Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

https://doi.org/10.1007/s40519-022-01452-0

ORIGINAL ARTICLE

The prevalence and risk factors of screen‑based disordered eating

among university students: a global systematic review, meta‑analysis,
and meta‑regression
Omar A. Alhaj1 · Feten Fekih‑Romdhane2,3 · Dima H. Sweidan1 · Zahra Saif4 · Mina F. Khudhair5 ·
Hadeel Ghazzawi6 · Mohammed Sh. Nadar7 · Saad S. Alhajeri8 · Michael P. Levine9 · Haitham Jahrami4,5

Received: 20 April 2022 / Accepted: 12 July 2022 / Published online: 4 August 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022

Abstract
Purpose The purpose of this review was to estimate the prevalence of screen-based disordered eating (SBDE) and several
potential risk factors in university undergraduate students around the world.
Methods An electronic search of nine data bases was conducted from the inception of the databases until 1st October 2021.
Disordered eating was defined as the percentage of students scoring at or above established cut-offs on validated screening
measures. Global data were also analyzed by country, research measure, and culture. Other confounders in this review were
age, BMI, and sex.
Results Using random-effects meta-analysis, the mean estimate of the distribution of effects for the prevalence of SBDE
among university students (K = 105, N = 145,629) was [95% CI] = 19.7% [17.9%; 21.6%], I2 = 98.2%, Cochran's Q p
value = 0.001. Bayesian meta-analysis produced an estimate of 0.24, 95% credible intervals [0.20, 0.30], τ = 92%. Whether
the country in which the students were studying was Western or non-Western did not moderate these effects, but as either
the mean BMI of the sample or the percentage of the sample that was female increased, the prevalence of SBDE increased.
Conclusions These findings support previous studies indicating that many undergraduate students are struggling with disor-
dered eating or a diagnosable eating disorder, but are neither receiver effective prevention nor accessing accurate diagnosis
and available treatment. It is particularly important to develop ever more valid ways of identifying students with high levels
of disordered eating and offering them original or culturally appropriate and effective prevention or early treatment.
Level of evidence I, systematic review and meta-analysis.

Keywords Adolescences · Body image · Body mass index · Eating disorders · Feeding and eating disorders

Introduction the Diagnostic and Statistical Manual of Mental Disorders

(DSM), currently in its 5th edition) [4]. The latter is used
Eating disorders are serious mental illnesses that usually much more often for research purposes, such as epidemio-
begin in adolescence [1], and in many instances recovery logical studies. Anorexia Nervosa (AN), Bulimia Nervosa
requires intensive professional treatment and support [2]. (BN), Binge Eating Disorder (BED), Avoidant/Restrictive
Ideally, these disorders are diagnosed by a professional Food Intake Disorder (ARFID), and Other Specified Feed-
multidisciplinary healthcare team after a comprehensive ing and Eating Disorders (OFSED) are the five major eating
physical and psychological assessments against established disorders listed in the DSM-5 [4]. Each eating disorder has
diagnostic criteria of mental disorders. The two commonly its own set of criteria based on extensive research and clini-
used diagnostic systems are The International Classifica- cal experience [4].
tion of Diseases (ICD), currently in its 10th edition [3], and The causes of eating disorders are unknown, although
there is general agreement that a variable and complicated
combination of biological, psychological, social, and cul-
* Haitham Jahrami tural risk factors increase the probability of eating disorder
[email protected] [2]. According to Global Burden of Diseases, Injuries, and
Extended author information available on the last page of the article Risk Factors Study (GBD) 2019 eating disorders impact

13
Vol.:(0123456789)
3216 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

about 42 million people globally [5], and they are frequently to ascertain the correspondence of those interview data
misdiagnosed and undertreated [6]. Moreover, their preva- with internationally agreed-upon diagnostic criteria [13]
lence worldwide is well-documented and appears to be For example, consider the findings of a recent systematic
increasing [7–9]. review and meta-analysis of the prevalence of eating dis-
The prevailing view of eating disorders is that they are orders and disordered eating in Western Asia [14]. The
categorically distinct patterns of maladaptive eating habits prevalence of disordered eating, as measured by three
linked to profound cognitive modification centered on the widely used screening tools (see Table 1) was: Eating
overvaluation of weight and shape as determinants of one's Attitudes Scale 26 (EAT-26) and Eating Attitudes Scale
identity and worthiness [1, 2]. In this regard, many experts 40 (EAT-40) = 22.1%; Sick, Control, One Stone, Fat, Food
in the field believe that “disordered eating” (defined and (SCOFF) questionnaire = 22.3%; and the Eating Disorder
discussed below) is qualitatively different from the eating Examination-Questionnaire (EDEQ) was 8.0%. On the
disorders [10–12]. For this reason, screening tests aimed other hand, from those studies using semi-structured inter-
at measuring patterns of disordered eating are thought not views against established criteria ICD/DSM, the estimated
to be a good proxy for estimating the prevalence of eating prevalence of anorexia nervosa was 1.6%, while the fig-
disorders based on large samples. Screening tests are best ures for bulimia nervosa and eating disorder not otherwise
used in a typical two-stage design, in which people who specified (EDNOS [BED + OSFED]) were 2.4% and 3.5%,
were found positive for the screening test criteria then par- respectively [14].
ticipate in semi-standardized or standardized interviews

Table 1  Detailed description of the clinical measures involved in the systematic review and meta-analysis of disordered eating among university
students, psychometric properties, cut-off points and full citation
Measure/Scale Cut-off point Psychometric properties

ANIS ≥ 65 Cronbach's α ranged from 0.80 to 0.90. In the three samples the ANIS total score correlated 0.41 to 0.51 with the
28- item General Health Questionnaire, and 0.15 to 0.26 with the percentage of ideal body weight [184]
BEDS-7 – Cohen’s Kappa = 0.827 [185]. Sensitivity = 100%, specificity = 38.7% [186]
DEBQ – Cronbach's α ranged from 0.80 to 0.95. All Pearson’s correlation coefficients assessing interrelationships between
scales (for restrained, emotional, and external eating) were significant, indicating that the measures have a high
internal consistency and factorial validity [187]
EAT-26 ≥ 20 Cronbach's α = 0.90. EAT-26 correlates highly with the original EAT-40 scale (r = 0.98)
EAT-40 30 Cronbach's α = 0.94. Sensitivity = 35.3%, specificity = 88.8%, positive predictive value = 24.0%, and negative
predictive value = 93.2% [182]
EDDS 16.5 Cronbach's α = 0.89. Anorexia nervosa: Sensitivity = 93%, specificity = 100%, positive predictive value = 93%,
negative predictive value = 100%
Bulimia nervosa: Sensitivity = 81%, specificity = 98%, positive predictive value = 97%, negative predictive
value = 96%
Binge-eating disorder: Sensitivity = 77%, specificity = 96%, positive predictive value = 95%, negative predictive
value = 93%
EDE-Q ≥4 Cronbach's α for the global score = 0.90
Women diagnosed with eating disorders scored significantly higher on the EDE-Q than the control women (sensi-
tivity = 0.83, specificity = 0.96, positive predictive value = 0.56) [188, 189]
EDI ≥ 50 Cronbach's α ranged from 0.82 to 0.90. Sensitivity = 52.9%, specificity = 85.2%, positive predictive value = 26.4%
EDS-5 – Cronbach's α ranged from 0.83 to 0.86. Sensitivity = 0.90 and specificity = 0.88
ORTO-11 < 25 Cronbach's α ranged between 0.74 and 0.83. Sensitivity = 75% and specificity = 84% [190]
ORTO-15 < 40 Cronbach's α = 0.83. The ORTO-15 showed significant associations with eating psychopathology (EAT-26 and
SR-YBC-EDS; range r = 0.64 – 0.29; p < 0.05) [191]
Q-EDD - Cohen’s Kappa = 0.94. Sensitivity = 0.97, specificity = 0.98, positive predictive power = 0.94, and negative predic-
tive power = 0.99
SCOFF ≥2 kappa statistic = 0.82. Sensitivity = 100%; specificity = 87.5%; and positive predictive value = 90.6%
WCS ≥ 52 Cronbach's α = 0.65, 0.61, and 0.63 at ages 5, 7, and 9 years [192]. Skewness values for all items ranged from
0.02 to 0.95; and Kurtosis values ranged between -0.83 and -0.53 [193]

ANIS Anorexia Nervosa Inventory for Self-Rating [184, 194]. BEDS-7 the 7-Item Binge-Eating Disorder Screener [186]. DEBQ The Dutch
Eating Behavior Questionnaire [187]. EAT-26 Eating Attitude Test-26 [195]. EAT-40 Eating Attitude Test-40. EAT-40 Eating Attitude Test-40
[195]. EDDS Eating Disorder Diagnostic Scale [196]. EDE-Q Eating Disorder Examination – Questionnaire [197]. EDI Eating Disorder Inven-
tory [195, 198, 199]. EDS-5 Eating Disorder Scale [200]. ORTO-11 ORTO-11 [190]. ORTO-15 ORTO-15 [201]. Q-EDD The Questionnaire for
Eating Disorder Diagnoses [202]. SCOFF Sick, Control, One Stone, Fat, Food [203]. WCS the Weight Concerns scale [204]

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3217

Disordered eating developmental period, one general context associated with

increased risk for eating disorders is the transition, for some
By comparison with the criteria for the eating disorders, people, from high school to college or university. University
the widely used phrase "disordered eating" is a broad con- students are under increased pressure to conform to body
struct that encompasses unhealthy (i.e., disordered) relation- and appearance ideals because romantic/sexuality expec-
ships with food, exercise, body weight, and one's body/body tations and peer comparisons, intensified by social media,
image. It sometimes refers to the presence of individual fea- grow alongside parental and academic demands for competi-
tures of the clinical syndromes, and other times to unhealthy tive excellence, if not perfection [21–23]. The confluence of
attitudes and practices such as negative body image and these pressures, along with the distinct possibility of spe-
calorie-restrictive dieting. Typically, it is assumed that the cific stressors such as parental divorce or sexual harassment/
constituents of disordered eating are less severe than the assault at college, amplify general sociocultural risk factors
syndromes codified by DSM and ICD. Nevertheless, as is for disordered eating and eating disorders.
the case for negative body image, disordered eating is in and The transition to college or university also increases the
of itself a public health problem in many countries because it probability of initiation or more frequent use of cognitive
is associated with a number of negative health consequences, enhancers and psychostimulants to boost weight/shape/
including depression, anxiety, and binge drinking [15–17]. stamina management and cognitive capabilities. Drugs with
Yet, “disordered eating” is used loosely in the literature, potential connections to disordered eating, as well to pur-
and rarely is it defined theoretically. Smolak and Levine have ported success at (or at least coping with) college and its
argued that disordered eating is defined by (a) “subclinical” social life, include nicotine, caffeine in coffee and energy
but unhealthy, maladaptive, and misery-inducing levels of drinks, alcohol, stimulants, and dietary and ergogenic sup-
negative body image, weight and shape concerns, and calo- plements [24–27]. Moreover, pressures contributing to dis-
rie-restrictive dieting and/or binge eating [10–12]; plus (b) ordered eating and eating disorders are greater for university
at least several of the following: individual eating disorder students who fall into one or more of the following catego-
symptoms such as self-induced vomiting after eating; abuse ries: identify as female; are LGTBQ + ; are involved in the
of laxatives, diuretics, diet pills, and exercise; unrealistic performative arts (e.g., dance) or certain competitive sports;
beauty standards, including an idealization of thinness; irra- or are overweight or obese [12, 28, 29].
tional and maladaptive beliefs about body fat and fat people, The reported prevalence of SBDE among various, differ-
often coupled with a high drive for thinness; and harsh self- ent types of samples of university students in the literature
surveillance and self-criticism, often in transaction with low varies from 3.1% (Liao et al. 2006) to 74.5% [30]. Variations
and unstable self-esteem [10–12]. between samples are likely due to variability in sampling
Longitudinal risk factor research consistently shows that methodology, sex, age, BMI, measure, and country, but this
negative body image and disordered eating are perhaps the has not been investigated in a systematic way. A recent sys-
best predictors of the development of full-blown eating dis- tematic review and meta-analysis by our group showed that
orders, at least in adolescent girls and adult women [13]. among medical students higher BMI, Westernized culture,
Another way of looking at disordered eating as an “at risk” and the research tool used were the main confounders [31].
status is the fact that its components constitute many of The explicit purpose of this meta-analysis is to provide
the items making up the measures used to screen people to an overall or 'absolute' estimate of the prevalence of SBDE
determine, relatively quickly and at low cost, who is “at risk” among university students as a population at risk for eating
for actually having an eating disorder upon closer examina- disorders and for the distress and comorbid problems atten-
tion using a structured diagnostic interview. Table 1 presents dant to disordered eating itself.
a list of these measures in alphabetic order. The principal A search of the literature and other registration platforms
purpose of this meta-analysis is to examine the prevalence of yielded no previous global meta-analytic review of SBDE
disordered eating, as assessed by these measures; hereafter among university students in general. Thus, this meta-anal-
we refer to this concept as screen-based disordered eating, ysis extends previous reviews by our research team of SBDE
using the acronym SBDE. in medical students [31–33] by evaluating the prevalence and
several potential confounders of disordered eating in more
Risk and the transition to college or university general and diverse samples of undergraduate university
students across the world. Specifically, the event rate was
Eating disorders can emerge at any age, ranging from early categorized using pre-defined cut-off scores from validated
childhood to older adulthood [18, 19]. However, given what screening devices, that is, continuous measurements of eat-
is known about risk factors and the modal ages of onset, ing disorder risk such as the EAT-26 and SCOFF (Table 1).
one high-risk period is late adolescence and emerging adult- To examine the sources of the expected heterogeneity in
hood, that is, roughly ages 17 through 22 [20]. Within that disordered eating estimates, confounder analyses, adjusting

