Abstract
Purpose
Meat avoidance has long been thought to be related to eating psychopathology; however, research does not necessarily support this notion. Furthermore, commonly used eating disorder scales may be picking up on normal meat-avoiding behaviours in vegetarians and vegans. As such, we systematically reviewed the association between vegetarianism, veganism, and disordered eating, and reviewed the psychometric properties of eating disorder scales for use in these populations.
Methods
We searched electronic databases MEDLINE, PsychINFO, and CINAHL for literature published until June 2021.
Results
Forty-eight studies met eligibility criteria, with no consensus as to whether meat avoidance was associated with higher rates of disordered eating. Most studies reported a significant positive association with both vegetarianism and veganism, and orthorexia nervosa. Six studies provided evidence for the use of eating disorder measures in vegetarians and vegans, reporting poor psychometric fit among all scales.
Conclusion
This systematic review highlights the extent to which vegetarians and vegans have been highly understudied, with limited research suggesting higher levels of orthorexia nervosa behaviours in vegetarians and vegans. Furthermore, our results provide tentative evidence that the factorial validity of commonly used eating disorder scales, such as the EDE-Q, may be poor in vegans.
Level of evidence
Level I, systematic review.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
It has long been thought that vegetarianism and veganism are related to an elevated risk of disordered eating; however, past research does not necessarily support this notion [1, 2]. Vegetarianism, defined as a dietary pattern involving the exclusion of red meat, and often poultry, fish, and seafood, is considered diverse and heterogenous in nature, encapsulating a number of wide-ranging dietary variations (i.e., lacto-vegetarianism, semi–vegetarianism) [3]. Additional restrictions are imposed on those following a vegan diet (i.e., the exclusion of all animal-derived products), with many incorporating lifestyle modifications such as rejecting clothing or entertainment involving exploitation of, or cruelty to, animals [4]. According to Rosenfeld and Burrow’s [5] dietary identity theory, vegetarians and vegans consider their diet to be a central part of their identity, display more positive feelings about their dietary in-group, and feel more negatively judged for following their dietary pattern relative to omnivores [6, 7]. These findings suggest that meat avoidance is considered by some to be more than just a diet, but rather an interlacing of one’s identity when it comes to food.
The idea that veg*ism, used herein to indicate the spectrum of meat avoidance, is associated with greater rates of disordered eating has not been well established. At its very core, veg*ism involves a high level of cognitive restraint to consciously regulate and restrict several food groups. While dietary motivations for veg*ism do not appear to influence disordered eating rates [8,9,10,11], it has been posited that veg*ism in and of itself may act as a socially acceptable method to restrict food intake and camouflage disordered eating behaviours [12,13,14,15,16]. Orthorexia nervosa, a form of disordered eating characterised by a pervasive obsession to eat “clean” and “pure” foods, has shown the strongest link with veg*ism due to similar overlapping food selection strategies [17]. For example, both orthorexia nervosa and veg*ism allow individuals to reduce their food intake according to specific nutritional rules (e.g., consumption of low sugar or gluten-free diet, abstaining from meat products) resulting in a diet of very few foods. While both facilitate efforts at dietary restriction, veg*ism may allow its followers to legitimize this food avoidance, potentially enabling greater disordered eating behaviours [16].
It is possible that some eating disorder scales may be capturing normal veg*an-motivated food choices and behaviours such as higher levels of cognitive restraint due to a heavy avoidance of certain food groups [1]. This could result in inaccurate estimates of the prevalence of eating disorders in these populations. Some examples of potentially confusing items in commonly used eating disorder scales to veg*ans are displayed in Table 1. For example, the Eating Disorder Examination-Questionnaire (EDE-Q) asks respondents to rate the degree they exclude foods and use food rules to influence weight or shape which may rely on insight that their dietary restrictions are indeed weight or shape motivated to be accurate. A limited number of studies have assessed the psychometric properties of commonly used eating disorder scales in veg*ans, producing largely inconsistent results. This demonstrates a significant gap within the literature, which is particularly important considering the utility of these tools (e.g., EDE-Q) as screening instruments and outcome measures.
The aim of this study is to conduct a systematic review of all published studies to ascertain the association between vegetarianism, veganism, and disordered eating. To our knowledge, only one systematic review has investigated the relationship between vegetarian diets and disordered eating. Sergentanis [24] collated the results of 20 studies finding a positive relationship between veg*ism and disordered eating in adolescents and young adults. Our systematic review represents an addition to Sergentanis [24]’s important work by investigating disordered eating along the meat-avoidance spectrum in a wider age range (i.e., adults aged 16 years and over included). Furthermore, we extend our systematic review to a secondary aim to examine the psychometric properties of eating disorder scales for use in vegetarians and vegans. Finally, we will then discuss the potential clinical implications of employing commonly used eating disorder scales to diagnose and monitor treatment progress in veg*ans with a suspected eating disorder.
Methods
This systematic review was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [25]. A priori registration to PROSPERO was received on 4 May 2021 (registration number CRD42021244365). A manual literature search was conducted in June 2021 in electronic bibliographic databases MEDLINE, PsychINFO, and CINAHL. A combination of search or MeSH terms, appearing in either title, abstract, subject heading, or keyword, were used, including vegan* OR vegetarian* OR “plant based” OR “meat avoid*” OR “meat abstain*”, with “eating disorder*” OR “disordered eating*” OR anorexi* OR bulimi* OR orthorexi* OR “binge eat*”. The reference lists of eligible papers were searched for relevant studies reporting on psychometric properties of eating disorder scales in veg*ans for inclusion.
Inclusion criteria
Inclusion criteria were based on the PICOS format (Participant, Intervention, Comparison, Outcomes, Study Design) [26], whereby articles eligible for inclusion in this review were required to meet the following criteria.
Types of participants
Studies that reported on individuals 16 years and over were eligible. This age range was chosen because we elected to focus on independent individual food choices rather than family-based food choices.
