Eating disorders in the Arab world: a literature review

Background The prevalence of eating disorders has been assumed to be low in the Arab world, due to the alleged absence of the thin ideal. However, the Arab world is undergoing rapid sociocultural changes, and there are reports of an increase of the desire to be thin. This literature review therefore provides point-prevalence of Arabs at high risk for eating disorders, and a comprehensive synthesis of correlates of eating disorder symptoms, eating disorder-related variables and of a high risk for eating disorders. Method Several electronic databases were searched for published, peer-reviewed studies between 1986 and 2019 involving several key terms. From 317 screened studies, 81, mainly cross-sectional, were included. Preferred Reporting Items for Systematic reviews and meta-analyses was used as guidance and the quality of studies were assessed using the Newcastle-Ottawa scale. Results Estimates of individuals at high risk for eating disorders ranged from 2 to 54.8%. The eating disorder-related variables identified were desire to be thin, body dissatisfaction, disturbed-, and dieting- eating behavior. Identified correlates were increased affluence, media use, western influences, and obesity. An additional finding was that in some cases eating disorders were expressed somatically rather than psychiatrically. Discussion In the Arab world, females were most at risk for eating disorders and eating disorder symptoms. Sociocultural changes gave rise to the thin ideal and the prevalence of obesity, increasing the risk for the development of eating disorder-related variables and eating disorders. The literature on eating disorders in the Arab world suffers from potential limitations due to the use of non-validated assessment tools. Further research is necessary, particularly on the development and validation of a culturally sensitive assessment tool. Improved knowledge is likely to increase the number of people seeking treatment and decrease the stigma of psychotherapy.

will facilitate recognition of eating disorders, encourage people to seek treatment and decrease the stigma of psychotherapy.