13
3218 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

for age and BMI, will be also conducted for country, per- inclusion and exclusion criteria. All duplicate studies were
centage of the sample that is female, culture (Western vs eliminated. Initial data extraction and quality assessment
non- Western), measure, and timeframe/year. Results of were conducted independently by two team members (DS
these meta-analyses should be useful in determining alloca- and OA). Any disputes regarding the suitability of a study
tion of resources in the development and dissemination of for inclusion in the review were resolved by a conversation
prevention programs for undergraduates. with the senior reviewer/expert clinician (ML or HJ), fol-
lowed by consensus of the research team.

Methods Eligibility inclusion and exclusion criteria

This study's protocol was registered on 2021–09-19 at Open This meta-analysis included the full text of original English-
Science Framework (OSF; https://d​ oi.o​ rg/1​ 0.1​ 7605/O
​ SF.I​ O/​ language articles, published before 1 October 2021, related
MB74E), an open-source platform that allows researchers to to SBDE among university (all countries) or college (in the
share their findings with others and get assistance throughout USA) students. The population was defined as undergraduate
their research. To make the review visible and avoid dupli- students from different disciplines and majors.
cation the protocol was also entered into the PROSPERO To cast the net widely, we included studies that met the
International prospective register of systematic reviews following criteria: (1) were published in an English-lan-
(CRD42022303882). guage journal; (2) the entire sample or a distinct subset of
This study was reported using Preferred Reporting Items the sample consisted of university or college (in the U.S.
for Systematic Reviews and Meta-Analyses (PRISMA2020; sense of the term, i.e., not a private high school) undergradu-
[34]. Statistical analyses were conducted and presented ate students; (3) participants completed one of the screening
according the Meta-analysis of Observational Studies in measures (see Table 1) for determining who is at-risk for
Epidemiology (MOOSE) protocol [35]. an eating disorder, such that scores could indicate endorse-
ment of the extremes of attitudes and behaviors that may
Search strategy exist in many cultures [10, 37, 38]; and, given the preceding
criterion, (4) participant responses to the screening measure
During September 2021 two authors (DS and MK) did an (e.g., the EAT-26) were scored and reported such that the
electronic search of the literature using nine databases: percentages of participant falling above and below estab-
PubMed/MEDLINE, American Psychological Association lished cut-off points could be determined.
PsycINFO, ScienceDirect, Springer, EBSCOhost, Embase, The following sources of data were excluded: (1) stud-
Cumulative Index to Nursing and Allied Health Literature ies of students in post-baccalaureate programs (e.g., those
(CINAHL), Scopus, and Web of Science. The full-text pursuing masters or doctoral degrees); (2) investigations of
search was conducted according to the following keywords mental health issues other than the prevalence of SBDE; and
and lists: List A: university student [OR] tertiary student (3) studies for which we were unable to get the necessary
[OR] college student [AND] List B: eating disorder* [OR] data even after contacting the authors. Figure 1 shows the
eating behavior/behaviour* [OR] feeding disorder* [OR] PRISMA 2020 [39] flow diagram for study selection.
eating symptom* [OR] eating attitude* [OR] eating prob-
lem*. The * ensures that the search term covers both the sin- Procedure
gular noun forms, as well as the reverse order of the words
in the phrase. For example, searching for “eating disorder” ASReview, a free online tool that combines digital tech-
encompasses “disordered eating” and “eating disorders”. nologies (e.g., natural language processing) with artificial
To verify that we included all relevant publications, we intelligence and machine learning, was used to screen and
also examined the reference lists of selected articles to iden- code the 89 studies selected for systematic review [40]. The
tify other potentially relevant articles and reviews. Meta- Abstrackr semi-automated abstract screening tool for sys-
analyses that do not include grey literature are more likely tematic reviews was used to increase the precision of abstract
to inflate effect size estimates, and produce less exact effect screening [41]. When necessary, data were extracted from
size estimates than those that do [36]. Consequently, while plot images using WebPlotDigitizer v4.5, an open-source
examining the reference sections we looked for organiza- web-based tool [42].
tional reports, unpublished studies, and studies published To standardize data description the following vari-
outside of widely known journals. ables, in addition to the key result of the event rate of
Three team members (DS, MJ, and SH) then indepen- screen-based disordered eating among university stu-
dently assessed the initial set of articles identified by screen- dents, were independently extracted by three members of
ing the titles, abstracts, and full-text articles according to the the research team (DS, MK, and SH): Author names, year

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3219

Fig. 1  PRISMA 2020 flow diagram for study selection

of publication, country in which the data were collected, Data analysis and data visualization
sample size, mean age (years), sex (percentage of female
participants), mean body mass index (kg/m2), and meas- A classical meta-analysis based on the random-effects model
ure used to determine presence or absence of SBDE. This was used, with the assumption that actual effects will vary
meta-analysis study included samples from 40 countries, over time [45]. We used the general inverse variance method
which were further coded into two categories, Western and with the logit transformed [PLO] proportions [46], and the
non-Western countries, according to regional groups of DerSimonian-Laird method was used to estimate and adjust
member states defined by the United Nations [43]. for the between-study variance in effects [47]. Random-
Consensus among the aforementioned three reviewers effects modelling was used because it assumes that, in using
was used to settle disagreements. If consensus could not different measures (e.g., EAT and SCOFF; Table 1), differ-
be reached, a fourth author (ZS) was involved in resolving ent sets of studies are estimating different, yet conceptually
the issue by discussion. If relevant data were missing from related, effects. For each study the pooled prevalence and the
a publication, the corresponding author of the article was 95% confidence interval are reported.
contacted. A forest plot was used to display data [48]. It is a dis-
advantage that forest plots may display only confidence
intervals at a significance level, such as p < 0.05. Confi-
Assessment of study quality and risk of bias dence intervals should also be used to determine whether a
research effect is substantial and therefore results are repro-
The Newcastle–Ottawa Scale (NOS) was used by two ducible, so drapery plots were also used [49] and analysis of
authors (ZS and HJ) independently to evaluate the quality the p curve was also reported [50]. The drapery plot depicts
of the studies included [44]. The NOS checklist consists the p value function as curves that provide the prediction
of three items: participants selection (sampling), compara- range for a single future study for all individual studies and
bility, and outcome and statistics. The NOS is based on a pooled values in a meta-analysis [49].
rating system [44] in which each item receives 1 to 3 (or 4) To further strengthen the results of the classical meta-
stars. This means that the maximum score for each study is analysis, Bayesian meta-analysis was also conducted and
either nine (cross-sectional and cohort studies) or 10 stars reported. Meta-analysis using Bayesian methods has three
(randomized controlled trials and case–control studies). principal advantages over many classical methods [51].
A study with 8 stars has good quality and low risk of bias, First, they account for the imprecision of the between-study
a study with a score of 5–7 stars has the moderate quality variance estimates [51]. Second, Bayesian methods take
and moderate risk of bias, and a study with a score of 0–4 into account “priors”, that is, what is previously known on
stars has low quality and high risk of bias. the topic [52]. Finally, Bayesian methods include external

13
3220 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

evidence, such as information about the effects of interven- investigate publication bias, funnel plots were employed as
tions or likely differences between studies [51, 52]. a preliminary visual tool [66]. To correct for funnel plot
Our prior distribution focused on eating disorder preva- asymmetry owing to probable publication bias, the trim and
lence in the absence of screen-based estimates and was pro- fill approach [67] was used to generate adjusted point esti-
posed to be µ = 10%, τ = 2%, and η = 5% according to previ- mates. To conduct a more rigorous analysis of publication
ous global estimates [53, 54]. The large eta was postulated bias, Peters’ correlations [68] and Egger's regression [66]
in our review due to the amount of "error" that could be were also utilized as gold standards.
obtained using a screen-based self-reported tool, as indi- Subgroup meta-analyses were used to investigate hetero-
cated by on our comprehensive review of the clinical meas- geneous outcomes and to answer specific queries regarding
ures commonly used (Table 1). distinct populations or study characteristics [69]. Subgroup
Bayesian meta-analysis uses the Bayesian hierarchical analyses were performed on categorical variables includ-
model, which relies on the same basic assumptions under- ing country, culture (Western vs. non-Western), and some
pinning the conventional random-effects model [55]. The of the clinical measures/scales used in various studies. To
difference is that in the Bayesian model the prior distribu- investigate the effect of time as a confounder the studies
tion (informative, weakly informative, or uninformative) is were clustered into 5-year intervals: 1985–1989, 1990–1994,
assumed for µ and τ2. The prior distribution describes the 1995–1999, 2001–2004, 2005–2009, 2010–2014,
uncertainty surrounding a particular effect measure within a 2015–2019, and 2020 onwards. The subgroup meta-analy-
meta-analysis, such as the odds ratio or the mean difference ses addressed any subgroup of five studies or more, and all
[55]. Uncertainty may be attributable to the researchers’ sub- results are reported graphically using forest plots.
jective beliefs about the size of the effect or to sources of Meta-regressions are, in essence, regression models in
evidence excluded from the meta-analysis. Quantity uncer- which the values of one or more explanatory factors are
tainty is reflected by the width of the prior distribution [56]. used to predict the outcome variable [70]. A meta-regres-
It is possible to use a non-informative prior when there is sion analysis' regression coefficient will indicate how the
little or no available information, such that all values are outcome variable changes as the explanatory variable (the
equally likely [56]. A credible interval (CrI) in Bayesian possible moderator/effect modifier/confounding variable)
statistics is a range of values where an unobserved parameter is increased by one unit [70]. A term for the interaction
value is likely to occur [57]. In our analyses we reported the between age, sex and BMI was tested. In statistically sig-
[95% CrI]. nificant meta-regression models effect size was reported
We assessed between-study heterogeneity using the I2 using R2, and percent of variance explained of 1–8%, 9–24%
statistic; a value between 75 and 100% represents a high and ≥ 25% were regarded as small, medium and large effect
degree of heterogeneity [52]. We also evaluated heterogene- size, respectively [71].
ity using Cochran's Q statistics [58], and ­tau2 (τ2) and tau (τ) R software for statistical computing was used to analyze
[52]. The H statistic [59] is the square root of the following: all data [72]. The packages ‘meta’ [73] and ‘metafor’ [74]
Cochran’s χ2 heterogeneity statistic divided by the degree of were used to perform all classical meta-analytics. Package
freedom [52]. To visualize heterogeneity we used a simple ‘bayesmeta’ was used to perform Bayesian random-effects
form of the Galbraith radial plot [60] in which the inverse of meta-analysis [75]. Using the package ‘robvis’, risk-of-bias
standard errors (horizontal axis) is plotted against observed plots were generated for quality assessment [76]. For all
effect sizes or outcomes standardized by their correspond- investigations, a summary plot (weighted) was generated to
ing standard errors (vertical axis). On the right-hand side of show the proportion of information inside each judgment
a full-scale Galbraith plot, an arc shows the corresponding for each domain [76]. Summary of all studies' risk of bias
effect sizes or outcomes [61, 62]. assessments. The risk of bias in each domain, as well as the
Meta-analysis' validity and robustness may be compro- overall risk, is depicted by a traffic light plot.
mised by the inclusion of outliers. Whenever the study's con-
fidence interval does not align with the pooled effect's confi-
dence interval, the study is classified as an outlier and can be Results
addressed by the sensitivity analysis [63]. Therefore, using a
Jackknife sensitivity analysis, we eliminated one study at a Descriptive
time to make sure we did not have any inordinate influence
from any single study [64]. This analysis involves repeating The initial literature search, conducted during Septem-
the main meta-analysis as many times as the number of stud- ber to October 2021, yielded 1523 studies, of which 89
ies included, discarding one different study each time [64]. independent studies [7, 30, 77–163] (across all times and
A publication bias occurs when the odds of research measures, K = 105 data points for analyses; N of partici-
being published are influenced by its findings [65]. To pants = 149,629) met the inclusion and exclusion criteria.