Types of interventions
Studies on individuals who follow a vegetarian (i.e., excluding red meat) or vegan (i.e., excluding animal-derived food products including meat, poultry, fish, eggs, and milk) diet were eligible.
Types of comparisons
Studies that used omnivores (i.e., consuming animal and plant products) as the reference group were eligible.
Types of outcomes
Studies assessing the relationship between vegetarianism, veganism, and disordered eating/eating disorders were eligible. We also elected to include studies assessing the relationship between vegetarianism, veganism, and orthorexia nervosa due to its characterisation is a form of disordered eating with overlapping symptoms as a subtype of anorexia nervosa [27].
Types of studies
Case studies, letters, conference abstracts or posters, systematic reviews, meta-analyses, extant reviews, and narrative reviews were restricted for inclusion. Grey literature dissertations were eligible. No publication date restrictions were imposed.
Exclusion criteria
Animal studies and studies that reported on samples under the age of 16 were excluded. Studies were restricted to those written in English language.
Study selection
First author independently screened studies against eligibility criteria using systematic review management platform, Covidence. Studies were screened in a hierarchical fashion, whereby titles and abstracts of searched literature were screened first, followed by full text to identify studies that met inclusion criteria. Eligible papers were imported and stored in database management software, Endnote, to allow for easy data extraction. If multiple articles for a single study were available, the most up-to-date publication was used.
Data extraction and synthesis
Tables were used to synthesise study characteristics and results. Study characteristics and results include first author name, date of publication, country of origin, study design, sample size, vegetarian and vegan sample size, veg*an age, veg*an gender, veg*an ethnicity/race, veg*an socioeconomic status, vegetarian and vegan study definition, study population, eating disorder measure, main findings including effect size of the correlation between veg*ism and disordered eating if available, and Newcastle–Ottawa scale quality rating. Due to large heterogeneity between study samples (e.g., N = 45 to 10,137), diet type (e.g., vegetarianism, semi–vegetarianism, laco-vegetarianism, veganism), and measures (e.g., orthorexia nervosa, restraint, disordered eating, binge eating, food addiction) [28], it was not possible to pool data into a statistical meta-analysis. Therefore, a quantitative synthesis is provided.
Quality assessment
To assess extraction bias, review author, GS blindly reviewed 10% of full-text articles to ensure inclusion and exclusion accuracy. This involved review author being unaware of the inclusion status of the manuscript during the assessment. If discrepancies did arise during the bias assessment, disagreements were discussed and resolved by consensus, or through the inclusion of third author.
To assess methodological quality, all studies were assessed using a modified Newcastle–Ottawa scale for cross-sectional [29], retrospective, or case–control studies [30]. The Newcastle–Ottawa scale allocates points to each study evaluating three domains: selection, comparability, and outcome/exposure. Cross-sectional studies were allocated a maximum of ten points, with higher scores consistent with greater methodological rigor (high = 9–10, good = 7–8, satisfactory = 5–6, poor = 0–4). Retrospective and case–control studies were allocated a maximum of 9 points, with an overall rating of high (> 6), moderate (4–5), or poor (0–3), with high ratings indicative of greater methodological rigor. Review author, GS blindly assessed 10% of studies. If discrepancies did arise during the quality assessment, disagreements were discussed and resolved by consensus, or through the inclusion of third author.
Results
Study selection
The initial literature search identified a total of 308 potentially relevant studies. After removal of duplicates, 199 were screened at title and abstract level with 119 deemed not relevant. A further 80 studies were evaluated for eligibility at full-text level with 32 excluded due to not meeting selection criteria. A discrepancy of four papers was identified during review author’s extraction check and were resolved through discussion and subsequently included in the review. Ultimately, a total of 48 studies were included in the systematic review. A PRISMA flow diagram describing the successive steps of the selection process is presented in Fig. 1.
Quality assessment of included studies
The methodological quality of most cross-sectional studies was satisfactory, followed by poor quality. The main limitation across studies was no description of the response rate or characteristics between respondents and non–respondents. A major strength across studies was the utilisation of valid and reliable measurement tools, such as the EDE-Q and Eating Attitudes Test (EAT). Of the three studies of retrospective design, in terms of overall rating, one study received a moderate-quality rating, and two received a poor-quality rating. The one case–control study was rated as high quality. The main limitation across retrospective and case–control studies was unclear follow-up rates or description of participants lost. A main strength across the studies was the representativeness of the veg*an samples. The quality ratings of the 48 studies are presented in Table 2.
Study characteristics
The study characteristics of the 48 included studies are presented in Table 2. Most studies (n = 35) were published within the last 10 years and primarily based in the U.S. (n = 20). Forty-four of the 48 studies used a cross-sectional design, with the remaining four studies using retrospective and case–control design. The total sample size of studies ranged from N = 45 to N = 10,137.
Relationship between veg*ism and disordered eating
Study characteristics
Thirty-six studies were identified that reported on the relationship between veg*ism and global disordered eating. Veg*an sample size of studies ranged from N = 5 to N = 822. Nineteen studies provided criteria for defining vegetarianism in their sample which ranged from excluding beef to all forms of animal flesh, and 17 providing criteria for defining veganism which ranged from excluding meat, egg, or dairy products to all animal-derived products. Mean veg*an sample age ranged from 19.0 (SD = 1.0) years in the youngest sample to 40.9 (SD = 15.5) years in the oldest sample. Most studies used mix-gender samples (n = 17), with 12 studies using female-only samples. No studies reported an even gender distribution among the veg*an group, with females significantly outweighing males in all studies. Six studies reported veg*an ethnicity/race, supporting predominantly White/Caucasian participants. Five studies reported veg*an SES. Community samples and university students were the most common study population (both n = 11), followed by mixed university student and community samples (n = 6), eating disorder patients (n = 3), university students and eating disorder patients (n = 2), mixed eating disorder patient and community samples (n = 1), and women in their third trimester of pregnancy (n = 1). The most common disordered eating measure was the EAT (n = 13), followed by the EDE-Q (n = 8) and Dutch Eating Behaviour Questionnaire (DEBQ; n = 6).
