-monitoring” and “working memory”, and ON(22). After controlling for anorexia nervosa and obsessive-compulsive variables, high- and low-scores ON groups differed regarding the total correct trials on the Wisconsin Card Sorting Test (WCST)(22). Another study explored the correlation between the scores of ON symptoms and measures of inhibitory control, in a group of 63 participants(35). This study did not show any correlation between percent error or response time in Go/No-Go task, Flanker Test, Stroop Task, and orthorexia symptoms(35). These results imply that ON is not associated with deficits in inhibitory control, which is a significant difference when comparing to anorexia nervosa or obsessive-compulsive disorder(35).
A positive correlation between mindfulness (evaluated through the administration of Freiburg Mindfulness Inventory) and healthy orthorexia was found in a cross-sectional survey (N=314 women and 75 men; mean age 27.17 years old)(36). Orthorexia nervosa was negatively associated with mindfulness, without any moderating impact of the gender(36). This study shows the potentially favorable impact of mindfulness-based therapy in encouraging healthy eating and protect against eating disorders like ON.
In the psychosocial model of orthorexia nervosa, individual factors, like perfectionism, restrictive eating and drive for thinness, but also neuroticism and obsessive-compulsive tendencies are combined with social and cultural factors, like obesity stigma, higher income, positive reinforcement from others, and availability of organic food(32). Societal influences include aestheticism, moral citizenship and social media, but also parental influences, partners opinions, and relational groups (e.g., fitness or clean eating groups)(35). These factors are conjugated with individual components, like health concerns, belief in food as medicine, past trauma, personality features, exposure to extreme views and behaviors while growing up, and moral concerns(37).
Regarding the perception of individuals with ON by others, they were more likely to “improve with treatment” than in patients with bulimia nervosa, but more of a “danger to others” and “harder to talk to” when compared to individuals with binge eating disorder, according to a study (N=505 participants randomly assigned to read a vignette depicting a woman with anorexia nervosa, ON, bulimia nervosa or binge eating disorder)(38). Orthorexia nervosa was considered as less distressing, less likely to evoke sympathy, but also more acceptable than other eating disorders(38). ON was predicted by internalization of appearance ideas (thinness and muscular) and media pressure, through need fulfillment and health anxiety(39).
A cross-sectional study that enrolled 627 medical students from Lebanon showed a correlation between higher psychic anxiety scores and lower orthorexia tendency and behavior, while somatic anxiety was associated with interest in healthy eating(40). These results suggest that psychic anxiety may not be related to the risk of ON onset. This study did not show a significant correlation between higher eating attitudes disturbances and higher levels of ON tendencies and behaviors, while other studies did support this association(40). These differences may be explained by cultural differences between the countries from where the subjects were enrolled(40,41).
In a case series (N=4 women consecutively admitted to an outpatient unit for the treatment of eating disorders), the diagnosis of ON was established through clinical interview and a psychometric tool (ORTO-15)(42). All patients included in this analysis presented in their personal history a diagnosis of psychiatric disorder (obsessive-compulsive disorder, bulimia nervosa, illness anxiety disorder, psychotic disorder) before the onset of ON(42). This case series raises important questions about the relationship between other psychiatric disorders and ON, and underlines the difficulty to distinguish between a causal relation, a comorbidity or a shared vulnerability for ON in patients with other psychiatric disorders. Also, the overlap between ON symptoms and other diagnoses makes even harder the process of defining the relationship between these pathologies. In case of obsessive-compulsive disorder, for example, obsessive thoughts are mostly egodystonic, while in ON they are mostly egosyntonic; therefore, the patient with orthorexia nervosa is less likely to request help from a healthcare specialist, unlike a patient with obsessive-compulsive disorder(42). Also, in the case previously diagnosed with a psychotic disorder, the content of her delusions were related to “poisoned food”, while after treatment they converted into excessive care about the content of preservatives, artificial flavors and additives, exposure to pesticides etc., symptoms more typical for ON(42).