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3221

Table 2  Selected descriptive results of the studies included in this systematic review and meta-analysis of disordered eating among university
students
S.No. Study label Citation Country Study characteristics Sample characteristics Quality score

1 Abdul Manaf (2016) [77] Malaysia Cross-sectional design. Sample Size %Female = 100%, 7
N = 206. ED Measure: EAT-26 Age = 19.5 years,
BMI = 22.2 kg/m2
2 Abo Ali (2020) [78] Egypt Cross-sectional design. Sample Size %Female = 67.2%, 8
N = 615. ED Measure: EAT-26 Age = 21 years,
BMI = 22 kg/m2
3 Akdevelioglu (2010) [79] Turkey Cross-sectional design. Sample Size %Female = 70%, 5
N = 577. ED Measure: EAT-40 Age = 21.2 years,
BMI = 21.2 kg/m2
4 Al Banna (2021) [80] Bangladesh Cross-sectional design. Sample Size %Female = 49.6%, 8
N = 365. ED Measure: EAT-26 Age = 21.1 years,
BMI = 22.2 kg/m2
5 Albrahim (2019) [81] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 6
N = 396. ED Measure: EAT-26 Age = 20.1 years,
BMI = 23.2 kg/m2
6 Alcaraz-Ibáñez (2019) [82] Spain Cross-sectional design. Sample Size %Female = 46%, 7
N = 545. ED Measure: SCOFF Age = 21.4 years,
BMI = 23 kg/m2
7 Alhazmi (2019) [83] Saudi Arabia Cross-sectional design. Sample Size %Female = 50%, 7
N = 342. ED Measure: EAT-26 Age = 21.2 years,
BMI = 22.2 kg/m2
8 Alkazemi (2018) [84] Kuwait Cross-sectional design. Sample Size %Female = 100%, 7
N = 1147. ED Measure: EAT-26 Age = 20.5 years,
BMI = 23.9 kg/m2
9 AlShebali (2020) [85] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 7
N = 503. ED Measure: EDE-Q Age = 19.8 years,
BMI = 23.4 kg/m2
10 Alwosaifer (2016) [86] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 7
N = 656. ED Measure: EAT-26 Age = 18.7 years,
BMI = 22.2 kg/m2
11 Azzouzi (2019) [87] Morocco Cross-sectional design. Sample Size %Female = 65.1%, 7
N = 710. ED Measure: SCOFF Age = 21.3 years,
BMI = 22.9 kg/m2
12 Badrasawi (2019) [88] Palestine Cross-sectional design. Sample Size %Female = 100%, 7
N = 154. ED Measure: BEDS-7 Age = 19.6 years,
BMI = 22.2 kg/m2
13 Barayan (2018) [89] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 5
N = 319. ED Measure: EDE-Q Age = 21.2 years,
BMI = 22 kg/m2
14 Barry (2021) [90] United States Cross-sectional design. Sample Size %Female = 50.4%, 8
N = 804. ED Measure: SCOFF Age = 21.2 years,
BMI = 22.2 kg/m2
15 Benítez (2019) [91] Spain Cross-sectional design. Sample Size %Female = 59.5%, 7
N = 600. ED Measure: EDI Age = 20.8 years,
BMI = 22.2 kg/m2
16 Bizri (2020) [92] Lebanon Cross-sectional design. Sample Size %Female = 53.4%, 7
N = 131. ED Measures: EAT-26; SCOFF Age = 23 years,
BMI = 22.2 kg/m2
17 Bo (2014) [93] Italy Cross-sectional design. Sample Size %Female = 54%, 7
N = 440. ED Measures: EAT-26; SCOFF Age = 19.8 years,
BMI = 16.9 kg/m2
18 Bosi (2016) [94] Brazil Cross-sectional design. Sample Size %Female = 100%, 7
N = 202. ED Measure: EAT-26 Age = 21.8 years,
BMI = 22.2 kg/m2

13
3222 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Table 2  (continued)
S.No. Study label Citation Country Study characteristics Sample characteristics Quality score

19 Brumboiu (2018) [95] Romania Cross-sectional design. Sample Size %Female = 82%, 7
N = 222. ED Measure: SCOFF Age = 21.5 years,
BMI = 21.3 kg/m2
20 Carriedo (2020) [96] Mexico Cross-sectional design. Sample Size %Female = 65.4%, 7
N = 911. ED Measure: EDE-Q Age = 21 years,
BMI = 22.6 kg/m2
21 Castejón (2020) [97] Spain Cross-sectional design. Sample Size %Female = 65.9%, 6
N = 604. ED Measure: EDI Age = 22.5 years,
BMI = 22.2 kg/m2
22 Chammas (2017) [98] Lebanon Cross-sectional design. Sample Size %Female = 37%, 6
N = 457. ED Measure: SCOFF Age = 21.3 years,
BMI = 22.2 kg/m2
23 Chan (2020) [99] Malaysia Cross-sectional design. Sample Size %Female = 51%, 8
N = 1017. ED Measure: EAT-26 Age = 20.7 years,
BMI = 22 kg/m2
24 Chaudhari (2017) [100] India Cross-sectional design. Sample Size %Female = 60.6%, 7
N = 193. ED Measure: EDE-Q Age = 23.4 years,
BMI = 24.5 kg/m2
25 Christensen (2021) [101] United States Cohort design. Sample Size N = 579. ED %Female = 76.3%, 7
Measure: EDDS Age = 21.8 years,
BMI = 25.1 kg/m2
26 Compte (2015) [102] Argentina Cross-sectional design. Sample Size %Female = 0%, 7
N = 472. ED Measure: EAT-26 Age = 21.2 years,
BMI = 24.8 kg/m2
27 Damiri (2021) [103] Palestine Cross-sectional design. Sample Size %Female = 61.3%, 8
N = 1047. ED Measures: EAT-26; SCOFF Age = 20.2 years,
BMI = 23.3 kg/m2
28 Din (2019) [104] Pakistan Cross-sectional design. Sample Size %Female = 56%, 7
N = 672. ED Measure: EAT-26 Age = 21.7 years,
BMI = 22.1 kg/m2
29 Ebrahim (2019) [105] Kuwait Cross-sectional design. Sample Size %Female = 0%, 7
N = 400. ED Measure: EAT-26 Age = 21.9 years,
BMI = 25.8 kg/m2
30 Erol (2019) [106] Turkey Cross-sectional design. Sample Size %Female = 70%, 7
N = 298. ED Measure: EAT-40 Age = 21.3 years,
BMI = 22.2 kg/m2
31 Falvey (2021) [107] Multi Cross-sectional design. Sample Size %Female = 65.9%, 7
N = 77,193. ED Measure: SCOFF Age = 23.1 years,
BMI = 24.4 kg/m2
32 Farchakh (2019) [30] Lebanon Cross-sectional design. Sample Size %Female = 50.4%, 8
N = 627. ED Measures: ORTO-15; EAT-26 Age = 21.8 years,
BMI = 23.4 kg/m2
33 Fatima (2018) [108] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 8
N = 120. ED Measure: EAT-26 Age = 21.2 years,
BMI = 22.2 kg/m2
34 Gramaglia (2019) [109] Multi Cross-sectional design. Sample Size %Female = 70%, 7
N = 664. ED Measures: EAT-26; ORTO-15 Age = 24 years,
BMI = 22.2 kg/m2
35 Greenleaf (2009) [110] United States Cross-sectional design. Sample Size %Female = 100%, 7
N = 204. ED Measure: QEDD Age = 20.2 years,
BMI = 23.1 kg/m2
36 Havemann (2011) [111] South Africa Cross-sectional design. Sample Size N = 26. %Female = 100%, 4
ED Measure: EAT-26 Age = 19 years,
BMI = 23.2 kg/m2
37 Herzog (1985) [112] United States Cross-sectional design. Sample Size %Female = 100%, 5
N = 121. ED Measure: SD Age = 25.1 years,
BMI = 22 kg/m2

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3223

Table 2  (continued)
S.No. Study label Citation Country Study characteristics Sample characteristics Quality score

38 Iyer (2021) [113] India Cross-sectional design. Sample Size %Female = 56.3%, 7
N = 332. ED Measure: EAT-26 Age = 22.3 years,
BMI = 22 kg/m2
39 Jamali (2020) [114] Pakistan Cross-sectional design. Sample Size %Female = 36.9%, 7
N = 407. ED Measures: EAT-26; SCOFF Age = 19.9 years,
BMI = 20.8 kg/m2
40 Jennings (2006) [115] Australia Cross-sectional design. Sample Size %Female = 100%, 4
N = 240. ED Measure: EAT-26 Age = 19.3 years,
BMI = 21.2 kg/m2
41 Joja (2012) [116] Germany Case–control design. Sample Size N = 110. %Female = 100%, 8
ED Measure: EDI Age = 20.3 years,
BMI = 21.5 kg/m2
42 Kiss-Toth (2018) [117] Multi Cross-sectional design. Sample Size %Female = 70%, 6
N = 1965. ED Measure: SCOFF Age = 21.2 years,
BMI = 22.2 kg/m2
43 Ko (2015) [118] Vietnam Cross-sectional design. Sample Size %Female = 100%, 7
N = 203. ED Measure: SCOFF Age = 18.8 years,
BMI = 19 kg/m2
44 Koushiou (2019) [119] Greece Cross-sectional design. Sample Size %Female = 90%, 7
N = 334. ED Measure: WCS; EDDS Age = 20.7 years,
BMI = 22.2 kg/m2
45 Kutlu (2013) [120] Turkey Cross-sectional design. Sample Size %Female = 59.5%, 7
N = 262. ED Measure: EAT-40 Age = 21.7 years,
BMI = 21.5 kg/m2
46 Ladner (2019) [121] France Cross-sectional design. Sample Size %Female = 69%, 7
N = 3076. ED Measure: SCOFF Age = 21.2 years,
BMI = 22.2 kg/m2
47 Le Grange (1998) [122] South Africa Cross-sectional design. Sample Size %Female = 75%, 6
N = 1402. ED Measure: EAT-40 Age = 19.2 years,
BMI = 22 kg/m2
48 Lee (2015) [123] Korea Cross-sectional design. Sample Size %Female = 52.3%, 7
N = 199. ED Measure: DEBQ Age = 29.2 years,
BMI = 22 kg/m2
49 Liao (2013) [124] China Cohort design (two data points). Sample %Female = 63%, 7
Size N = 487. ED Measure: EAT-26 Age = 20.5 years,
BMI = 20.2 kg/m2
50 Mancilla-Diaz (2007) [125] Mexico Cross-sectional design. Sample Size %Female = 100%, 7
N = 1402. ED Measure: EAT-40 Age = 19.2 years,
BMI = 22.4 kg/m2
51 Marciano (1988) [126] Canada Cross-sectional design. Sample Size %Female = 84.5%, 6
N = 994. ED Measure: EAT-26 Age = 20.4 years,
BMI = 22.2 kg/m2
52 Mazzaia (2018) [127] Brazil Cross-sectional design. Sample Size %Female = 84.2%, 7
N = 120. ED Measure: EAT-26 Age = 21.9 years,
BMI = 23.3 kg/m2
53 Mealha (2013) [7] Portugal Cross-sectional design. Sample Size %Female = 100%, 6
N = 189. ED Measures: EAT-26; EDI Age = 20.3 years,
BMI = 21.2 kg/m2
54 Momeni (2020) [128] Iran Cross-sectional design. Sample Size %Female = 47%, 7
N = 385. ED Measure: EAT-26 Age = 21.8 years,
BMI = 22.4 kg/m2
55 Ngan (2017) [129] Malaysia Cross-sectional design. Sample Size %Female = 65%, 5
N = 263. ED Measure: EAT-26 Age = 22.8 years,
BMI = 22 kg/m2
56 Nichols (2009) [130] West Indies Cross-sectional design. Sample Size %Female = 48%, 6
N = 383. ED Measure: EAT-26 Age = 21.2 years,
BMI = 22.2 kg/m2