Vegetarian samples
A range of statistical tests showed a significant positive association between vegetarianism and disordered eating in eighteen of 33 studies [8, 9, 12,13,14,15,16, 31,32,33,34,35,36,37,38,39,40,41]. Sixteen studies reported no association between vegetarianism and disordered eating [2, 9, 17, 22, 31, 34, 35, 39, 42,43,44,45,46,47,48], and seven studies reported a negative association between vegetarianism and disordered eating [9, 45, 49,50,51,52,53]. Of note, some studies reported differing associations depending on the specific eating disorder measure used. For example, Dorard and Mathieu [45] found higher disordered eating when measured by the Body Shape Questionnaire (BSQ), but not when measured by the EAT-26. There is also little consensus within eating disorder measures. For example, when measured using the EAT, five studies reported greater disordered eating in the vegetarian sample than omnivores, compared to eight studies that reported no association between vegetarianism and disordered eating.
Vegan samples
Most studies (8 of 13) reported no association between veganism and disordered eating [2, 9, 10, 17, 42,43,44, 54]. Six studies reported lower disordered eating in the vegan sample [9, 10, 33, 49, 51, 53], with two studies reporting greater disordered eating in the vegan sample compared to omnivore controls [9, 15]. Like the vegetarian results, findings of studies differed depending on the eating disorder measure or subscale used. For example, Heiss [10] reported lower disordered eating in vegans when measured by the EDE-Q, but no association when measured by the Eating Disorder Inventory (EDI), Binge Eating Scale (BES), Yale Food Addiction Scale (YFAS), and DEBQ. Similarly, Timko [9] reported no association in veg*ans when measured by the EAT-26 and DEBQ emotional eating subscale, but a negative association between veganism and the DEBQ dietary restraint and emotional eating subscales.
Relationship between veg*ism and orthorexia nervosa
Study characteristics
Eighteen studies were identified that reported on the relationship between veg*ism and orthorexia nervosa, a form of disordered eating characterised by a pervasive obsession to eat “clean” and “pure” foods [17]. The sample size of studies ranged from N = 62 to N = 2,826, with veg*an sample sizes ranging from N = 6 to N = 1,346. Of the 17 studies with vegetarian participants, eight provided criteria for defining vegetarianism in their sample. Of the 14 studies with vegan participants, eight provided criteria for defining veganism in their sample. Veg*an sample age was predominately based in the mid to late 20s. In line with Relationship between Veg*ism and Disordered Eating, gender distribution was heavily dominated by females. Information on veg*an ethnicity/race were rarely provided, with one study reporting a predominately White/Caucasian sample (88.0–90.6%). Three studies reported veg*an SES. The most common measure to assess orthorexia nervosa was the ORTO (n = 8), followed by the Dusseldorf Orthorexia Scale (DOS; n = 5), and Eating Habits Questionnaire (EHQ; n = 3).
Vegetarian samples
A range of statistical tests showed a significant positive association between vegetarian adherence and orthorexia nervosa pathology (15 of 16) [17, 42, 49, 50, 56,57,58, 60, 62,63,64,65,66]. One study reported no association between vegetarianism and orthorexia nervosa pathology [43], and no studies reported a negative association between vegetarianism and orthorexia nervosa pathology.
Vegan samples
Nine of 12 studies reported a significant positive association between veganism and orthorexia nervosa pathology [17, 42, 49, 51, 55, 58, 60, 66, 67]. One study reported a negative association with orthorexia nervosa pathology [59], while two studies reported no association between veganism and orthorexia nervosa pathology compared to omnivore controls [43, 62]. Notably, those that reported a negative or no association between veganism and orthorexia had substantially smaller veg*an sample sizes. For example, Dunn [59] reported veganism was associated with lower orthorexia nervosa pathology measured by the ORTO-15 in a sample of six vegans (equivalent to 2% of the total sample size). Similarly, Herranz Valera [62] reported no association between veganism and orthorexia nervosa pathology using a sample of seven vegans.
Psychometric properties of eating disorder scales in veg*ans
To address the secondary aim of this systematic review, the psychometric properties of eating disorder scales applied in veg*an samples were reviewed and synthesized (see Table 3). Of the 48 studies included in this review, six studies provided evidence for the use of eating disorder scales in veg*ans. Most studies assessed the EDE-Q (n = 5), followed by the DEBQ and EDI (n = 2). One study examined an orthorexia nervosa measure, the ORTO-15. Notably, four studies are reported by the same author, with two studies using the same sample [54, 61].
In terms of reliability, good to excellent internal consistency of the EDE-Q was reported in vegans [10, 51, 54, 61], and good internal consistency was reported in combined veg*ans [41]. Among the other reported measures, acceptable to good internal consistency was reported for the DEBQ and TFEQ, and good to excellent internal consistency was reported for the EDI, EAT, Drive for Muscularity, BES, and YFAS in vegetarians, vegans, and combined samples [10, 46]. Internal consistency for the ORTO was found to be unacceptable in both vegetarian and vegan groups [51].
In terms of model fit, three studies conducted confirmatory factor analysis (CFA) using the EDE-Q reporting contradictory results. Heiss [54] found difficulties in replicating model fit of the four-factor model in addition to the three-, two-, and full one- and brief one-factor models in a sample of vegans. Further research found adequate but slightly less support for the brief three-factor model than the omnivore group [61].
Measurement invariance was documented in two studies using the EDE-Q in samples of vegans and vegetarians. Heiss [61] reported the factor loadings of the EDE-Q were not equivalent between the vegan and omnivore group. However, Zickgraf [41] supported full measurement invariance between non–vegetarians, weight-motivated vegetarians, and non–weight-motivated vegetarians.