With a prevalence estimated to 6.9% in the general population, but with higher incidence in healthcare professionals and performance artists (up to 35-57.8%), orthorexia nervosa is considered a pathology that still escape the efforts of delineating its importance for mental and physical health(28). The accuracy of these epidemiological data is highly debated because the existing tools designed for ON detection have low specificity(43). For example, when personal relevance of eating behaviors and limitations in everyday life are also assessed, beside the administration of a validated questionnaire, the prevalence rate of orthorexic eating behaviors drop under 1%(28,43). Therefore, the available research tools do not identify with enough accuracy the boundary between individuals with an excessive interest in healthy lifestyle and those with dysfunctions in daily life secondary to their dietary restrictions(1).
A 10-item self-administered questionnaire was initially described for ON in 2000 (Bratman’s Orthorexia Test), with a Yes/No format, but without psychometric properties being researched for it(8). This questionnaire was designed as a screening instrument, and its author did not publish any interpretation guideline(8). Another questionnaire has been subsequently created, ORTO-15, consisting of 15 multiple-choice questions, six being taken from the previously mentioned instrument(8,44,45). Responses are scored on a four-point Likert scale, based on the frequency of symptoms manifestation(44,45). Scores above 40 are suggested to indicate the absence of orthorexia nervosa(44,45). Different versions of ORTO-15 have been created by translation in different languages – e.g., ORTO-11 (a Turkish version) or ORTO-9-GE (the German version)(46,47).
Orthorexia Nervosa Inventory (ONI) has 24 items scored on a four-point Likert scale and a three-factor structure(48). ONI subscales are “physical and social impairment”, “behavior and absorption”, and “emotional stress”(48). ONI scores were significantly higher among vegetarians and vegans, and in people exercising compulsively(48). There were reported no significant differences in the ONI scores between men and women, but these scores were negatively correlated with the Body Mass Index(48).
Düsseldorf Orthorexia Scale (DOS) includes 10 items assessing concrete behavioral aspects of this disorder(18). Cronbach’s α and retest reliability were acceptable (0.94 and 0.42, respectively)(18). Self-rated health of eating behavior and subjective importance of healthy eating correlated with DOS scores, indicating a good validity of this scale(18).
Based on evaluations conducted with ORTO-15 in nonclinical samples, a large proportion of individuals, up to 88%, scored in the orthorexia range, raising doubts about the ability of this instrument to distinguish between healthy eating and “pathologically healthy” eating(49-51). For example, although the prevalence of ON was over 70% in a college sample, only 20% of the sample endorsed a dietary practice of removing a particular food type from their diet(49). Also, less than 1% of the sample acknowledged impairment in everyday activities and medical problems(49).
People who followed a vegan diet had a less pathological mean ORTO-15 score(49). However, based on a literature review, following a vegetarian diet was related to orthorexic eating behaviors (in 11 out of 14 studies)(52).
A study focused on the comparison of four psychometric tools used for measuring ON severity (respectively Bratman’s Orthorexia Test, the ORTO-15, the Eating Habits Questionnaire [EHQ] and the DOS) and enrolled 511 adults who completed all these questionnaires(53). The originally suggested factorial structure was good for Brotman’s Orthorexia Test, EHQ and DOS, but not for ORTO-15(53). Also, the internal reliability was good for these instruments, except for ORTO-15(53). Therefore, with the exception of ORTO-15, orthorexia nervosa may be reliably measured with the currently available tools(53). The problems of ORTO-15 seem to be related to the scoring system, which need to be revised in order to increase its reliability.