Background
Eating disorders (EDs) have a significant impact on the well-being of affected individuals [1]. This includes comorbid obesity [2], depressive symptoms, anxiety [3] substance abuse, suicide attempts [4], and high rates of mortality and relapse [5]. Since EDs have historically been associated with Caucasian females in developed Western countries with high socio-economic status [6], they have been perceived as culturally bound syndromes [7]. Taking this in consideration most studies regarding EDs have been heavily concentrated in Western countries [8].
Cultural factors are essential to understanding the development of EDs [9] and the main feature identified associated with EDs in the West is a thin body ideal [6], which is presented to the society as achievable by dieting and exercising [8]. Traditional Arab notions of beauty are different, with the curvy body ideal associated with fertility and wealth [10] EDs were assumed not to afflict Arabs 1 [10,11] hence EDs were not reported in the Arab world until 1986 [12]. However, from then on [10,11] reports on the thin ideal in the Arab world steadily increased [13][14][15].
Recent studies have shown that EDs occur globally [16,17], EDs occur particularly in cultures in transition as they adopt Western values [7,18], illuminating the interplay between culture and psychopathology [19]. This is relevant to the Arab world, since the oil boom of the 1970s and the consequent increased affluence [20,21], it has been undergoing rapid sociocultural and socioeconomic changes [14,22]. The oil boom led to the arrival of Western companies, Western expatriates [18,23], and to increased exposure to Western culture [8].
The sociocultural changes associated with acculturation include adopting the language, lifestyle, values, and beliefs of other cultures [24]. In the Arab world, increased exposure to Western media [20,[25][26][27], and increased contact with expatriates [28,29] has also led to a rise in the popularity of the thin ideal [10,30], and to increased levels of dieting, body dissatisfaction and EDs [8]. However, the theory of Westernization assumes that the Western culture is transferrable to Arab cultures [31], the oil boom has also been associated with elevated levels of industrialization [18] including increased technology, affluence, and higher education [32,33]. Industrialization in particular has coincided with changes in the types of food available [7,34], and these are instrumental in a rise in obesity [28,35,36]. Together, these changes have contributed to elevated levels of non-communicable diseases such as diabetes mellitus, hypertension [20], and psychological problems [37] including ED-related variables and EDs [18,[38][39][40].
Kraemer and colleagues [41][42][43] have provided us with a terminology to address how culture and EDs may be associated. An important distinction is that between a correlate and a risk factor. A correlate is a potential risk factor that is measured at the same time as an ED. The correlate and ED are associated with each other, but precedence of the correlate to the onset of an ED is not proven [41,42]. Therefore correlates cannot be defined as risk factors. Correlates are identified in cross sectional studies [41][42][43]. Currently identified correlates are increased affluence [6,43], obesity [44,45], and societal changes such as industrialization [46], globalization [6], and acculturation [38,46]. If it is established that an associated variable precedes the onset of an ED, this variable may be indicated as a risk factor [42]: gender is an example [41]. A risk factor identifies individuals who are at elevated risk for future emergence of an ED [43], and risk factors can best be identified by longitudinal studies [41,42]. Initiation of the onset of an ED can be indicated by ED-related variables [43]. ED-related variables are an early sign of an ED which indicate vulnerability to develop an ED. However, on their own ED-related variables do not meet the criteria for an ED as described in the DSM-5 [47]. Examples of ED-related variables are disturbed eating behavior, self-reported dieting, an elevated desire to be thin, and body dissatisfaction [41].
To distinguish between various levels of ED-pathology, commonly the following types are described: ED symptoms, at high risk for ED and full syndrome ED [41][42][43]. ED symptoms are symptoms that are defining features of an ED [43] such as binge eating, and compensatory behaviors, which can be measured by self-report and clinical interviews. If the frequency of ED symptoms meets the criteria of an ED, including clinically significant functional impairment [6], an individual can be classified as suffering from a full syndrome ED [43,47]. Full syndrome EDs are best diagnosed with a clinical interview [43]. Individuals who score above a clinical cut-off on an ED-screening instrument are considered to be at high risk for an ED [43,48]. This review will differentiate between ED-related variables, ED symptoms, and high risk for EDs. Arabs with elevated levels of ED-related variables and ED symptoms are at risk for EDs, as are Arabs who score above a clinical cut-off on an ED-screening instrument, who will be referred to as Arabs at high risk for EDs. Correlates of ED-related variables, ED symptoms, and scores above a clinical cut-off on an ED-screening instrument will also be identified. Identification of correlates and ED-related variables is likely to facilitate identification of risk groups for prevention programs [43,49]. At the same time, preventative and treatment programs cannot be copied from Western societies [50]. They need to be adapted to the relevant sociocultural context [51,52].
Although EDs have been reported in the Arab world [28,29,36,[53][54][55], there is limited valid data [56], and the substantial differences in the expression of EDs among different cultures [7,52,57,58] are yet to be taken into account. For example, somatic expression of EDs is more likely in the Arab world than in the West [52,58]. Some Arabs attribute a restrictive food pattern to somatic complaints [59], while there are also reports of body dissatisfaction and fear of fatness [52,58]. Examples of somatic complaints are stomach ache, feeling bloated or an absence of appetite [59]. In addition, perceptions of body image also differ between cultures [52].
There are a number of comprehensive reviews of the nature of EDs in Western countries [5,43,51,60,61]. However, as the Arab world has undergone rapid sociocultural changes [14,22], EDs could be more culturally reactive than culturally bound [8]. An overview of EDs in the Arab world is needed. Unfortunately, due to a lack of appropriate and valid data, it is not feasible to examine how the prevalence of EDs in the Arab world has changed over the years. Still, data from ED-screening instruments can help identify individuals at high risk for EDs, and enable examination of associations between ED-related variables, correlates and ED symptoms. In this review, we therefore aim to provide estimates regarding the point-prevalence of Arabs at high risk for EDs, with ED symptoms above threshold values for caseness, and to provide a comprehensive synthesis of relevant studies of correlates and ED-related variables. The countries included here are those together referred to as "the Arab world", which is a part of the Eastern Mediterranean Region [62], also referred to as the Middle East [63]. Non-Arab countries such as Cyprus and Turkey are excluded [64]. To the best of our knowledge, this is the first review summarizing EDs in the Arab world.