Discussion
We present the first systematic review to examine the association between vegetarianism, veganism, and disordered eating in adults of all ages. This review is also the first to synthesize the psychometric properties of eating disorder scales for use in vegetarian and vegan populations. A total of 48 studies met eligibility criteria, with most studies being cross-sectional in nature. Samples comprised primarily community samples of young adult women of Caucasian descent. Importantly, this review highlights the extent to which vegetarians and vegans have been highly understudied in the eating disorder/disordered eating research field. As veg*ism is likely proving to be to be more than just another “passing fad” [68], future research is vital to untie the complex relationship between disordered eating and veg*an eating behaviours and attitudes.
Our review showed there is no consensus whether vegetarianism or veganism is associated with higher levels of disordered eating. Specifically, our review was unable to confirm whether diet type (e.g., semi–vegetarianism, veganism) influenced rates of disordered eating. For example, we found studies reporting higher disordered eating in vegans [e.g., 9, 51, 60, 66], vegetarians [e.g., 58, 63, 65], semi–vegetarians [e.g., 9, 33], meat reducers [e.g., 51], and omnivores [e.g., 9] compared to other subtypes among a range of scales. While restricting dietary groups based on animal products is expected to serve as a risk factor to developing an eating disorder, it may be the case that the number of dietary groups excluded (e.g., meat, diary) may not play as greater role [17]. Future research is required to disentangle the extent to which diet type serves as a risk factor to disordered eating habits, with longitudinal research required to confirm the causality of this association. Our review also showed that the relationship between vegetarianism, veganism, and disordered eating differed depending on the assessment measure employed [9, 10]. It is clear disordered eating is a broad construct, whereby different measures may be detecting slightly different aspects of disordered eating attitudes and behaviours. Future research should aim to examine the relationship between vegetarianism, veganism, and specific aspects of disordered eating, as this was seemingly the exception, rather than the norm in the studies reviewed.
While motivations for dietary adherence do not appear to influence disordered eating rates (i.e., being veg*an for health reasons vs. ideological/ethical reasons) [8,9,10, 69], lack of agreement discovered through this review may also be explained by additional factors, such as poor methodological quality. For example, smaller sample sizes of veg*an samples meant studies were likely statistically underpowered. Another possible explanation is a lack of consensus in defining subtypes of veg*ans. While vegetarians, vegans, and their respective subtypes have been shown to differ in meaningful ways, many studies continue to group them into one singular category, potentially masking true associations between each group. Stringency in ensuring distinct vegetarian and vegan samples is recruited will ensure greater confidence when drawing on results of future studies.
In contrast to the lack of agreement between veg*ism and disordered eating, vegetarianism and veganism appear to be associated with greater orthorexia nervosa pathology. Characterised by a fixation on eating “healthy” foods, such as through an increased concern about the health of ingredients, compulsive checking of ingredients lists, and cutting of food groups (e.g., sugar, gluten), orthorexia nervosa is not formally recognised in the Diagnostic and Statistical Manual of Mental Disorders (DMS) [70]. This lack of formalised diagnostic criteria is in part due to disagreement around how best to classify orthorexia, with some suggesting it shares common features with anorexia nervosa and obsessive compulsive disorder (OCD) [71]. Overlapping similarities between veg*ism and orthorexia nervosa have been characterised to include specific food selection according to nutritional rules and food-related issues becoming a large part of one’s day-to-day [42]. It remains unclear, however, as to whether orthorexia scores of those following a veg*an diet exceed cut-off for pathological orthorexia behaviours. Some found not only do vegetarians and vegans display greater orthorexia behaviours compared to omnivore controls, but they also reach preliminary cut-off for orthorexia nervosa, with vegans being more affected [17, 51].
The positive results of the systematic review must be considered with limitations of orthorexia nervosa scales in mind. For example, the most widely used scale, the ORTO, is broadly criticized for having inconsistent psychometric properties [72], and other scales, such as the DOS and EHQ, have been shown to detect both orthorexia nervosa and healthy orthorexia [73]. As veg*ism is associated with greater health consciousness (i.e., consume alcohol less frequently, exercise more often, have a higher daily intake of fruit and vegetables) [10, 39], it may be the case that veg*ans are high in healthy orthorexia, rather than orthorexia nervosa itself, but commonly used orthorexia tools are unable to disentangle these distinct constructs [74]. Future work must pay attention to measuring the multidimensional constructs of orthorexia nervosa to ensure we correctly account for psychopathology in populations of interest, including veg*ans. Ultimately, until validated and widely supported diagnostic criteria have been established, true orthorexia nervosa remains difficult to classify and the results of this systematic review must be considered in view of this.
Our systematic review found support for overall good psychometric properties of eating disorder scales (e.g., EAT, DEBQ) in veg*ans. These results should be considered in light of preliminary research that calls into question difficulties in replicating the factor structure of these tools. For example, our systematic review supported good reliability and validity of the EDE-Q but found limited support for an appropriate factor structure of the tool in vegetarians and vegans [75]. Given full measurement invariance of the EDE-Q was supported across samples of omnivores, weight-motivated vegetarians, and non–weight-motivated vegetarians, but not samples of vegans and omnivores [41], it may be the case that the EDE-Q is not measuring the same latent construct of disordered eating across vegetarian, vegan, and omnivore groups. Furthermore, the psychometric properties of orthorexia nervosa scale, the ORTO, was provided in one study revealing poor internal consistency and insufficient discriminant validity. Taken together with the findings that vegans had greater orthorexia nervosa scores, it may be the case that the ORTO is not detecting true pathological orthorexic eating behaviours but rather normal dietary adherence within this population. Taken together, this area presents a critical gap in eating disorder research, particularly considering these measures are being used to estimate eating disorder prevalence in the community, as well as support diagnostic options for vegetarians and vegans with a suspected eating disorder.