Diagnostic and associated features
Diagnostic criteria for ON are focused upon obsessive thoughts regarding “healthy” eating, manifested by: (1) compulsive behavior and/or mental preoccupation about restrictive dietary practice believed by the individual to promote optimum health; (2) rigid self-imposed rules which trigger exaggerated fear of disease, feelings of personal impurity, and/or physical negative sensations ± anxiety and shame if they are violated; (3) dietary restrictions are increasingly severe, and this involves elimination of entire food groups, fasts or various purifying methods; (4) the desire to lose weight is absent, hidden or minor when reported to ideation about healthy eating; (5) the compulsive behavior and mental preoccupation impair the physical functioning (malnutrition, severe weight loss, medical complications), create interpersonal/social/academic/professional distress, or excessively influence body image, self-worth, identity and/or satisfaction(17).
Another set of criteria for orthorexia nervosa includes: (A) enduring and intensive preoccupation with healthy nutrition/foods/eating; (B) marked anxieties for extensive avoidance of foods labeled as “unhealthy” according to subjective beliefs; (C1) more than two overvalued ideas referring to the effectiveness or potential benefits of foods; (C2) ritualized preoccupation with buying, preparing, and consuming foods, which is not induced by culinary reasons but arise from over-valued ideas; intensive fears appear as a consequence to impossibility to adhere to nutritional self-imposed rules; (D1) the fixation on healthy eating causes suffering or impairments of clinical relevance in social, occupational or other important areas of life and/or negatively affects children; (D2) deficiency syndrome due to disordered eating behavior; insight into the illness may be present or not; (E) intended weight loss and underweight may be present, but worries about weight and body shape are not dominant(54). For a diagnosis of ON, criteria A, B, C and E must be clearly fulfilled, criteria D must be at least partially fulfilled, while the last criteria (E) differentiates orthorexia nervosa from atypical anorexia nervosa(54).
Associated traits of ON are obsessive thoughts related to food choice, planning, purchase, preparing, and consumption; distress when the individual is exposed to prohibited foods; moral judgments about others based on their dietary preferences; persistent belief that dietary restrictions are healthy despite the evidence of malnutrition(17). Obsessional or pathological preoccupation with healthy nutrition, emotional consequences of nonadherence to self-imposed nutritional rules, and psychosocial impairments in relevant areas of life, malnutrition and weight loss have been reported as defining orthorexia nervosa (29). ON symptoms were significantly correlated with eating disorder manifestations, obsessive-compulsive symptoms, anxiety and depression(55). Eating disorder symptoms was the only significant variable that remained correlated with ON in the regression analysis(55).
A cross-sectional study (N=519 adults) showed that depression, stress and anxiety were positively correlated with ON(9). These three factors play a mediating role between impulsivity and orthorexia nervosa/healthy orthorexia(9). It is hypothesized that orthorexic behavior can lead to obsessive eating habits and social isolation, which may be confounded with the tendency to isolation determined by the depression itself(9). A high level of anxiety may be detected in both pathological and healthy orthorexia, because restrictive eating patterns can be motivated by anxiety related to health, while an excess of worry about the foods can create or amplify the anxiety(9).
Orthorexic individuals experience intense discomfort when their eating habits are disrupted, disgust when food purity is compromised, and guilt or self-loathing when they commit food transgressions, all these in combination with chronic worries about imperfection and non-optimal health(23,56).
The existence of a by proxy type of ON has been suggested in children of parents who are imposing a restrictive diet (mainly vegan), leading to severe undernutrition(57). Infants and toddlers subjected to uncontrolled vegan diets may be at risk, and their health status should be regularly monitored(57). Although media reported several cases that could be classified as such, orthorexia nervosa by proxy remains an elusive construct, at least for the time being, because well-supported criteria for ON are not established(57).
The goal of any therapeutic approach in ON patients should be healthy eating without obsessive thoughts and ritualistic eating behaviors(8). Because no clinical trial could be identified in the literature assessing the efficacy of any therapeutic interventions for orthorexia nervosa, only theoretical considerations could be made regarding this domain(58). The treatment should be based on extensive psychoeducation and nutritional counseling(58).