Search strategy
The primary search was conducted in Web of Science, PubMed (Medline) and Google Scholar databases from 1986 up to July 2019 by BM, and involved key terms related to ED prevalence in the Arab world. All combinations of ("Eating disorders" OR "Anorexia Nervosa" OR "Bulimia Nervosa" OR "Binge Eating Disorder" OR "Disturbed Eating Behavior" OR "Eating Attitudes" OR "Dieting" OR "Body Image" OR" Body Satisfaction" OR "Obesity") AND ("Arab" OR "Middle East" OR "Gulf" OR "United Arab Emirates (UAE)" OR "Saudi" OR "Oman" OR "Qatar" OR "Bahrain" OR "Kuwait" OR "Lebanon" OR "Palestine" OR "Jordan") were searched. To be included, articles had to be peer reviewed. This resulted in 14,656 hits. Subsequently, more records were added from other sources (n = 13). After duplicates were removed, 14,629 studies remained. The titles were then screened for eligibility. Eligible studies were conducted in Arab countries in the Middle East, providing estimates regarding the point prevalence of individuals at high risk for EDs, ED symptoms and ED-related variables, and reporting correlates of scores above a clinical cut-off in ED-screening instruments, ED symptoms, and ED-related variables. Records (n = 14,312) studying, for example, diabetes mellitus, metabolic syndrome or nutrition related to other diseases, were excluded. When there was doubt, the abstract and discussion section were further screened. This resulted in 72 relevant studies, plus 9 additional studies identified from their lists of reference. The additional sources of studies were four websites involving statistical data such as GCC stat (Fig. 1). Most studies were conducted in the UAE (n = 11), followed by Jordan and Saudi Arabia (n = 7). A further six studies were conducted in Kuwait, five in Egypt, four each in Oman, and Qatar. Three among Palestinian populations, two each in Bahrain, Lebanon, Iran, and one each in Algeria, and Libya.

Inclusion criteria
Inclusion criteria were that the studies involved Arab Middle Eastern countries [63] of the Eastern Mediterranean Region [62]. These are West Asian countries (Lebanon, Jordan, Iran), the Gulf (Bahrain, Kuwait, Qatar, UAE, Saudi Arabia, Oman), Egypt, and Palestinian populations. Though, a non-Arab country, Iran is included because of the several cultural similarities (e.g. hijab and full body coverage for females in public) [64]. All studies investigated at least one ED symptom or ED-related variable. In cases where obesity was studied, study of at least one additional ED behavior was required. There were no restrictions on age or gender. All studies reported point prevalence, the prevalence at a certain point in time [51]. Additional studies examined correlates of EDs, ED symptoms or ED-related variables.

Exclusion criteria
Studies involving non-Middle Eastern countries, and non-Arab Middle Eastern countries were excluded [63], as were studies focusing only on obesity, and qualitative studies without empirical data.

Assessments
While correlates and ED-related variables can be accurately measured by self-report [43], diagnostic interviews are preferable for ED symptoms. Since diagnostic concordance between diagnostic interviews and self-report is moderate at best [65], and EDs can best be assessed by diagnostic interviews [43], questionnaires including the SCOFF, EAT, and EDE-Q may at best provide prevalence estimates of Arabs at high risk for EDs, indicated by elevated scores at ED-screening instruments. However, since for several Arab populations norms of ED-screening instruments were not available, norms of either Western populations or other Arab populations were used. Therefore, these instruments indicated elevated scores on ED-screening instruments in the target country using the norms of another country. In this study, Arabs who scored above a clinical cut-off are referred to as Arabs at high risk for EDs, regardless of the norms used.

Data abstraction and quality assessment
This review is based on studies published in English between 1986 and 2019; a large timeframe was chosen due to the limited number of studies on EDs conducted in the Arab world. Since only studies on point prevalence were found and none on incidence rate and lifetime prevalence, prevalence refers to point prevalence. Most of the studies were published in peer reviewed journals. Two theses were also included [66,67]. In total, 81 studies met the inclusion criteria. Almost all the studies were quantitative; most were cross sectional, although three were case control studies [35,68,69]. One validation study was included [54], and longitudinal data were not available. Eight cross sectional studies involved cross cultural aspects [11,23,26,39,52,58,64,70]. Author BM assessed the risk of bias on study validity. Discrepancies or uncertainties were resolved by discussion with either the second or the third author. Data extraction was done independently by BM, taking into consideration study characteristics, demographics, study results, and key study outcomes. Quality of included studies was assessed using the Newcastle-Ottawa scale [71], a seven item scale that investigates power, research design, sample, recruitment, and statistical analysis (see supplementary material). The maximum score was 10 stars, based on selection (maximum five stars), comparability (maximum two stars) and outcome (maximum three stars). Only studies with at least four stars were included in the review. Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRIS MA) was used as guidance ( Fig. 1) [72].