13
3224 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Table 2  (continued)
S.No. Study label Citation Country Study characteristics Sample characteristics Quality score

57 Padmanabhan (2017) [131] United Arab Cross-sectional design. Sample Size %Female = 52.6%, 5
Emirates N = 156. ED Measure: EAT-26 Age = 23.3 years,
BMI = 22.2 kg/m2
58 Parra-Fernández (2019) [132] Spain Cross-sectional design. Sample Size %Female = 70%, 7
N = 492. ED Measure: EDI Age = 20 years,
BMI = 22.6 kg/m2
59 Parreño-Madrigal (2020) [133] Spain Cross-sectional design. Sample Size %Female = 72.6%, 8
N = 481. ED Measure: SCOFF Age = 20.1 years,
BMI = 22.4 kg/m2
60 Pereira (2011) [134] Brazil Cross-sectional design. Sample Size %Female = 100%, 7
N = 214. ED Measure: EAT-26 Age = 21 years,
BMI = 21.1 kg/m2
61 Pitanupong (2017) [135] Thailand Cross-sectional design. Sample Size %Female = 56%, 7
N = 885. ED Measure: EAT-26 Age = 20.8 years,
BMI = 21.2 kg/m2
62 Plichta (2019) [136] Poland Cross-sectional design. Sample Size %Female = 70.4%, 7
N = 1120. ED Measure: ORTO-15 Age = 21.4 years,
BMI = 22 kg/m2
63 Polanco (2020) [137] Mexico Cross-sectional design. Sample Size N = 90. %Female = 66.4%, 6
ED Measure: EAT-26 Age = 20 years,
BMI = 22 kg/m2
64 Radwan (2018) [138] United Arab Cross-sectional design. Sample Size %Female = 61.4%, 7
Emirates N = 662. ED Measure: EAT-26 Age = 20.4 years,
BMI = 24.1 kg/m2
65 Ramaiah (2015) [139] India Cross-sectional design. Sample Size %Female = 65%, 7
N = 172. ED Measure: EAT-26 Age = 21 years,
BMI = 21.6 kg/m2
66 Rasman (2018) [140] Malaysia Cross-sectional design. Sample Size %Female = 75.3%, 8
N = 279. ED Measure: SCOFF Age = 21.9 years,
BMI = 22.5 kg/m2
67 Rathner (1994) [141] Austria Cross-sectional design. Sample Size %Female = 40.9%, 7
N = 379. ED Measures: EDI; ANIS Age = 22 years,
BMI = 21 kg/m2
68 Reyes-Rodríguez (2011) [142] Puerto Rico Cross-sectional design. Sample Size %Female = 0%, 5
N = 709. ED Measure: EAT-26 Age = 18.3 years,
BMI = 24.4 kg/m2
69 Roshandel (2012) [143] Iran Cross-sectional design. Sample Size %Female = 100%, 7
N = 400. ED Measure: EAT-26 Age = 22.1 years,
BMI = 21.2 kg/m2
70 Rostad (2021) [144] Norway Cross-sectional design. Sample Size %Female = 70.9%, 8
N = 1044. ED Measure: EDS Age = 21.2 years,
BMI = 22.8 kg/m2
71 Safer (2020) [145] Tunisia Cross-sectional design. Sample Size %Female = 69.9%, 7
N = 974. ED Measure: SCOFF Age = 22.8 years,
BMI = 22.2 kg/m2
72 Saleh (2018) [146] Palestine Cross-sectional design. Sample Size %Female = 100%, 7
N = 2001. ED Measures: EAT-26; SCOFF Age = 19.5 years,
BMI = 21.7 kg/m2
73 Sepúlveda (2007) [147] Spain Cross-sectional design. Sample Size %Female = 67.9%, 8
N = 2551. ED Measure: EDI Age = 21 years,
BMI = 22 kg/m2
74 Sharifian (2021) [148] Finland Cross-sectional design. Sample Size %Female = 52.6%, 7
N = 3110. ED Measure: SCOFF Age = 21.2 years,
BMI = 22.2 kg/m2
75 Sharma (2019) [149] India Cross-sectional design. Sample Size %Female = 42.4%, 8
N = 370. ED Measure: EAT-26 Age = 20.3 years,
BMI = 22 kg/m2

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3225

Table 2  (continued)
S.No. Study label Citation Country Study characteristics Sample characteristics Quality score

76 Shashank (2016) [150] India Cross-sectional design. Sample Size %Female = 100%, 8
N = 134. ED Measures: EAT-26; SCOFF Age = 21.4 years,
BMI = 22.4 kg/m2
77 Spillebout (2019) [151] France Cross-sectional design. Sample Size %Female = 69.9%, 7
N = 731. ED Measure: SCOFF Age = 20 years,
BMI = 22.1 kg/m2
78 Taha (2018) [152] Saudi Arabia Cross-sectional design. Sample Size %Female = 100%, 7
N = 1200. ED Measure: EAT-26, SCOFF Age = 21 years,
BMI = 22.2 kg/m2
79 Tavolacci (2015) [153] France Cross-sectional design. Sample Size %Female = 63.6%, 7
N = 3457. ED Measure: SCOFF Age = 20.5 years,
BMI = 21.4 kg/m2
80 Tavolacci (2018) [154] France Cross-sectional design. Sample Size %Female = 61%, 7
N = 1225. ED Measure: SCOFF Age = 21.6 years,
BMI = 22 kg/m2
81 Tavolacci (2020) [155] France Cross-sectional design. Sample Size %Female = 63.4%, 7
N = 1493. ED Measure: SCOFF Age = 20.1 years,
BMI = 22.2 kg/m2
82 Thangaraju (2020) [156] India Cross-sectional design. Sample Size %Female = 100%, 7
N = 199. ED Measure: EDE-Q Age = 20.4 years,
BMI = 23.8 kg/m2
83 Tury (2020) [157] Hungary Cohort design (two data points). Sample %Female = 53.9%, 7
Size N = 538. ED Measures: ANIS; EDI Age = 21.4 years,
BMI = 21.4 kg/m2
84 Uriegas (2021) [158] United States Cross-sectional design. Sample Size %Female = 56%, 7
N = 150. ED Measure: EDI Age = 19.9 years,
BMI = 25.2 kg/m2
85 Uzun (2006) [159] Turkey Cross-sectional design. Sample Size %Female = 100%, 6
N = 414. ED Measure: EAT-40 Age = 19.9 years,
BMI = 22.2 kg/m2
86 Weigel (2016) [160] Germany Cross-sectional design. Sample Size %Female = 58.2%, 7
N = 304. ED Measure: EDI Age = 22.6 years,
BMI = 20.1 kg/m2
87 Yoneda (2020) [161] Japan Cross-sectional design. Sample Size %Female = 100%, 7
N = 469. ED Measure: EAT-26 Age = 19.9 years,
BMI = 20.7 kg/m2
88 Yu (2015) [162] China Cross-sectional design. Sample Size %Female = 64.2%, 6
N = 1328. ED Measure: EAT-26 Age = 21.2 years,
BMI = 22.2 kg/m2
89 Zhou (2020) [163] United States RCT design. Sample Size N = 130. ED %Female = 100%, 7
Measure: EDE-Q Age = 20.8 years,
BMI = 24.4 kg/m2

FEDS feeding and eating disorders. Quality score was computed based on Newcastle–Ottawa quality assessment scale total score for cross-
sectional studies
EAT-26 Eating Attitudes Test-26, EAT-40 Eating Attitudes Test-40, SCOFF Sick, Control, One Stone, Fat, Food, EDE-Q Eating Disorder Exam-
ination- Questionnaire, BEDS-7 Binge Eating Disorder Screener-7, ORTO-15 ORTO-15, QEDD Questionnaire for Eating Disorder Diagnoses,
EDDS The Eating Disorder Diagnostic Scale, SD Self-developed, WCS The Weight Concern Scale, DEBQ Dutch Eating Behavior Question-
naire, EDI Eating Disorder Inventory-I/II, ORTO-11 ORTO-11, ANIS Anorexia Nervosa Inventory for Self-Rating

Details of the studies included are shown in Table 2. Of (2%), and 11% presented data collected during the COVID-
the 89 studies only two (2.2%) were grey literature: [117] 19 pandemic. Furthermore, the Eating Attitudes Test-26
and [121]. They were of a similar quality compared to the (EAT-26) and Sick, Control, One Stone, Fat, Food (SCOFF)
published studies. measures were the most common scales, making up 64%
Most of the studies analyzed were cross-sectional (95%), of total studies (see Table 3). The mean percentage of
although a few used cohorts (3%) or other methodology participants self-identifying as female was approximately

13
 3226

Table 3  A meta-analysis of disordered eating among university students

13
Analysis K N Random effects model Heterogeneity Confounders Publication bias
2a 2b c
Pooled results [95% CI or CrI] Figure I τ τ H Q Cochran's Age Sex BMI Egger's ­teste Peter's test
Q P ­valued

Prevalence of studies 105 145,629 19.7% [17.9%; 21.6%] Figure 4 98.2 0.6 0.34 7.39 5696.85 0.001 0.49 0.04 0.001 0.90 0.06
Bayesian analysis 105 145,629 Odds 0.24 [0.20; 30] Figure 5 98% 0.9 – – – – – – – –
Prevalence by country
Saudi Arabia 8 4736 21.2% [14.1%; 30.5%] Figure 12 97.7% 0.70 0.48 – 307.42 0.001 – – – NS NS
India 7 1534 18.1% [14.7%; 22.0%] 70.1% 0.27 0.076 – 20.05 – – – NS NS
United states of America 6 1988 37.1% [26.3%; 49.5%] 95.8% 0.61 0.37 – 117.83 – – – NS NS
Spain 6 5235 31.7% [20.4%; 45.6%] 98.8% 0.73 0.53 – 404.30 – – – NS NS
Palestine 5 6250 32.8% [26.2%; 40.2%] 96.8% 0.35 0.13 – 124.57 – – – NS NS
Lebanon 5 1966 33.2% [15.9%; 56.7%] 98.8% 1.09 1.19 – 338.83 – – – NS NS
France 5 9982 21.0% [18.7%; 23.6%] 88.4% 0.16 0.025 – 34.52 – – – NS NS
Prevalence by culture (Western)
No 55 29,363 20.9% [17.8%; 24.4%] Figure 13 97.9% 0.74 0.55 – 2711.00 0.001 – – – NS NS
Yes 50 115,966 18.4% [16.4%; 20.6%] 97.8% 0.51 0.26 – 2264.44 – – – NS NS
Prevalence by measure
EAT-26 45 23,821 17.0% [13.9%; 20.3%] Figure 14 97.6% 0.75 0.56 – 1905.43 0.001 – – – NS NS
SCOFF 22 100,638 27.6% [24.1%; 31.5%] 98.4% 0.44 0.19 – 1413.76 – – – NS NS
EDI 10 6394 16.9% [9.6%; 28.2%] 98.8% 1.04 1.08 – 729.14 – – – NS NS
EAT-40 6 4355 10.6% [7.4%; 14.9%] 93.3% 0.45 0.21 – 75.17 – – – NS NS
EDE-Q 6 2255 18.1% [8.3%; 35.0%] 97.8% 1.09 1.20 – 223.88 – – – NS NS
Prevalence by Timeframe/Year
2020 Onwards 31 97,625 20.8% [17.6%; 24.5%] Figure 15 98. 4% 0.58 0.34 – 1869.51 0.001 – – – NS NS
2015–2019 50 35,006 23.8% [20.7%; 27.2%] 97.9% 0.63 0.39 – 2376.86 – – – NS NS
2010–2014 11 3256 13.0% [8.4%; 19.7%] 94.6% 0.77 0.60 – 222.67 – – – NS NS
2005–2009 8 6167 10.6% [7.3%; 15.1%] 95.7% 0.56 0.31 – 164.13 – – – NS NS

K Represents the number of included studies, N Represents the number of included samples
a 2
I statistic referred to the percentage of variation across samples due to heterogeneity rather than chance
b 2
τ Describe the extent of variation among the effects observed in different samples (between-sample variance)
c
H Describes confidence intervals of heterogeneity
d
Significant differences between samples in meta-analysis
e
Detects publication bias in meta-analyses
f
Represents the correlation between effect sizes and sample variation
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3227