Strengths and limits
There are several strength and limitations to this systematic review. This systematic review offers the most comprehensive examination of the association between vegetarianism, veganism, and disordered eating to date, and is the first to synthesise all available literature to inform the use of eating disorder scales in assessing veg*ans. Ultimately, this systematic review highlights that vegetarian and vegan research has been wholly understudied in the eating disorder research field. An important limitation is that due to a heavy reliance on cross-sectional studies, the causal relationship between veg*ism and disordered eating cannot be confirmed. It therefore remains unclear whether veg*ism increases the risk of developing disordered eating, or whether developing disordered eating increases the chances of transitioning to a vegetarian or vegan diet. Second, the significant amount of missing demographic information (e.g., gender, ethnicity/race) across studies raises concerns regarding the representativeness of the samples. From the available demographic information, there is also limited variability with participants comprising predominately young Caucasian women. While this is in line with vegan population estimates from existing literature [76, 77], generalisability of these findings is constrained. Furthermore, we defined vegetarianism as the exclusion of red meat compared to the exclusion of meat in general, resulting in an additional four known studies. This means we may have captured participants who are not strict vegetarians, such as pescatarians or semi–vegetarians, but does ensure inclusivity of the meat-avoidance spectrum. We also acknowledge that many studies did not provide a definition of veg*ism in their study, meaning that the number of combined vegetarian, semi–vegetarian, or pescatarian samples may be underestimated in our results. We encourage researchers to incorporate rigorous processes, such as the two-tier process to defining meat avoidance described in Asher [78], to assist in minimising ambiguity in definitions and creating a streamline approach to the research field. Finally, quality ratings within studies were overall low, indicating a high degree of risk of bias, likely characteristic of a heavy reliance on self-report data.
Clinical implications
This systematic review has important clinical implications for veg*an populations. Although further clinical research is very much needed, it is possible that veg*ism, in and of itself, may not necessarily be a risk factor for the development of eating disorders. Furthermore, our results provide tentative evidence that the factorial validity of commonly used eating disorder scales, such as the EDE-Q, may be poor in veg*ans. According to our review, at this stage, the brief-three-factor model (also known as the short-three-factor model) of the EDE-Q provides the only promising lead in appropriately measuring eating disorder symptoms in veg*ans. Specifically, the brief-three-factor model contains items that query restraint to influence body weight or shape (i.e., “Have you tried to exclude from your diet any foods that you like in order to influence your shape or weight?”, “Have you tried to follow definite rules regarding your eating in order to influence your shape or weight?”) so as not to pick up on normal veg*an-related eating restraint. Though this may be troublesome itself, as it relies on a high level of insight from respondents that their dietary restrictions are indeed weight motivated to provide accurate responses. Furthermore, as the model contains seven items under three factors, it does not meet minimum recommendations for factor analysis (e.g., four items per latent variable), limiting reliability estimates and generalizability of the questionnaire [79]. Until eating disorder scales have undergone stringent psychometric testing in both vegetarian and vegan populations, caution must be taken when interpreting their results in research and clinical settings.
In the meantime, it remains important for clinicians to gain an in-depth understanding of veg*an patients reasons around food exclusion and dietary rules to ensure their eating habits are not being over-pathologized for simply following a veg*an diet. Furthermore, clinicians should also enquire about when they began their diet adherence to determine where exactly the diet fits into the patient’s history of disordered eating (i.e., did veg*ism predate the onset of the eating disorder or did veg*ism start as a result of the eating disorder) [41]. While weight restoration is best achieved through the reintroduction of meat [80], doing so in veg*ans who attribute their diet to reasons unrelated to weight loss or health (e.g., ethical, environmental, religion) may be detrimental to their treatment as a whole [1, 81]. Questions about motivations for veg*ism and sequence of events should be re-asked throughout treatment as patients may initially minimise the extent to which their veg*an diet is weight control or health motivated and the role it plays in the development and/or maintenance of their eating disorder [41]. As veg*ism is also known to be highly intertwined with ones sense of identity [6, 7], clinicians should focus on developing dietary flexibility in the context of a vegetarian or vegan diet to ensure the patients sense of self is maintained during treatment, and subsequent recovery [1, 41]. However, patients will gain greater insight into these concepts as treatment progresses.
Future research
Research must establish whether commonly used eating disorder scales are accurately quantifying pathological eating behaviours in veg*ans. Based on the findings of this review, we recommend future research conduct an exploratory factor analysis of these scales in veg*ans to assess whether a different factor structure provides a more appropriate fit. We also recommend future research provides support for construct validity by comparing test scores using a semi–structured interview such as the widely supported EDE [75]. If, through confirming the psychometric properties of commonly used eating disorders scales, they continue to perform poorly, this will demonstrate the need to develop novel eating disorder scales or subscales suitable for vegetarians and vegans. Once such tools have been validated in these populations, longitudinal research must be conducted to track eating behaviours and attitudes in individuals as they transition to a vegetarian or vegan diet. In doing so, researchers would be able to establish a potential causal or bidirectional relationship between veg*ism and disordered eating, in turn guiding evidence-based treatment approaches for these populations. Without longitudinal research, it remains unclear whether veg*ism increases the risk of developing an eating disorder, or vice versa.
Conclusion
This study presents the first systematic review to examine the association between vegetarianism, veganism, and disordered eating. Our findings demonstrate a lack of agreement whether veg*ism is associated with greater rates of global disordered eating but does potentially demonstrate an association with greater rates of orthorexia nervosa. It is clear there are additional factors influencing the association between global disordered eating, vegetarianism, and veganism which may be related to the use of eating disorder scales that are not psychometrically fit for these populations. Understanding the psychometric properties of eating disorder scales in veg*ans is vital to ensure such measures are accurately measuring eating psychopathology. In the meantime, caution must be taken when interpreting eating disorder results in both research and clinical settings to ensure veg*ans are not being unfairly pathologized for their dietary adherence.
What is already known on this subject?
Veg*ism may act as a socially acceptable way to restrict food intake and camouflage disordered eating. Some ED scales may be capturing normal veg*an-motivated food choices and behaviours.
What does this study add?