Quality assessment
Of 317 studies assessed for eligibility, 81 studies met the minimum criteria of four stars on the Newcastle Ottawa  [71]. However, the 81 studies included had several limitations, these studies either involved a selected group of users or had a selective sample. For example, the association between media and EDs in Kuwait was assessed only on female adolescents at one university [73], disturbed eating behavior only amongst adolescent girls in the UAE, and only among the inhabitants of al Ain province [20], and in Saudi Arabia, only in the city of Hail [67], all involving academically inclined schoolgirls [39]. Most of the studies involved a population of university students [74]. In addition, reporting in some studies was incomplete (lack of demographic data, such as mean age, and marital status) [67,75]. In other studies mean and SD [26,29,39,54,55,64,[76][77][78][79] or percentage that scored above the clinical cutoff [52,58,64] on the ED-screening instrument were not reported.
Another limitation was that power calculations were not reported in some studies, so their sample size was not justified and was low. This was the case (N=100) in an Omani [29], a Saudi [67], and (N=104) an Iranian [64] sample. Studies examining the association between Westernization and ED had only small samples (N<100) in the UAE [40], and in Iran [64].Another limitation was that several studies used non-validated ED-screening instruments. Several studies, for instance, translated the EAT-40 without adaptations, and used Western norms [20,28,39]. In another study on ED in Saudi Arabia, the EAT-26 was translated without adaptations, and applied using Western norms [67]. The Saudi version of the EAT-26, which included the Western norms was also used in non-Saudi countries [14,18,26,40,52,53,58,74,79,80]. Other studies investigated adults using a version of the Saudi EAT-26 validated in a high school population [26]. Musaiger and Al-Mannai [73] also examined a Kuwaiti sample using questions validated in a Bahraini sample. In addition, two studies assessed disturbed eating behavior among adolescent Omani and European expatriates residing in Oman using the EAT26 and the EDI-2. These studies found that Omanis had significant higher EAT26 scores, while European expatriates had significantly higher EDI-2 scores. Omanis and European expatriates scored significantly different on both ED-screening instruments [52,58], possibly due to the fact that a Saudi version including Saudi norms were used for the EAT26 and EDI-2 [78,81]. Several studies also failed to provide essential information on the norms used, and some studies provided no information at all. One study, in Kuwait [75], did not provide information about how their statistical analysis was conducted. Furthermore, none of the studies on binge eating behavior mentioned the frequency of binge eating episodes, or whether the binges were of a subjective or objective nature [29,70,74,77]. All the studies discussed relied on self-report measures, which are prone to socially desirable responding [75]. The collectivistic culture of the Arab world makes people hesitant to report individual desires [76], so the use of self-report measures makes studies vulnerable to under-or overestimation of complaints [77].
In conclusion, most of the studies included in this synthesis have limitations. These include the absence of psychometric validation of assessment tools for the Arab world, and the absence of necessary data such as demographic data, power calculation proper assessment, and description of ED-pathology.
Eating disorder-related variables Table 1 presents a summary of 19 studies including EDrelated variables. These were: desire to be thin, body dissatisfaction, disturbed eating behavior, and dieting [43]. Disturbed eating behavior was involved in nine studies, body dissatisfaction in ten studies, desire to be thin in seven, and dieting in five cross sectional studies. EDrelated variables were identified by self-report in all studies, except for one, which used both interview ánd self-report data [39]. All ED-related variables were mainly assessed in female adolescents.

Dieting
The findings suggested that, as a whole, approximately 40% of the Arab population is on a diet [20,35,70,88], and this is true for adults and adolescents of both genders [20,35,70]. Saudis were at the lowest risk of dieting [13].

Desire to be thin
Though findings were quite inconsistent, overall Arabs from the UAE had the highest desire to be thin [13][14][15], whereas Qataris appeared to be minimally affected by the thin ideal [15]. A risk factor of the desire to be thin was adulthood over adolescence [10, 13-15, 20, 87, 89]. Table 2 provides a summary of 11 studies that examined eating disorder symptoms in the Arab world. Binge eating and restrained eating behavior were each identified in seven cross sectional studies and compensatory behavior (n = 2) in a few studies. All ED symptoms were identified by self-report among adolescents: high school and/or university students.

Binge eating
Kuwaiti and Egyptian Arabs appeared to be most at risk of binge eating. However, binge eating was only assessed among adolescents, and the findings with regard to gender were inconsistent [11,29,74,77,84,90,92].

Restrained eating behavior
Off the seven studies included in this synthesis that studied restrained eating behavior, all of them reported restrained eating behavior in about one-third of Arabic females. Arab adolescents who reported emotional eating and Arabs with obesity were particularly at risk of restrained eating behavior [20,23,29,66,67,77,87].