70% [95% CI = 66–75%], while, as expected, the average τ = 92%. An odd estimate of 0.24 equals an SBSE prevalence
respondent's age was 21 years old [95% CI = 20–22; range of to approximately 1: 4 or more simply 24–25%.
18–29 years), with a median sample mean BMI of 22 kg/m2 There is no publication bias in our data, as evidenced by
[95% CI = 21–24]; range 17–26 kg/m2). visual examination of the funnel (Fig. 6) and Galbraith radial
The studies represented samples from 40 countries and plots (Fig. 7), as well as Egger's regression test at 0.90 and
territories, and the majority (54%) of studies reported data Peter's test at 0.05. The Jackknife sensitivity analysis showed
from non-Western countries. The countries and territories that excluding one study at a time from this meta-analysis
were: Argentina, Australia, Austria, Bangladesh, Brazil, did not affect the prevalence of SBDE in university students
Canada, China, Egypt, Finland, France, Germany, Greece, by more than 0.5% (Fig. 8), suggesting that our weighted
Hungary, India, Iran, Italy, Japan, Kuwait, Lebanon, Malay- prevalence findings are robust and relatively insensitive to
sia, Mexico, Morocco, Norway, Pakistan, Palestine, Poland, outliers. Another indication that the results from all studies
Portugal, Puerto Rico, Republic of Korea, Romania, Saudi are reproducible is seen in a drapery plot based on p values
Arabia, South Africa, Spain, Thailand, Tunisia, Turkey, (Fig. 9), which eliminates the need to rely on the p < 0.05
United Arab Emirates, United States, Vietnam, and West significance threshold when interpreting the results of any
Indies. given study.
Seven countries accounted for 42% of the studies qualify-
ing for this meta-analysis: Saudi Arabia (8%), United States Confounder analyses
of America (7%), Spain (7%), India (7%), France (5%),
Malaysia (4%), and Turkey (4%). Age, BMI, and sex

Meta-regression analysis (Figs. 10 and 11) showed that BMI

Quality assessment of included studies
and sex are statistical confounders, p = 0.001 and p = 0.04,
respectively, for the prevalence of SBDE in university stu-
Figure 2 presents summary plots for the quality and risk
dents. As the mean BMI of the sample increased, or as the
of bias of the studies qualifying for this meta-analysis. The
percentage of the sample that was female increased, so did
overall risk of bias in this sample of studies was moderate, as
the weighted prevalence of SDBE. The effect size was large
the categorization of bias was low (17%), moderate (81%),
for BMI (R2 =  ~ 0.50), but small for the proportion of the
and high (2%). Figure 3 provides a summary of all studies'
sample that was female (R2 =  ~ 20). There was no statisti-
risk of bias assessments the risk of bias in each domain, as
cally significant relationship between mean age of the sam-
well as the overall risk, is depicted by a traffic light plot.
ple and level of disordered eating (p = 0.49; see Table 3).
The study concerned university college students; thus, it is
Meta‑analysis of the overall prevalence highly unlikely that age could have any impact on the esti-
of disordered eating mates because of the small age range.
The interaction term between age, sex and BMI yielded
Figure 4 presents the raw prevalence data for each study a statistically significant result, p = 0.01, but the interaction
and the results of the basic meta-analysis. The prevalence was not explored further because the effect size was negli-
of SBDE among university students (k = 108, N = 146,210) gible, R2 = 0.10.
using random-effects meta-analysis was (K = 105,
N = 145,629) was [95% CI] = 19.7% [17.9%; 21.6%], Country and culture
I2 = 98.2%, τ [95% CI] = 0.6 [0.64; 0.94], τ2 [95% CI] = 0.34
[0.41; 0.88], H [95% CI] = 7.39 [7.1; 7.7], Cochran's Q Figure 12 shows the weighted prevalence levels as a function
p-value = 0.001. Bayesian meta-analysis, shown in Fig. 5, of country in which the data were collected. These varied
yielded disordered eating odds of 0.24 95% CrI [0.20, 0.30], tremendously, and, as noted above, the number of studies (k)

Fig. 2  Summary plot of the

assessment of the risk of bias

13
3228 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Events per 100

Study Number Total Prevalence (%) 95%CI observations

Abdul Manaf , 2016 13 206 6.3 [ 3.4; 10.5]

Abo Ali, 2020 203 615 33.0 [29.3; 36.9]
Akdevelioglu, 2010 34 577 5.9 [ 4.1; 8.1]
Al Banna, 2021 84 365 23.0 [18.8; 27.7]
Albrahim, 2019 145 396 36.6 [31.9; 41.6]
Alcaraz-Ibáñez, 2019 223 545 40.9 [36.8; 45.2]
Alhazmi, 2019 98 342 28.7 [23.9; 33.8]
Alkazemi, 2018 532 1147 46.4 [43.5; 49.3]
AlShebali, 2020 35 503 7.0 [ 4.9; 9.5]
Alwosaifer, 2016 179 656 27.3 [23.9; 30.9]
Azzouzi, 2019 233 710 32.8 [29.4; 36.4]
Badrasawi, 2019 77 154 50.0 [41.8; 58.2]
Barayan, 2018 45 319 14.1 [10.5; 18.4]
Barry, 2021 233 804 29.0 [25.9; 32.3]
Benítez, 2019 231 600 38.5 [34.6; 42.5]
Bizri, 2020 Q1 22 131 16.8 [10.8; 24.3]
Bizri, 2020 Q2 24 124 19.4 [12.8; 27.4]
Bosi, 2016 20 202 9.9 [ 6.2; 14.9]
Brumboiu, 2018 57 222 25.7 [20.1; 31.9]
Carriedo, 2020 109 911 12.0 [ 9.9; 14.3]
Castejón, 2020 335 604 55.5 [51.4; 59.5]
Chammas, 2017 146 457 31.9 [27.7; 36.4]
Chan, 2020 141 1017 13.9 [11.8; 16.1]
Chaudhari, 2017 27 193 14.0 [ 9.4; 19.7]
Compte, 2015 18 472 3.8 [ 2.3; 6.0]
Damiri, 2021 Q1 329 1047 31.4 [28.6; 34.3]
Damiri, 2021 Q2 221 1047 21.1 [18.7; 23.7]
Din, 2019 103 672 15.3 [12.7; 18.3]
Ebrahim, 2019 185 400 46.2 [41.3; 51.3]
Erol, 2019 33 298 11.1 [ 7.7; 15.2]
Farchakh, 2019 Q1 467 627 74.5 [70.9; 77.9]
Farchakh, 2019 Q2 189 627 30.1 [26.6; 33.9]
Fatima, 2018 32 120 26.7 [19.0; 35.5]
Gramaglia, 2019 Q1 56 664 8.4 [ 6.4; 10.8]
Gramaglia, 2019 Q2 246 664 37.0 [33.4; 40.8]
Grange, 1998 153 1402 10.9 [ 9.3; 12.7]
Greenleaf, 2009 56 204 27.5 [21.5; 34.1]
Havemann, 2011 14 26 53.8 [33.4; 73.4]
Herzog, 1985 18 121 14.9 [ 9.1; 22.5]
Iyer, 2021 50 332 15.1 [11.4; 19.4]
Jamali, 2020 Q1 146 407 35.9 [31.2; 40.7]
Jamali, 2020 Q2 199 407 48.9 [43.9; 53.9]
Jennings, 2006 26 240 10.8 [ 7.2; 15.5]
Joja, 2012 17 110 15.5 [ 9.3; 23.6]
Kara, 2021 228 579 39.4 [35.4; 43.5]
Kiss-Toth, 2018 519 1965 26.4 [24.5; 28.4]
Ko, 2015 99 203 48.8 [41.7; 55.9]
Koushiou, 2019 Q1 102 334 30.5 [25.6; 35.8]
Koushiou, 2019 Q2 60 340 17.6 [13.7; 22.1]
Kutlu, 2013 13 262 5.0 [ 2.7; 8.3]
Ladner, 2019 732 3076 23.8 [22.3; 25.3]
Lee, 2015 31 199 15.6 [10.8; 21.4]
Liao, 2006 T1 15 487 3.1 [ 1.7; 5.0]
Liao, 2008 T2 17 486 3.5 [ 2.1; 5.5]
Mancilla-Diaz, 2007 246 1402 17.5 [15.6; 19.6]
Marciano, 1988 147 994 14.8 [12.6; 17.1]
Mazzaia, 2018 30 120 25.0 [17.5; 33.7]
Mealha, 2013 Q1 8 189 4.2 [ 1.8; 8.2]
Mealha, 2013 Q2 31 189 16.4 [11.4; 22.5]
Momeni, 2020 73 385 19.0 [15.2; 23.2]
Nancy, 2021 68 150 45.3 [37.2; 53.7]
Ngan, 2017 29 263 11.0 [ 7.5; 15.5]
Nichols, 2009 16 383 4.2 [ 2.4; 6.7]
Padmanabhan, 2017 73 156 46.8 [38.8; 54.9]
Parra-Fernández, 2018 76 454 16.7 [13.4; 20.5]
Parreño-Madrigal, 2020 136 481 28.3 [24.3; 32.5]
Pereira, 2011 48 214 22.4 [17.0; 28.6]
Pitanupong, 2017 141 885 15.9 [13.6; 18.5]
Plichta, 2019 317 1120 28.3 [25.7; 31.0]
Polanco, 2020 7 90 7.8 [ 3.2; 15.4]
Radwan, 2018 220 662 33.2 [29.7; 37.0]
Ramaiah, 2015 29 172 16.9 [11.6; 23.3]
Rasman, 2018 119 279 42.7 [36.8; 48.7]
Rathner, 1994 Q1 18 379 4.7 [ 2.8; 7.4]
Rathner, 1994 Q2 38 379 10.0 [ 7.2; 13.5]
Reyes-Rodríguez, 2011 36 709 5.1 [ 3.6; 7.0]
Roshandel, 2012 86 400 21.5 [17.6; 25.9]
Rostad, 2021 191 1044 18.3 [16.0; 20.8]
Safer, 2020 341 974 35.0 [32.0; 38.1]
Saleh, 2018 Q1 573 2001 28.6 [26.7; 30.7]
Saleh, 2018 Q2 767 2001 38.3 [36.2; 40.5]
Sarah, 2021 15671 77193 20.3 [20.0; 20.6]
Sepulveda, 2007 480 2551 18.8 [17.3; 20.4]
Sharifian, 2021 286 3110 9.2 [ 8.2; 10.3]
Sharma, 2019 78 370 21.1 [17.0; 25.6]
Shashank, 2016 Q1 39 134 29.1 [21.6; 37.6]
Shashank, 2016 Q2 26 134 19.4 [13.1; 27.1]
Simona Bo, 2014 Q1 40 440 9.1 [ 6.6; 12.2]
Simona Bo, 2014 Q2 114 440 25.9 [21.9; 30.3]
Spillebout, 2019 122 731 16.7 [14.1; 19.6]
Taha , 2018 Q1 424 1200 35.3 [32.6; 38.1]
Taha, 2018 Q2 131 1200 10.9 [ 9.2; 12.8]
Tavolacci, 2015 717 3457 20.7 [19.4; 22.1]
Tavolacci, 2018 231 1225 18.9 [16.7; 21.2]
Tavolacci, 2020 370 1493 24.8 [22.6; 27.1]
Thangaraju, 2020 27 199 13.6 [ 9.1; 19.1]
Tury, 2020 Q1T1 42 538 7.8 [ 5.7; 10.4]
Tury, 2020 Q1T2 134 969 13.8 [11.7; 16.2]
Tury, 2020 Q2T1 24 538 4.5 [ 2.9; 6.6]
Tury, 2020 Q2T2 68 969 7.0 [ 5.5; 8.8]
Uzun, 2006 71 414 17.1 [13.6; 21.1]
Weigel, 2016 32 304 10.5 [ 7.3; 14.5]
Yoneda, 2020 41 469 8.7 [ 6.3; 11.7]
Yu, 2015 60 1328 4.5 [ 3.5; 5.8]
Zhou, 2020 96 130 73.8 [65.4; 81.2]

Random effects model 145629 19.7 [17.9; 21.6]

Heterogeneity: I 2 = 98.17%,  2 = 0.34, p = 0
0 20 40 60 80 100
Prevalence (%)

Fig. 3  Traffic light plot of the assessment of the risk of bias Fig. 4  Classical random-effects meta-analysis of disordered eating in
university students