We found a positive relationship between veg*ism and orthorexia nervosa, but no consensus with global disordered eating. There is not enough evidence to support the use of ED scales in veg*ans.
Data availability
Data sharing is not applicable as no new data were created or analysed in this study.
References
McLean CP, Kulkarni J, Sharp G (2021) Eating disorders and meat avoidance: a call to understand the “why” instead of the “what.” Aust N Z J Psychiatry. https://doi.org/10.1177/00048674211037889
Heiss S et al (2021) Vegetarians and omnivores with diagnosed eating disorders exhibit no difference in symptomology: a retrospective clinical chart review. Eat Weight Disord 26:1007–1012. https://doi.org/10.1007/s40519-020-00903-w
Craig WJ, Mangels AR (2009) Position of the American dietetic association: vegetarian diets. J Am Diet Assoc 109:1266–1282. https://doi.org/10.1016/j.jada.2009.05.027
The Vegan Society. Definition of Veganism. n.d; Available from: https://www.vegansociety.com/go-vegan/definition-veganism.
Rosenfeld DL, Burrow AL (2017) The unified model of vegetarian identity: a conceptual framework for understanding plant-based food choices. Appetite 112:78–95. https://doi.org/10.1016/j.appet.2017.01.017
Rosenfeld DL (2019) A comparison of dietarian identity profiles between vegetarians and vegans. Food Qual Prefer 72:40–44. https://doi.org/10.1016/j.foodqual.2018.09.008
Rosenfeld DL (2019) Psychometric properties of the dietarian identity questionnaire among vegetarians. Food Qual Prefer 74:135–141. https://doi.org/10.1016/j.foodqual.2019.01.020
Sieke EH et al (2021) To meat or not to meat: disordered eating and vegetarian status in university students. Eat Weight Disord. https://doi.org/10.1007/s40519-021-01202-8
Timko CA, Hormes JM, Chubski J (2012) Will the real vegetarian please stand up? An investigation of dietary restraint and eating disorder symptoms in vegetarians versus non–vegetarians. Appetite 58:982–990. https://doi.org/10.1016/j.appet.2012.02.005
Heiss S, Coffino JA, Hormes JM (2017) Eating and health behaviors in vegans compared to omnivores: dispelling common myths. Appetite 118:129–135. https://doi.org/10.1016/j.appet.2017.08.001
McLean CP et al (2021) Characteristics and clinical implications of the relationship between veganism and pathological eating behaviours. Eat Weight Disord. https://doi.org/10.1007/s40519-021-01330-1
Klopp SA, Heiss CJ, Smith HS (2003) Self-reported vegetarianism may be a marker for college women at risk for disordered eating. J Am Diet Assoc 103:745–747. https://doi.org/10.1053/jada.2003.50139
Trautmann J et al (2008) Vegetarian students in their first year of college: are they at risk for restrictive or disordered eating behaviors? Coll Stud J 42:340
Lindeman M, Stark K, Latvala K (2000) Vegetarianism and eating-disordered thinking. Eat Disord 8:157–165. https://doi.org/10.1080/10640260008251222
Zuromski KL et al (2015) Increased prevalence of vegetarianism among women with eating pathology. Eat Behav 19:24–27. https://doi.org/10.1016/j.eatbeh.2015.06.017
Bardone-Cone AM et al (2012) The inter-relationships between vegetarianism and eating disorders among females. J Acad Nutr Diet 112:1247–1252. https://doi.org/10.1016/j.jand.2012.05.007
Barthels F, Meyer F, Pietrowsky R (2018) Orthorexic and restrained eating behaviour in vegans, vegetarians, and individuals on a diet. Int J Eat Weight Disord 23:159–166. https://doi.org/10.1007/s40519-018-0479-0
Garner DM et al (1982) The eating attitudes test: psychometric features and clinical correlates. Psychol Med 12:871–878. https://doi.org/10.1017/s0033291700049163
Fairburn CG, Beglin SJ (1994) Assessment of eating disorders: interview or self-report questionnaire? Int J Eat Disord 16:363–370
Donini LM et al (2005) Orthorexia nervosa: validation of a diagnosis questionnaire. Eat Weight Disord 10:e28–e32. https://doi.org/10.1007/BF03327537
Stunkard AJ, Messick S (1985) The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 29:71–83. https://doi.org/10.1016/0022-3999(85)90010-8
Forestell CA, Spaeth AM, Kane SA (2012) To eat or not to eat red meat. A closer look at the relationship between restrained eating and vegetarianism in college females. Appetite 58:319–325. https://doi.org/10.1016/j.appet.2011.10.015
Gearhardt AN, Corbin WR, Brownell KD (2009) Preliminary validation of the Yale food addiction scale. Appetite 52:430–436. https://doi.org/10.1016/j.appet.2008.12.003
Sergentanis TN et al (2020) Vegetarian diets and eating disorders in adolescents and young adults: a systematic review. Children 8:12. https://doi.org/10.3390/children8010012
Moher D et al (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 339:332–336. https://doi.org/10.1136/bmj.b2535
Liberati A et al (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 339:b2700. https://doi.org/10.1136/bmj.b2700
Barthels F, Meyer F, Pietrowsky R (2015) Orthorexic eating behavior. A new type of disordered eating. Sci Res 62:156–161. https://doi.org/10.4455/eu.2015.029
Esterhuizen TM, Thabane L (2016) Con: meta-analysis: some key limitations and potential solutions. Nephrol Dial Transplant 31:882–885. https://doi.org/10.1093/ndt/gfw092
Modesti PA et al (2016) Panethnic differences in blood pressure in Europe: a systematic review and meta-analysis. PLoS ONE 11:e0147601. https://doi.org/10.1371/journal.pone.0147601
Wells GA et al (2021) The Newcastle-Ottawa scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 7th August 2021]; Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.