Compensatory behavior
Around 7.5% of female Arab adolescents reported usage of laxatives [70,93]. Table 3 provides a summary of 27 studies that examined prevalence estimates of Arabs at high risk for EDs. EDscreening instruments were conducted among both males and females, and among adolescents and adults. Arabs at high risk for anorexia nervosa (AN) and bulimia nervosa (BN) were identified in different countries by seven cross-sectional studies. Only one study reported Arabs at high risk for binge eating disorder (BED) and other specified feeding or eating disorders (OSFED) [90]. While most ED assessment tools were self-report measures, in one cross sectional study participants were interviewed using a self-developed interview [39]. Another study used the Structured Clinical Interview for DSM-IV (SCID) [76]. In one case control study among health care clinic attendees, EDs were not  [69]. Scores above a clinical cut-off on EDscreening instruments were reported without classification specification in 18 studies (Table 3).
Correlates Table 4 presents a summary of 32 studies on correlates of a high risk for EDs, ED symptoms and ED-related variables. In Western countries, increased affluence and obesity appear to be the major correlates [44]. These factors may also play a role in Arab communities [76], as the Arab world experiences Western and media influences (social media, TV, western advertising and magazines) [84], which may be correlates of the development of EDs [23]. Increased affluence was identified in five cross sectional studies, Western influences in 11 cross sectional studies, media use in five cross sectional   studies, and obesity in nine cross-sectional studies. Most were identified by self-report. However, one study investigated the association between media use and EDs in several countries with a self-developed interview [27], and another study examined EDs with the SCID [76].

Western influences
Exposure to Western influences was associated with EDrelated variables, especially the desire to be thin [13,22,37,70]. Examples of such Western influences were media [14,31], traveling abroad [35], living abroad [90,92,97], and contact with expatriates [13,18,35,37]. Arabs with greater assimilation with the Western culture were potentially at more at risk for EDs [22]. Arabs in Gulf countries (Oman, Saudi Arabia, UAE) were particularly vulnerable to Western influences. However, besides Western influences vulnerability for EDs might also be associated with industrialization, as the Gulf also has the highest level of industrialization within the Arab world [13,37].

Obesity
A higher BMI was positively associated with ED-related variables [74,[104][105][106], ED symptoms [24,62] and high risk for EDs [22]. This is of concern because the Arab world has one of the highest rates of obesity in the world [107], in Saudi Arabia in particular [108], followed by Bahrain, Egypt, Jordan, Syria and Oman [91].

Cultural differences in eating disorders
Some studies suggest that, rather than causing a rise in EDs, Western influences are associated with a shift in symptomatology. The four cross-sectional studies summarized in Table 4 examined cultural differences in EDs.

Cultural differences
Interestingly, predominant identification with a Middle Eastern culture was associated with a risk for AN [29,64], while predominant identification with a Western culture was associated with a risk for BN [64], for instance among European expatriates resident in Oman, among Omani Arabs [29] and among Iranians [64].
Omani Arabs tended to express their ED symptoms somatically (e.g. bloated stomach, nausea, reduced, fluttering feeling in the stomach, throat discomfort, etc.), while Western populations were more likely to express them psychiatrically [52,58].

Discussion
The aim of this study was to provide estimates regarding the point-prevalence of Arabs at high risk for EDs, with ED symptoms and ED-related variables. An additional aim was to provide a comprehensive synthesis of relevant studies of correlates and ED-related variables. Although EDs occur both in Western [60] and in Arab societies [76], there are no official statistics available for in the Arab world [96] and prevalence has rarely been studied [90]. The prevalence of full syndrome EDs have not been reported in the Arab world, except for one case study [97]. Perhaps this is because EDs are not viewed as common disorders in the Arab world [69]. This synthesis found that in the Arab world, 13-55% is at high risk for EDs, the prevalence was higher among females than in males [16, 18, 20, 26, 28, 29, 36, 39, 40, 52-55, 58, 64, 66, 67, 69, 70, 74, 76, 77, 79, 80, 84, 87, 89, 91], and the prevalence appeared to increase during adolescence [20,76,98,99]. There were also some indications symptomatology was culturally reactive [12,29,52,58,64,87]. EDs are perceived as diseases of globalization [6]. The rapid sociocultural changes in the Arab world [20] since   Use of internet and reading magazines associated with dieting to lose weight, media influence 2-3 times higher in obese than in non-obese females, watching TV not associated with body shape concern.