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3229

Fig. 5  Bayesian meta-analysis quoted estimate shrinkage estimate

of disordered eating in univer- study

Abdul Manaf , 2016
estimate
0.0674
95% CI
[0.0384, 0.1181]
sity students Abo Ali, 2020 0.4927 [0.4165, 0.5829]
Akdevelioglu, 2010 0.0626 [0.0443, 0.0885]
Al Banna, 2021 0.2989 [0.2343, 0.3814]
Albrahim, 2019 0.5777 [0.4709, 0.7087]
Alcaraz-Ibáñez, 2019 0.6925 [0.5838, 0.8215]
Alhazmi, 2019 0.4016 [0.3177, 0.5077]
Alkazemi, 2018 0.8650 [0.7703, 0.9715]
AlShebali, 2020 0.0748 [0.0530, 0.1054]
Alwosaifer, 2016 0.3753 [0.3160, 0.4456]
Azzouzi, 2019 0.4885 [0.4176, 0.5713]
Badrasawi, 2019 1.0000 [0.7291, 1.3715]
Barayan, 2018 0.1642 [0.1198, 0.2251]
Barry, 2021 0.4081 [0.3504, 0.4752]
Benítez, 2019 0.6260 [0.5311, 0.7379]
Bizri, 2020 Q1 0.2018 [0.1277, 0.3191]
Bizri, 2020 Q2 0.2400 [0.1537, 0.3747]
Bosi, 2016 0.1099 [0.0693, 0.1744]
Brumboiu, 2018 0.3455 [0.2556, 0.4668]
Carriedo, 2020 0.1359 [0.1113, 0.1660]
Castejón, 2020 1.2454 [1.0607, 1.4621]
Chammas, 2017 0.4695 [0.3857, 0.5715]
Chan, 2020 0.1610 [0.1347, 0.1923]
Chaudhari, 2017 0.1627 [0.1083, 0.2443]
Compte, 2015 0.0396 [0.0248, 0.0635]
Damiri, 2021 Q1 0.4582 [0.4022, 0.5221]
Damiri, 2021 Q2 0.2676 [0.2306, 0.3104]
Din, 2019 0.1810 [0.1467, 0.2233]
Ebrahim, 2019 0.8605 [0.7069, 1.0474]
Erol, 2019 0.1245 [0.0867, 0.1788]
Farchakh, 2019 Q1 2.9187 [2.4391, 3.4928]
Farchakh, 2019 Q2 0.4315 [0.3638, 0.5118]
Fatima, 2018 0.3636 [0.2426, 0.5450]
Gramaglia, 2019 Q1 0.0921 [0.0701, 0.1211]
Gramaglia, 2019 Q2 0.5885 [0.5028, 0.6889]
Grange, 1998 0.1225 [0.1036, 0.1449]
Greenleaf, 2009 0.3784 [0.2782, 0.5146]
Havemann, 2011 1.1667 [0.5396, 2.5224]
Herzog, 1985 0.1748 [0.1059, 0.2883]
Iyer, 2021 0.1773 [0.1313, 0.2395]
Jamali, 2020 Q1 0.5594 [0.4568, 0.6850]
Jamali, 2020 Q2 0.9567 [0.7877, 1.1620]
Jennings, 2006 0.1215 [0.0809, 0.1825]
Joja, 2012 0.1828 [0.1090, 0.3065]
Kara, 2021 0.6496 [0.5498, 0.7674]
Kiss-Toth, 2018 0.3589 [0.3247, 0.3968]
Ko, 2015 0.9519 [0.7229, 1.2535]
Koushiou, 2019 Q1 0.4397 [0.3483, 0.5549]
Koushiou, 2019 Q2 0.2143 [0.1621, 0.2832]
Kutlu, 2013 0.0522 [0.0299, 0.0912]
Ladner, 2019 0.3123 [0.2874, 0.3393]
Lee, 2015 0.1845 [0.1258, 0.2707]
Liao, 2006 T1 0.0318 [0.0190, 0.0531]
Liao, 2008 T2 0.0362 [0.0223, 0.0588]
Mancilla-Diaz, 2007 0.2128 [0.1854, 0.2442]
Marciano, 1988 0.1736 [0.1457, 0.2068]
Mazzaia, 2018 0.3333 [0.2205, 0.5039]
Mealha, 2013 Q1 0.0442 [0.0218, 0.0897]
Mealha, 2013 Q2 0.1962 [0.1335, 0.2883]
Momeni, 2020 0.2340 [0.1813, 0.3019]
Nancy, 2021 0.8293 [0.6013, 1.1437]
Ngan, 2017 0.1239 [0.0843, 0.1823]
Nichols, 2009 0.0436 [0.0264, 0.0719]
Padmanabhan, 2017 0.8795 [0.6422, 1.2046]
Parra-Fernández, 2018 0.2011 [0.1572, 0.2572]
Parreño-Madrigal, 2020 0.3942 [0.3232, 0.4807]
Pereira, 2011 0.2892 [0.2097, 0.3987]
Pitanupong, 2017 0.1895 [0.1583, 0.2269]
Plichta, 2019 0.3948 [0.3466, 0.4496]
Polanco, 2020 0.0843 [0.0390, 0.1824]
Radwan, 2018 0.4977 [0.4234, 0.5851]
Ramaiah, 2015 0.2028 [0.1361, 0.3023]
Rasman, 2018 0.7437 [0.5867, 0.9429]
Rathner, 1994 Q1 0.0499 [0.0311, 0.0800]
Rathner, 1994 Q2 0.1114 [0.0797, 0.1558]
Reyes-Rodríguez, 2011 0.0535 [0.0383, 0.0748]
Roshandel, 2012 0.2739 [0.2158, 0.3477]
Rostad, 2021 0.2239 [0.1914, 0.2620]
Safer, 2020 0.5387 [0.4723, 0.6145]
Saleh, 2018 Q1 0.4013 [0.3642, 0.4421]
Saleh, 2018 Q2 0.6216 [0.5680, 0.6802]
Sarah, 2021 0.2547 [0.2503, 0.2592]
Sepulveda, 2007 0.2318 [0.2099, 0.2560]
Sharifian, 2021 0.1013 [0.0897, 0.1144]
Sharma, 2019 0.2671 [0.2081, 0.3429]
Shashank, 2016 Q1 0.4105 [0.2828, 0.5960]
Shashank, 2016 Q2 0.2407 [0.1569, 0.3694]
Simona Bo, 2014 Q1 0.1000 [0.0723, 0.1384]
Simona Bo, 2014 Q2 0.3497 [0.2825, 0.4328]
Spillebout, 2019 0.2003 [0.1649, 0.2433]
Taha , 2018 Q1 0.5464 [0.4854, 0.6150]
Taha, 2018 Q2 0.1225 [0.1022, 0.1469]
Tavolacci, 2015 0.2617 [0.2410, 0.2841]
Tavolacci, 2018 0.2324 [0.2014, 0.2682]
Tavolacci, 2020 0.3295 [0.2930, 0.3705]
Thangaraju, 2020 0.1570 [0.1046, 0.2355]
Tury, 2020 Q1T1 0.0847 [0.0618, 0.1160]
Tury, 2020 Q1T2 0.1605 [0.1337, 0.1926]
Tury, 2020 Q2T1 0.0467 [0.0310, 0.0703]
Tury, 2020 Q2T2 0.0755 [0.0590, 0.0966]
Uzun, 2006 0.2070 [0.1603, 0.2673]
Weigel, 2016 0.1176 [0.0816, 0.1697]
Yoneda, 2020 0.0958 [0.0695, 0.1320]
Yu, 2015 0.0473 [0.0365, 0.0613]
Zhou, 2020 2.8235 [1.9095, 4.1751]
mean 0.2420 [0.2020, 0.2897]
prediction 0.2420 [0.0388, 1.5071]
0 0.5 1 1.5 2 2.5 3 3.5 4
Heterogeneity (tau): 0.92 [0.79, 1.06] odds

13
3230 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Fig. 6  Funnel plot of disordered

0.0
eating in university students

0.1
Standard Error

0.2
0.3
0.4

-3 -2 -1 0 1

Logit Transformed Proportion

Fig. 7  Galbraith radial plot of 1.07

disordered eating in university
students
0.32
yi 2
zi  0
vi -2

-0.44

-1.19

-1.94

-2.70

-3.45

0 2 4 6 8 10 12

xi  1 vi

was very low for many countries. Lebanon (k = 1, N = 627) Table 3 suggests that, at the very least, a greater percentage
reported the highest SBDE prevalence of 74.5% [70.1; 77.8], of university students in the USA are reporting SBDE than
while Argentina (k = 1, N = 472) and China (k = 3, N = 2,301) their counterparts in India (which has a low heterogeneity
reported the lowest percentages of 3.8 [2.4; 6.0], and 4.0 index), France, and Saudi Arabia. The difference between
[3.2; 5.0], respectively. A subgroup meta-analysis, conducted different countries was statistically significant, p = 0.001.
for the eight countries with at least 5 studies (see Table 3), Table 3 shows that non-Western countries (k = 55,
yielded evidence of statistically significant heterogeneity N = 29,663) have a slightly higher weighted mean preva-
(p = 0.001) in the prevalence of SBDE. Visual inspection of lence of SBDE, 20.9% [17.8; 24.5], than Western countries

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3231

Fig. 8  Sensitivity plot of

disordered eating in university
students

13
3232 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Fig. 9  Drapery plot of dis-

1.0
0.0

Ba

C
ordered eating in university

av

as
dr

em

te
as

jó
aw

an

n,
20

n,
students

i,

20
20

20

20

20
Q

19

11
2
0.8
0.2

0.6
0.4

Confidence level
P-value

0.4
0.6

0.2

0.8

p = 0.1 90%-CI

p = 0.05 95%-CI

p = 0.01 99%-CI
0.0

1.0

0.0 0.2 0.4 0.6 0.8

Fig. 10  Meta-regression
between sex and disordered eat-
1

ing in university students

0
Proportion of ED Symptoms

-1
-2
-3

0 20 40 60 80 100

Sex

(k = 50, N = 115,966), 18.4% [16.4; 20.6], but the difference least 5 studies were SCOFF (k = 22, N = 100,638) = 27.6%
is not statistically significant (p = 0.52; see Fig. 13). [24.1; 31.5], EAT-26 (k = 45, N = 23,821) = 16.9% [13.9;
20.3], EDE-Q (k = 6, N = 2255) = 18.1% [8.4; 35.0], EDI
Measure of screen‑based disordered eating (k = 10, N = 6,394) = 16.9% [9.6; 28.2], EAT-40 (k = 6,
N = 4355) = 10.6% [7.5; 14.9].
There was significant heterogeneity across the various meas-
ures (Table 3) used by the sample of this studies in this meta- Timeframe/years
analysis, I2 = 98.2%, τ2 = 0.36, p = 0.001 (Fig. 14). Consid-
ering the measures as 15 categories, the BEDS-7 (k = 1, Ninety-three percent of the studies were published after
N = 154) yielded the highest SBDE prevalence at 50.0% 2009, while 23% were published in 2020, 2021, or 2022.
[42.2; 57.8], while the EDE-Q (k = 1, N = 503) yielded No studies meeting the inclusion and exclusion criteria were
the lowest prevalence at 7.0% [5.04; 9.54]. In descending published between 1995 and 2004. Results of subgroup
order, the prevalence of SBDE for the measures used in at meta-analysis showed a statistically significant (p = 0.001)