Barrack MT et al (2019) Disordered eating among a diverse sample of first-year college students. J Am Coll Nutr 38:141–148. https://doi.org/10.1080/07315724.2018.1487345
Bas M, Karabudak E, Kiziltan G (2005) Vegetarianism and eating disorders: association between eating attitudes and other psychological factors among Turkish adolescents. Appetite 44:309–315. https://doi.org/10.1016/j.appet.2005.02.002
Collins M, Quinton S (2020) The inter-relationship between diet, selflessness, and disordered eating in Australian women. Brain Behav 10:e01774. https://doi.org/10.1002/brb3.1774
Gilbody SM, Kirk SFL, Hill AJ (1999) Vegetarianism in young women: another means of weight control? Int J Eat Disord 26:87–90. https://doi.org/10.1002/%28SICI%291098-108X%28199907%2926:1%3C87::AID-EAT11%3E3.0.CO;2-M
Kadambari R, Cowers S, Crisp A (1986) Some correlates of vegetarianism in anorexia nervosa. Int J Eat Disord 5:539–544. https://doi.org/10.1002/1098-108x(198603)5:3%3c539::aid-eat2260050310%3e3.0.co;2-o
McLean JA, Barr SI (2003) Cognitive dietary restraint is associated with eating behaviors, lifestyle practices, personality characteristics and menstrual irregularity in college women. Appetite 40:185–192. https://doi.org/10.1016/S0195-6663(02)00125-3
Micali N et al (2012) Nutritional intake and dietary patterns in pregnancy: a longitudinal study of women with lifetime eating disorders. Br J Nutr 108:2093–2099. https://doi.org/10.1017/S0007114512000256
Paslakis G et al (2020) Prevalence and psychopathology of vegetarians and vegans - results from a representative survey in Germany. Sci Rep 10:6840–6840. https://doi.org/10.1038/s41598-020-63910-y
Robinson-O’Brien R et al (2009) Adolescent and young adult vegetarianism: better dietary intake and weight outcomes but increased risk of disordered eating behaviors. J Am Diet Assoc 109:648–655. https://doi.org/10.1016/j.jada.2008.12.014
Yackobovitch-Gavan M et al (2009) An integrative quantitative model of factors influencing the course of anorexia nervosa over time. Int J Eat Disord 42:306–317. https://doi.org/10.1002/eat.20624
Zickgraf HF et al (2020) Examining vegetarianism, weight motivations, and eating disorder psychopathology among college students. Int J Eat Disord 53:1506–1514. https://doi.org/10.1002/eat.23335
Brytek-Matera A et al (2019) Strict health-oriented eating patterns (orthorexic eating behaviours) and their connection with a vegetarian and vegan diet. Int J Eat Weight Disord 24:441–452. https://doi.org/10.1007/s40519-018-0563-5
Çiçekoğlu P, Tunçay GY (2018) A comparison of eating attitudes between vegans/vegetarians and nonvegans/nonvegetarians in terms of orthorexia nervosa. Arch Psychiatr Nurs 32:200–205. https://doi.org/10.1016/j.apnu.2017.11.002
Delucca GF (2014) An investigation of vegetarianism and eating attitudes based on different types of vegetarian eating patterns. The Chicago School of Professional Psychology: ProQuest Dissertations Publishing
Dorard G, Mathieu S (2021) Vegetarian and omnivorous diets: a cross-sectional study of motivation, eating disorders, and body shape perception. Appetite 156:104972. https://doi.org/10.1016/j.appet.2020.104972
Fisak B et al (2006) Challenging previous conceptions of vegetarianism and eating disorders. Int J Eat Weight Disord 11:195–200. https://doi.org/10.1007/BF03327571
Lacey JM, Zotter DU (2009) Zinc-specific food frequency questionnaire to assess college women’s eating habits. Can J Diet Pract Res 70:204–208. https://doi.org/10.3148/70.4.2009.204
O’Connor MA et al (1987) Vegetarianism in anorexia nervosa? A review of 116 consecutive cases. Med J Aust 147:540–542. https://doi.org/10.5694/j.1326-5377.1987.tb133677.x
Brytek-Matera A (2020) Restrained eating and vegan, vegetarian and omnivore dietary intakes. Nutrients 12:2133. https://doi.org/10.3390/nu12072133
Brytek-Matera A (2020) Interaction between vegetarian versus omnivorous diet and unhealthy eating patterns (orthorexia nervosa, cognitive restraint) and body mass index in adults. Nutrients 12:646. https://doi.org/10.3390/nu12030646
Heiss S, Coffino JA, Hormes JM (2019) What does the ORTO-15 measure? Assessing the construct validity of a common orthorexia nervosa questionnaire in a meat avoiding sample. Appetite 135:93–99. https://doi.org/10.1016/j.appet.2018.12.042
Janelle KC, Barr SI (1995) Nutrient intakes and eating behavior scores of vegetarian and nonvegetarian women. J Am Diet Assoc 95:180–188. https://doi.org/10.1016/s0002-8223(95)00045-3
Norwood R et al (2019) The psychological characteristics of people consuming vegetarian, vegan, paleo, gluten free and weight loss dietary patterns. Obes Sci Pract 5:148–158. https://doi.org/10.1002/osp4.325
Heiss S, Boswell JF, Hormes JM (2018) Confirmatory factor analysis of the eating disorder examination-questionnaire: a comparison of five factor solutions across vegan and omnivore participants. Int J Eat Disord 51:418–428. https://doi.org/10.1002/eat.22848
Barthels F et al (2020) Orthorexic eating behavior in vegans is linked to health, not to animal welfare. Eat Weight Disord 25:817–820. https://doi.org/10.1007/s40519-019-00679-8
Dell’Osso L et al (2018) Prevalence and characteristics of orthorexia nervosa in a sample of university students in Italy. Eat Weight Disord 23:55–65. https://doi.org/10.1007/s40519-017-0460-3
Dell’Osso L et al (2016) Orthorexia nervosa in a sample of Italian university population. Riv Psichiatr 51:190–196. https://doi.org/10.1708/2476.25888
Dittfeld A et al (2017) A study on the relationship between orthorexia and vegetarianism using the BOT (Bratman test for orthorexia). Psychiatr Pol 51:1133–1144. https://doi.org/10.12740/PP/75739