Clinical impact
Development of this synthesis of EDs in the Arab world has important clinical implications, as most studies examining ED prevalence used Western assessment tools [90]. Such utilization of potentially insensitive assessment tools and potentially inappropriate norms may lead to underestimation of symptom severity [107], elevated rates of undiagnosed EDs, and to a lack of knowledge and (public) awareness about EDs and its correlates [108]. This in turn hampers timely and proper treatment of EDs [69]. In the Arab world, the situation is further complicated by the lack of specialized therapists and treatment facilities [97]. Another problematic aspect of the lack of popular knowledge about EDs is stigmatization, leading to delayed help seeking [108]. Since it is culturally unacceptable to discuss personal matters with non-family members, only 0.3% of Arabs with an ED seek help, compared to 20% of Westerners who do so [70]. More knowledge about EDs might decrease the current preference for self-or family treatment, and therefore stigma associated with psychotherapy [70]. In order to increase knowledge psycho-education can be offered at highschools, including a parental program. Once college and high-school social workers are better educated in recognizing EDs they can refer their students to mental health care clinics. In addition, overweight Arabs tend to seek bariatric surgery, while currently they are in general not screened for EDs. Screening for EDs and refer Arabs at high risk for EDs might reduce delayed help seeking. Last, to decrease stigmatization, it might be beneficial to let influencers share their personal story of recovery of an ED, since Arabs are extensive social media users.
Two studies in Oman reported that, as in several Asian studies [59,105,106], EDs were explained somatically rather than psychiatrically [52,58]. This may lead to a failure to accurately recognize behaviors as ED symptoms, and so to delayed help-seeking and greater symptom severity [109,110]. In addition, in Saudi Arabia people with EDs tended to seek help only after experiencing somatic complaints, such as kidney failure or diabetes mellitus.
Although EDs have been studied in the Arab world, there are methodological shortcomings to many of the studies currently available. Arabs who display an increased risk for EDs, are likely to benefit from preventative programs, as selective prevention programs display larger effects than universal prevention programs [49,[111][112][113]. The risk groups in the Arab world are individuals with elevated levels of body dissatisfaction, desire to be thin, and self-reported dieting [13,20,23,26,70,76,85,103]. In addition, the more high affluent societies and Arabs with extensive (social) media use, and more prone to Western influences are also at risk [38,104]. Counteracting the elevated prevalence of obesity in order to prevent associated health risks [80], accompanied by psycho-education in order to prevent maladaptive weight-loss strategies, may be beneficial.

Future directions
Most assessment tools included in this synthesis have been devised for use in Western populations. Due to the potential bias, it is recommended to use ED-screening instruments adapted for use in the Arab world. Future research might therefore consider the development and validation of cultural sensitive assessment tools [66]. One relevant adaptation of self-report measures is that fasting may be motivated by religion [77,96] and should only be considered pathological if motivated by weight and shape concern. Other relevant adaptations in some Arab countries might relate to the strict gender separations and related cultural norms and behaviors, e.g., a proper attention to distinguishing between situations where females do and don't cover themselves [77].
Another potential bias of the studies included in this synthesis is the usage of Western norms. In general, scores on ED-screening instruments are significantly lower in Western populations than in Arab populations [28,80,114,115], usage of Western norms might therefore reflect bias. While some assessment tools have been validated in other Arab countries, they potentially may not be sensitive to nuances across various Arab communities. Efforts to assess EDs in the Arab world have therefore potentially been limited due to the absence of psychometric validation of assessment tools, especially in clinical samples. In order to understand the culture, validation studies should be supplemented with interview data. A culturally sensitive assessment tool will help minimize inconsistencies found among studies in the Arab world.
To elucidate correlates and ED-related variables, researchers need to examine levels of acculturation, internalization of the thin ideal, increased affluence and media exposure, as they may mediate body image dissatisfaction and disturbed eating behavior. As most studies included in this synthesis were conducted in countries dealing with rapid sociocultural changes, additional studies in the Arab world with limited exposure to Western influences and limited industrialization are necessary [116,117].
Recently, studies are beginning to address the possible somatic symptom presentation of EDs in the Arab world [52,58]. We recommend that, during diagnostic interviews, clinicians pay attention to the possibility of this understudied phenomena. In addition, it is worth further investigating the somatic attribution of EDs in the Arab world. Confirmation of this theory will facilitate recognition of EDs in the Arab world.