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3233

Fig. 11  Meta-regression
between BMI and disordered

1
eating in university students

0
Proportion of ED Symptoms

-1
-2
-3

18 20 22 24 26

BMI

increase in the prevalence of SBDE among university stu- (approximately 10) [53, 54]. This is perhaps due to the
dents (see Fig. 15). Specifically, there appears to be an margin of error of screen-based measurement tools com-
increase between 2005–2014 and 2015–2022; for the catego- bined with the fact that disordered eating focuses on the
ries 2005–2009, 2010–2014, 2015–2019 and 2020 onwards presence of individual symptoms while an eating disor-
the weighted pooled prevalence of screen-based disordered der focuses on meeting a group of symptoms for a mini-
eating was 10.6% [07.3; 15.1], 13.0% [8.3; 19.8], 23.8% mum period of time to meet established diagnostic cri-
[20.7; 27.2] and 20.8% [17.6; 24.5], respectively. To fur- teria. However, a figure of 15–20% is also what would
ther illustrate the effect of year on the prevalence of SBDE be expected if a number of those sometimes co-occurring
among university students, a meta-regression showed that symptoms were normally distributed within the popula-
time of publication is a statistically significant predictor tions that have been studied to date.
(p = 0.001) of increased prevalence rate of SBDE among At 19.7%, our overall prevalence rate of screen-based
university students (see Fig. 16). disordered eating corresponds with perfectly Levine and
Smolak’s (2021) conclusion based on their narrative review
[12]. Furthermore our results are consistent with the findings
Discussion of the two studies with the largest sample sizes: 20.3% and
20.7% in the studies by Falvey, Hahn, Anderson, Lipson, and
This meta-analysis of 89 studies (total N = 145,629) from Sonneville (2021; N = 77,193) and Tavolacci et al. (2015;
40 countries suggests that the prevalence of screen-based N = 3457), respectively [107, 153]. We acknowledge that
disordered eating among university undergraduate students further research is needed, because the prevalence may be
is 19.7%. We also found that increasing BMI is a strong lower. If we assume that the sensitivity of measures such as
statistical confounder, while female sex is a statistically the EAT-26 and the SCOFF is around 85% [6], this means
significant but weak confounder. Age, which typically does that at least 15% of 20%, or at least 3%, have disordered
not vary a great deal for undergraduates, had an insignifi- eating beliefs, anxieties, and behaviors that are correlated
cant effect. Non-Western countries have a slightly higher with a wide variety of health problems and that put them
risk prevalence of screen-based disordered eating (20.9%) at risk for a possible eating disorder [12, 164]. However,
compared to Western countries (18.4%), but the difference regardless of the psychometrics of the screening measures,
is not statistically significant. Although slight asymmetry epidemiological studies indicate that it is highly unlikely
to the right was apparent in the funnel plot, using Egger’s that 20−3% = 17%, that is, 1 in 6, university undergraduates
regression test we ruled out significant heterogeneity. have a diagnosable but not yet diagnosed eating disorder.
The overall screen-based prevalence rate of disor- If we place the point prevalence of DSM-5-defined eating
dered eating (nearly 20%) is approximately twice the disorders at a conservative estimate of 8–10% [8], then our
global prevalence of eating disorders estimates of around meta-analytic findings suggest, again conservatively, that

13
3234 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Fig. 12  Subgroup meta-analysis

Events per 100
Study Number Total Prevalence (%) 95%CI observations

Country = Malasyia

by Country
Abdul Manaf , 2016 13 206 6.3 [ 3.4; 10.5]
Chan, 2020
Ngan, 2017
141
29
1017
263
13.9 [11.8; 16.1]
11.0 [ 7.5; 15.5] Events per 100
Rasman, 2018
Random effects model
119 279
1765
42.7 [36.8; 48.7]
15.3 [ 6.3; 32.6] Study Number Total Prevalence (%) 95%CI observations
Heterogeneity: I 2 = 97.82%,  2 = 0.97, p < 0.001

Country = Egypt
Abo Ali, 2020
Random effects model
203 615
615
33.0 [29.3; 36.9]
33.0 [29.4; 36.8]
Western = No
Heterogeneity: not applicable
Abdul Manaf , 2016 13 206 6.3 [ 3.4; 10.5]
Abo Ali, 2020
Country = Turkey
Akdevelioglu, 2010
Erol, 2019
34
33
577
298
5.9 [ 4.1; 8.1]
11.1 [ 7.7; 15.2]
203 615 33.0 [29.3; 36.9]
Kutlu, 2013
Uzun, 2006
13
71
262
414
5.0 [ 2.7; 8.3]
17.1 [13.6; 21.1]
Akdevelioglu, 2010 34 577 5.9 [ 4.1; 8.1]
Random effects model 1551
Heterogeneity: I 2 = 92.42%,  2 = 0.40, p < 0.001
8.9 [ 4.9; 15.8]
Al Banna, 2021 84 365 23.0 [18.8; 27.7]
Country = Bangladesh Albrahim, 2019 145 396 36.6 [31.9; 41.6]
Al Banna, 2021 84 365 23.0 [18.8; 27.7]
Random effects model
Heterogeneity: not applicable
365 23.0 [19.0; 27.6] Alhazmi, 2019 98 342 28.7 [23.9; 33.8]
Country = Saudi Arabia Alkazemi, 2018 532 1147 46.4 [43.5; 49.3]
Albrahim, 2019
Alhazmi, 2019
145
98
396
342
36.6
28.7
[31.9; 41.6]
[23.9; 33.8] AlShebali, 2020 35 503 7.0 [ 4.9; 9.5]
AlShebali, 2020 35 503 7.0 [ 4.9; 9.5]
Alwosaifer, 2016
Barayan, 2018
179
45
656
319
27.3
14.1
[23.9; 30.9]
[10.5; 18.4]
Alwosaifer, 2016 179 656 27.3 [23.9; 30.9]
Fatima, 2018
Taha , 2018 Q1
32
424
120
1200
26.7
35.3
[19.0; 35.5]
[32.6; 38.1]
Azzouzi, 2019 233 710 32.8 [29.4; 36.4]
Taha, 2018 Q2
Random effects model
131 1200
4736
10.9
21.2
[ 9.2; 12.8]
[14.1; 30.5] Badrasawi, 2019 77 154 50.0 [41.8; 58.2]
Heterogeneity: I 2 = 97.72%,  2 = 0.48, p < 0.001

Country = Spain
Barayan, 2018 45 319 14.1 [10.5; 18.4]
Alcaraz-Ibáñez, 2019
Benítez, 2019
223
231
545
600
40.9
38.5
[36.8; 45.2]
[34.6; 42.5]
Bizri, 2020 Q1 22 131 16.8 [10.8; 24.3]
Castejón, 2020
Parra-Fernández, 2018
335
76
604
454
55.5
16.7
[51.4; 59.5]
[13.4; 20.5] Bizri, 2020 Q2 24 124 19.4 [12.8; 27.4]
Chammas, 2017 146 457 31.9 [27.7; 36.4]
Parreño-Madrigal, 2020 136 481 28.3 [24.3; 32.5]
Sepulveda, 2007 480 2551 18.8 [17.3; 20.4]
Random effects model 5235 31.7 [20.4; 45.6]
Heterogeneity: I 2 = 98.76%,  2 = 0.54, p < 0.001 Chan, 2020 141 1017 13.9 [11.8; 16.1]
Country = Kuwait
Alkazemi, 2018 532 1147 46.4 [43.5; 49.3] Chaudhari, 2017 27 193 14.0 [ 9.4; 19.7]
Ebrahim, 2019
Random effects model
185 400
1547
46.2 [41.3; 51.3]
46.3 [43.9; 48.8] Damiri, 2021 Q1 329 1047 31.4 [28.6; 34.3]
Heterogeneity: I 2 = 0%,  2 = 0, p = 0.964

Country = Morocco
Damiri, 2021 Q2 221 1047 21.1 [18.7; 23.7]
Azzouzi, 2019
Random effects model
233 710
710
32.8 [29.4; 36.4]
32.8 [29.5; 36.4]
Din, 2019 103 672 15.3 [12.7; 18.3]
Heterogeneity: not applicable
Ebrahim, 2019 185 400 46.2 [41.3; 51.3]
Country = Palestine
Badrasawi, 2019
Damiri, 2021 Q1
77
329
154
1047
50.0
31.4
[41.8; 58.2]
[28.6; 34.3]
Erol, 2019 33 298 11.1 [ 7.7; 15.2]
Damiri, 2021 Q2
Saleh, 2018 Q1
221
573
1047
2001
21.1
28.6
[18.7; 23.7]
[26.7; 30.7]
Farchakh, 2019 Q1 467 627 74.5 [70.9; 77.9]
Saleh, 2018 Q2
Random effects model
767 2001
6250
38.3
32.8
[36.2; 40.5]
[26.2; 40.2] Farchakh, 2019 Q2 189 627 30.1 [26.6; 33.9]
Fatima, 2018 32 120 26.7 [19.0; 35.5]
Heterogeneity: I 2 = 96.79%,  2 = 0.13, p < 0.001

Country = United States

Barry, 2021
Greenleaf, 2009
233
56
804
204
29.0
27.5
[25.9; 32.3]
[21.5; 34.1]
Herzog, 1985 18 121 14.9 [ 9.1; 22.5]
Herzog, 1985
Kara, 2021
18
228
121
579
14.9
39.4
[ 9.1; 22.5]
[35.4; 43.5] Iyer, 2021 50 332 15.1 [11.4; 19.4]
Nancy, 2021
Zhou, 2020
68
96
150
130
45.3
73.8
[37.2; 53.7]
[65.4; 81.2] Jamali, 2020 Q1 146 407 35.9 [31.2; 40.7]
Random effects model 1988 37.2 [26.3; 49.5]
Heterogeneity: I 2 = 95.76%,  2 = 0.37, p < 0.001 Jamali, 2020 Q2 199 407 48.9 [43.9; 53.9]
Country = Lebanon
Bizri, 2020 Q1 22 131 16.8 [10.8; 24.3]
Ko, 2015 99 203 48.8 [41.7; 55.9]
Bizri, 2020 Q2
Chammas, 2017
24
146
124
457
19.4 [12.8; 27.4]
31.9 [27.7; 36.4] Kutlu, 2013 13 262 5.0 [ 2.7; 8.3]
Lee, 2015 31 199 15.6 [10.8; 21.4]
Farchakh, 2019 Q1 467 627 74.5 [70.9; 77.9]
Farchakh, 2019 Q2 189 627 30.1 [26.6; 33.9]
Random effects model 1966 33.2 [15.9; 56.7]
Heterogeneity: I 2 = 98.82%,  2 = 1.20, p < 0.001
Liao, 2006 T1 15 487 3.1 [ 1.7; 5.0]
Country = Brazil
Bosi, 2016 20 202 9.9 [ 6.2; 14.9] Liao, 2008 T2 17 486 3.5 [ 2.1; 5.5]
Momeni, 2020 73 385 19.0 [15.2; 23.2]
Mazzaia, 2018 30 120 25.0 [17.5; 33.7]
Pereira, 2011 48 214 22.4 [17.0; 28.6]
Random effects model 536 18.2 [10.7; 29.3]
Heterogeneity: I 2 = 86.41%,  2 = 0.26, p < 0.001 Ngan, 2017 29 263 11.0 [ 7.5; 15.5]
Country = Romania
Brumboiu, 2018 57 222 25.7 [20.1; 31.9] Nichols, 2009 16 383 4.2 [ 2.4; 6.7]
Random effects model
Heterogeneity: not applicable
222 25.7 [20.4; 31.8]
Padmanabhan, 2017 73 156 46.8 [38.8; 54.9]
Country = Mexico
Carriedo, 2020 109 911 12.0 [ 9.9; 14.3]
Pitanupong, 2017 141 885 15.9 [13.6; 18.5]
Mancilla-Diaz, 2007
Polanco, 2020
246
7
1402
90
17.5 [15.6; 19.6]
7.8 [ 3.2; 15.4]
Radwan, 2018 220 662 33.2 [29.7; 37.0]
Random effects model 2403
Heterogeneity: I 2 = 88.2%,  2 = 0.11, p < 0.001
13.1 [ 9.0; 18.7]
Ramaiah, 2015 29 172 16.9 [11.6; 23.3]
Country = India
Chaudhari, 2017 27 193 14.0 [ 9.4; 19.7]
Rasman, 2018 119 279 42.7 [36.8; 48.7]
Iyer, 2021
Ramaiah, 2015
50
29
332
172
15.1
16.9
[11.4; 19.4]
[11.6; 23.3]
Roshandel, 2012 86 400 21.5 [17.6; 25.9]
Sharma, 2019
Shashank, 2016 Q1
78
39
370
134
21.1
29.1
[17.0; 25.6]
[21.6; 37.6] Safer, 2020 341 974 35.0 [32.0; 38.1]
Saleh, 2018 Q1 573 2001 28.6 [26.7; 30.7]
Shashank, 2016 Q2 26 134 19.4 [13.1; 27.1]
Thangaraju, 2020 27 199 13.6 [ 9.1; 19.1]
Random effects model 1534 18.1 [14.7; 22.0]
Heterogeneity: I 2 = 70.07%,  2 = 0.08, p = 0.003 Saleh, 2018 Q2 767 2001 38.3 [36.2; 40.5]
Country = Argentina
Compte, 2015 18 472 3.8 [ 2.3; 6.0] Sharma, 2019 78 370 21.1 [17.0; 25.6]
Random effects model
Heterogeneity: not applicable
472 3.8 [ 2.4; 6.0]
Shashank, 2016 Q1 39 134 29.1 [21.6; 37.6]
Country = Pakistan
Din, 2019 103 672 15.3 [12.7; 18.3]
Shashank, 2016 Q2 26 134 19.4 [13.1; 27.1]
Jamali, 2020 Q1
Jamali, 2020 Q2
146
199
407
407
35.9 [31.2; 40.7]
48.9 [43.9; 53.9]
Taha , 2018 Q1 424 1200 35.3 [32.6; 38.1]
Random effects model 1486
Heterogeneity: I 2 = 98.5%,  2 = 0.70, p < 0.001
31.5 [15.0; 54.4]
Taha, 2018 Q2 131 1200 10.9 [ 9.2; 12.8]
Country = Multi
Gramaglia, 2019 Q1 56 664 8.4 [ 6.4; 10.8]
Thangaraju, 2020 27 199 13.6 [ 9.1; 19.1]
Gramaglia, 2019 Q2
Kiss-Toth, 2018
246
519
664
1965
37.0
26.4
[33.4; 40.8]
[24.5; 28.4]
Uzun, 2006 71 414 17.1 [13.6; 21.1]
Sarah, 2021
Random effects model
15671 77193
80486
20.3
21.3
[20.0; 20.6]
[15.1; 29.1] Yoneda, 2020 41 469 8.7 [ 6.3; 11.7]
Yu, 2015 60 1328 4.5 [ 3.5; 5.8]
Heterogeneity: I 2 = 98.52%,  2 = 0.18, p < 0.001