Dunn TM et al (2017) Prevalence of orthorexia nervosa is less than 1 %: data from a US sample. Eat Weight Disord 22:185–192. https://doi.org/10.1007/s40519-016-0258-8
Ferreira C, Coimbra M (2020) To further understand orthorexia nervosa DOS validity for the Portuguese population and its relationship with psychological indicators, sex, BMI and dietary pattern. Eat Weight Disord. https://doi.org/10.1007/s40519-020-01058-4
Heiss S, Timko CA, Hormes JM (2020) Confirmatory factor analysis of the EDE-Q in vegans and omnivores: support for the brief three factor model. Eat Behav 39:101447–101447. https://doi.org/10.1016/j.eatbeh.2020.101447
Herranz Valera J et al (2014) Prevalence of orthorexia nervosa among ashtanga yoga practitioners: a pilot study. Eat Weight Disord 19:469–472. https://doi.org/10.1007/s40519-014-0131-6
Hessler-Kaufmann JB et al (2021) Orthorexic tendencies moderate the relationship between semi–vegetarianism and depressive symptoms. Eat Weight Disord 26:623–628. https://doi.org/10.1007/s40519-020-00901-y
Luck-Sikorski C et al (2019) Is orthorexic behavior common in the general public? A large representative study in Germany. Eat Weight Disord 24:267–273. https://doi.org/10.1007/s40519-018-0502-5
Missbach B et al (2015) When eating right, is measured wrong! A validation and critical examination of the ORTO-15 questionnaire in German. PLoS ONE 10:e0135772–e0135772. https://doi.org/10.1371/journal.pone.0135772
Oberle CD, De Nadai AS, Madrid AL (2020) Orthorexia nervosa inventory (ONI): development and validation of a new measure of orthorexic symptomatology. Eat Weight Disord. https://doi.org/10.1007/s40519-020-00896-6
Parra-Fernández ML et al (2020) Pathological preoccupation with healthy eating (orthorexia nervosa) in a Spanish sample with vegetarian, vegan, and non–vegetarian dietary patterns. Nutrients 12:3907. https://doi.org/10.3390/nu12123907
Braunsberger K, Flamm OR (2019) The case of the ethical vegan: motivations matter when researching dietary and lifestyle choices. J Manag Issues 31:228–222
Dyett PA et al (2013) Vegan lifestyle behaviors. An exploration of congruence with health-related beliefs and assessed health indices. Appetite 67:119–124. https://doi.org/10.1016/j.appet.2013.03.015
American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders, 5th edn. American Psychiatric Publishing, Arlington, VA
Ryman FVM et al (2019) Orthorexia nervosa: disorder or not? opinions of Dutch health professionals. Front Psychol 10:555–555. https://doi.org/10.3389/fpsyg.2019.00555
Roncero M, Barrada JR, Perpiñá C (2017) Measuring orthorexia nervosa: psychometric limitations of the ORTO-15. Span J Psychol 20:E41. https://doi.org/10.1017/sjp.2017.36
Hallit S et al (2021) The relation of orthorexia with lifestyle habits: Arabic versions of the eating habits questionnaire and the Dusseldorf orthorexia scale. J Eat Disord 9:1–102. https://doi.org/10.1186/s40337-021-00455-z
Zickgraf HF, Barrada JR (2021) Orthorexia nervosa vs. healthy orthorexia: relationships with disordered eating, eating behavior, and healthy lifestyle choices. Eat Weight Disord. https://doi.org/10.1007/s40519-021-01263-9
Berg KC et al (2012) Psychometric evaluation of the eating disorder examination and eating disorder examination-questionnaire: a systematic review of the literature. Int J Eat Disord 45:428–438. https://doi.org/10.1002/eat.20931
Rosenfeld DL (2018) The psychology of vegetarianism: recent advances and future directions. Appetite 131:125–138. https://doi.org/10.1016/j.appet.2018.09.011
Ruby MB (2012) Vegetarianism. A blossoming field of study. Appetite 58:141–150. https://doi.org/10.1016/j.appet.2011.09.019
Asher K et al. (2014) Study of current and former vegetarians and vegans: companion to the initial findings. Faunalytics. https://faunalytics.org/wp-content/uploads/2014/12/HRC-Study-of-Current-Former-Vegetarians-Vegans-Dec-2014-Tables-Methodology-1.pdf
Costello A, Osborne J (2005) Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Pract Assess Res Eval 10:1–9. https://doi.org/10.7275/jyj1-4868
Heiss S, Hormes JM, Timko CA (2017) Vegetarianism and eating disorders. Vegetarian and plant-based diets in health and disease prevention. Elsevier Inc, pp 51–69
Sharp G et al (2017) Potential in-class strategies to increase children’s vegetable consumption. Public Health Nutr 20:1491–1499. https://doi.org/10.1017/S136898001700012X
Acknowledgements
The authors wish to thank Alex Lavale of the Monash Alfred Psychiatry research centre for her statistical advice.
Funding
Open Access funding enabled and organized by CAUL and its Member Institutions. The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Author information
Authors and Affiliations
Contributions
CM and GS designed the study. CM performed the systematic search, data extraction, and initial draft of the manuscript. GS conducted extraction and quality bias checks and drafting of the manuscript revisions. All authors assisted with drafting for the final version of the manuscript, including critical revisions.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non–financial interests to disclose.
Ethical approval
The following systematic review did not require Institutional Review Board approval.
Informed consent
Formal consent was not required for this systematic review.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
McLean, C.P., Kulkarni, J. & Sharp, G. Disordered eating and the meat-avoidance spectrum: a systematic review and clinical implications. Eat Weight Disord 27, 2347–2375 (2022). https://doi.org/10.1007/s40519-022-01428-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40519-022-01428-0