However, many ON patients do not perceive themselves as having a health problem, therefore their adherence to treatment is questionable from the start(14). Even more, ON patients may consider themselves as being on a path to an idealized health, therefore they might perceive a therapeutic intervention as an intrusion into their lifestyle and core life beliefs(14).
A multidisciplinary approach, including pharmacotherapy, psychotherapy and psychoeducation, is considered the best approach(15,30,59,60). A balanced diet is considered the key element of the treatment plan, but also an antidepressant (e.g., selective serotonin reuptake inhibitors) ± an antipsychotic (e.g., olanzapine) – depending on the severity of clinical manifestations, and cognitive behavioral therapy should be taken into consideration(15). Cognitive restructuring in the ON patients should address the significance of “healthy foods”, the relation between health and nutrition, the catastrophic anticipation of the “unhealthy foods” consumption etc.(58) A confrontation with “feared foods”, in a graded exposure therapy, could be part of the treatment as in any phobia, and working with hypochondriacal ideas is another important aspect that may alleviate central ON symptoms(58). The main objective of the psychotherapy in these patients is creating a realistic “health concept”, both regarding the nutrition and the body functioning(58,61). Exposure and response prevention, in combination with habit reversal training, may be successful for treating obsessive and compulsive aspects of orthorexia nervosa(23). Relaxation training may be useful for alleviating pre- and post-prandial anxiety and other types of health-related anxiety(23).
In patients suffering from malnutrition due to excessive food restriction, hospitalization may be required. When ON patients’ list allows only very few foods and/or when there is a severe undersupply with essential nutrients, hospitalization is also a preferred option(58). The participation of family members and relatives is recommended when children are being harmed as a consequence of patients presenting ON(58). Antipsychotics, like olanzapine, have been used to decrease the obsessive nature of magical food-related thinking(23,59). In an 18-year-old patient with ON (suppression of sugar and fat from her diet, tightly counted meal calorie content, eating only self-made meals, avoidance of eating in public, unacceptance of other person’s opinion on diet, social isolation and a 15 kg weight loss) and comorbid depression (depressed mood, anxiety, anhedonia, fatigue, insomnia), the treatment with mirtazapine for 11 months resulted in the remission of the disordered eating behavior(62). Also, the patient regained her weight, she had an increase in the ORTO-15 score, and her depressive symptoms remitted(62).
Because evidence-based treatment strategies are lacking, the emphasis should be placed on early detection of orthorexia nervosa, using screening methods in vulnerable population. For this purpose, general practitioners, advanced practice nurses, registered nurses, and dieticians should be prepared to recognize ON symptoms(63). Training sessions for these healthcare staff should include data about the diagnosis, evolution, risk factors, assessment methods, treatment options and medical aspects of the treatment(63,64). An emphasis is placed by several authors on the need for caution when recommendations about dietary changes/regimens are made to vulnerable individuals, because such recommendations might trigger orthorexic behaviors(63).
Orthorexia nervosa is still a controversial diagnosis, but data about its importance for both mental and physical health are gathering. Several sets of diagnosis criteria have been formulated and multiple psychometric tools have been created(8,17,18,48,54). Not all the instruments for quantifying ON severity have equal validity, therefore care is needed when using them. Especially the validity of ORTO-15 has been questioned, and doubts about its scoring strategy have been raised(53). Also, cultural differences between populations have been detected by researchers as factors to impact these tools psychometric value(40), so that using such an instrument without prior validation on a specific population may be misleading(40).
Treatment strategies for orthorexia nervosa are not based on solid evidence, therefore care should be taken when approaching these patients. Exploring psychiatric comorbidities is of significant importance in orthorexia nervosa. Due to lack of insight in their eating behavior, ON patients’ involvement in a therapeutic plan is expected to be low. Psychotherapy remains the main recommendation, but in cases where malnutrition is observed, pharmacological treatment under medical surveillance should be initiated.
Disclaimer: The authors declare no conflicts of interest.