Country = South Africa

Grange, 1998
Havemann, 2011
153
14
1402
26
10.9 [ 9.3; 12.7]
53.8 [33.4; 73.4]
Random effects model 29663 20.9 [17.8; 24.4]
Random effects model 1428
Heterogeneity: I 2 = 96.81%,  2 = 2.46, p < 0.001
26.8 [ 3.9; 76.9]
Heterogeneity: I 2 = 97.99%,  2 = 0.55, p = 0
Country = Australia
Jennings, 2006 26 240 10.8 [ 7.2; 15.5]
Random effects model
Heterogeneity: not applicable
240 10.8 [ 7.5; 15.4] Western = Yes
Country = Germany
Alcaraz-Ibáñez, 2019 223 545 40.9 [36.8; 45.2]
Joja, 2012
Weigel, 2016
17
32
110
304
15.5 [ 9.3; 23.6]
10.5 [ 7.3; 14.5] Barry, 2021 233 804 29.0 [25.9; 32.3]
Random effects model 414 12.4 [ 8.4; 17.7]
Heterogeneity: I 2 = 46.19%,  2 = 0.04, p = 0.173 Benítez, 2019 231 600 38.5 [34.6; 42.5]
Country = Vietnam
Ko, 2015 99 203 48.8 [41.7; 55.9]
Bosi, 2016 20 202 9.9 [ 6.2; 14.9]
Random effects model
Heterogeneity: not applicable
203 48.8 [42.0; 55.6]
Brumboiu, 2018 57 222 25.7 [20.1; 31.9]
Country = Greece Carriedo, 2020 109 911 12.0 [ 9.9; 14.3]
Koushiou, 2019 Q1 102 334 30.5 [25.6; 35.8]
Koushiou, 2019 Q2
Random effects model
60 340
674
17.6 [13.7; 22.1]
23.6 [13.2; 38.4]
Castejón, 2020 335 604 55.5 [51.4; 59.5]
Heterogeneity: I 2 = 93.35%,  2 = 0.24, p < 0.001
Compte, 2015 18 472 3.8 [ 2.3; 6.0]
Country = France
Ladner, 2019 732 3076 23.8 [22.3; 25.3] Gramaglia, 2019 Q1 56 664 8.4 [ 6.4; 10.8]
Spillebout, 2019 122 731 16.7 [14.1; 19.6]
Tavolacci, 2015
Tavolacci, 2018
717
231
3457
1225
20.7
18.9
[19.4; 22.1]
[16.7; 21.2]
Gramaglia, 2019 Q2 246 664 37.0 [33.4; 40.8]
Tavolacci, 2020
Random effects model
370 1493
9982
24.8
21.0
[22.6; 27.1]
[18.7; 23.6]
Grange, 1998 153 1402 10.9 [ 9.3; 12.7]
Heterogeneity: I 2 = 88.41%,  2 = 0.02, p < 0.001
Greenleaf, 2009 56 204 27.5 [21.5; 34.1]
Country = Korea
Lee, 2015
Random effects model
31 199
199
15.6 [10.8; 21.4]
15.6 [11.2; 21.3]
Havemann, 2011 14 26 53.8 [33.4; 73.4]
Heterogeneity: not applicable
Jennings, 2006 26 240 10.8 [ 7.2; 15.5]
Country = China
Liao, 2006 T1 15 487 3.1 [ 1.7; 5.0] Joja, 2012 17 110 15.5 [ 9.3; 23.6]
Kara, 2021 228 579 39.4 [35.4; 43.5]
Liao, 2008 T2 17 486 3.5 [ 2.1; 5.5]
Yu, 2015 60 1328 4.5 [ 3.5; 5.8]
Random effects model 2301 4.0 [ 3.2; 5.0]
Heterogeneity: I 2 = 13.02%,  2 = < 0.01, p = 0.317 Kiss-Toth, 2018 519 1965 26.4 [24.5; 28.4]
Country = Canada
Marciano, 1988 147 994 14.8 [12.6; 17.1] Koushiou, 2019 Q1 102 334 30.5 [25.6; 35.8]
Random effects model
Heterogeneity: not applicable
994 14.8 [12.7; 17.1]
Koushiou, 2019 Q2 60 340 17.6 [13.7; 22.1]
Country = Portugal
Mealha, 2013 Q1 8 189 4.2 [ 1.8; 8.2]
Ladner, 2019 732 3076 23.8 [22.3; 25.3]
Mealha, 2013 Q2
Random effects model
31 189
378
16.4 [11.4; 22.5]
8.8 [ 2.2; 29.2]
Mancilla-Diaz, 2007 246 1402 17.5 [15.6; 19.6]
Heterogeneity: I 2 = 92.39%,  2 = 1.03, p < 0.001
Marciano, 1988 147 994 14.8 [12.6; 17.1]
Country = Iran
Momeni, 2020
Roshandel, 2012
73
86
385
400
19.0 [15.2; 23.2]
21.5 [17.6; 25.9]
Mazzaia, 2018 30 120 25.0 [17.5; 33.7]
Random effects model
Heterogeneity: I 2 = 0%,  2 = 0, p = 0.376
785 20.3 [17.6; 23.2]
Mealha, 2013 Q1 8 189 4.2 [ 1.8; 8.2]
Country = West Indies Mealha, 2013 Q2 31 189 16.4 [11.4; 22.5]
Nancy, 2021 68 150 45.3 [37.2; 53.7]
Nichols, 2009 16 383 4.2 [ 2.4; 6.7]
Random effects model 383 4.2 [ 2.6; 6.7]
Heterogeneity: not applicable

Country = United Arab Emirates

Parra-Fernández, 2018 76 454 16.7 [13.4; 20.5]
Padmanabhan, 2017
Radwan, 2018
73
220
156
662
46.8 [38.8; 54.9]
33.2 [29.7; 37.0] Parreño-Madrigal, 2020 136 481 28.3 [24.3; 32.5]
Random effects model 818
Heterogeneity: I 2 = 89.96%,  2 = 0.15, p = 0.002
39.4 [27.2; 53.2]
Pereira, 2011 48 214 22.4 [17.0; 28.6]
Country = Thailand
Pitanupong, 2017 141 885 15.9 [13.6; 18.5]
Plichta, 2019 317 1120 28.3 [25.7; 31.0]
Random effects model
Heterogeneity: not applicable
885 15.9 [13.7; 18.5]
Polanco, 2020 7 90 7.8 [ 3.2; 15.4]
Country = Poland Rathner, 1994 Q1 18 379 4.7 [ 2.8; 7.4]
Rathner, 1994 Q2 38 379 10.0 [ 7.2; 13.5]
Plichta, 2019 317 1120 28.3 [25.7; 31.0]
Random effects model 1120 28.3 [25.7; 31.0]
Heterogeneity: not applicable

Country = Austria
Reyes-Rodríguez, 2011 36 709 5.1 [ 3.6; 7.0]
Rathner, 1994 Q1
Rathner, 1994 Q2
18
38
379
379
4.7 [ 2.8; 7.4]
10.0 [ 7.2; 13.5] Rostad, 2021 191 1044 18.3 [16.0; 20.8]
Sarah, 2021 15671 77193 20.3 [20.0; 20.6]
Random effects model 758 7.1 [ 3.3; 14.3]
Heterogeneity: I 2 = 86.46%,  2 = 0.28, p = 0.007

Country = Puerto Rico

Reyes-Rodríguez, 2011 36 709 5.1 [ 3.6; 7.0]
Sepulveda, 2007 480 2551 18.8 [17.3; 20.4]
Random effects model
Heterogeneity: not applicable
709 5.1 [ 3.7; 7.0]
Sharifian, 2021 286 3110 9.2 [ 8.2; 10.3]
Country = Norway Simona Bo, 2014 Q1 40 440 9.1 [ 6.6; 12.2]
Rostad, 2021 191 1044 18.3 [16.0; 20.8]
Random effects model
Heterogeneity: not applicable
1044 18.3 [16.1; 20.8] Simona Bo, 2014 Q2 114 440 25.9 [21.9; 30.3]
Country = Tunisia
Spillebout, 2019 122 731 16.7 [14.1; 19.6]
Safer, 2020
Random effects model
341 974
974
35.0 [32.0; 38.1]
35.0 [32.1; 38.1] Tavolacci, 2015 717 3457 20.7 [19.4; 22.1]
Tavolacci, 2018 231 1225 18.9 [16.7; 21.2]
Heterogeneity: not applicable

Country = Finland
Sharifian, 2021
Random effects model
286 3110
3110
9.2 [ 8.2; 10.3]
9.2 [ 8.2; 10.3]
Tavolacci, 2020 370 1493 24.8 [22.6; 27.1]
Heterogeneity: not applicable
Tury, 2020 Q1T1 42 538 7.8 [ 5.7; 10.4]
Tury, 2020 Q1T2 134 969 13.8 [11.7; 16.2]
Country = Italy
Simona Bo, 2014 Q1 40 440 9.1 [ 6.6; 12.2]
Simona Bo, 2014 Q2 114 440 25.9 [21.9; 30.3]
Random effects model 880
Heterogeneity: I 2 = 97.49%,  2 = 0.76, p < 0.001
15.8 [ 5.2; 39.1] Tury, 2020 Q2T1 24 538 4.5 [ 2.9; 6.6]
Country = Hungary Tury, 2020 Q2T2 68 969 7.0 [ 5.5; 8.8]
Tury, 2020 Q1T1
Tury, 2020 Q1T2
42
134
538
969
7.8 [ 5.7; 10.4]
13.8 [11.7; 16.2] Weigel, 2016 32 304 10.5 [ 7.3; 14.5]
Tury, 2020 Q2T1 24 538 4.5 [ 2.9; 6.6]
Tury, 2020 Q2T2
Random effects model
68 969
3014
7.0 [ 5.5; 8.8]
7.8 [ 4.8; 12.5]
Zhou, 2020 96 130 73.8 [65.4; 81.2]
Heterogeneity: I 2 = 93.26%,  2 = 0.26, p < 0.001
Random effects model 115966 18.4 [16.4; 20.6]
Country = Japan
Yoneda, 2020 41 469 8.7 [ 6.3; 11.7] Heterogeneity: I 2 = 97.84%,  2 = 0.23, p = 0
Random effects model 469 8.7 [ 6.5; 11.7]
Heterogeneity: not applicable

Random effects model

2 2
Heterogeneity: I = 98.17%,  = 0.34, p = 0
145629 19.7 [17.9; 21.6]
Random effects model 145629 19.7 [17.9; 21.6]
0 20 40 60
Prevalence (%)
80 100
Heterogeneity: I 2 = 98.17%,  2 = 0.34, p = 0
0 20 40 60 80 100
Prevalence (%)

Fig. 13  Subgroup meta-analysis by Culture.

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3235

Events per 100

Study Number Total Prevalence (%) 95%CI observations

Measure = EAT-26 Events per 100

Abdul Manaf , 2016 13 206 6.3 [ 3.4; 10.5] Study Number Total Prevalence (%) 95%CI observations
Abo Ali, 2020 203 615 33.0 [29.3; 36.9]
Al Banna, 2021 84 365 23.0 [18.8; 27.7]
Albrahim, 2019 145 396 36.6 [31.9; 41.6] YearCat = 2015-2019
Alhazmi, 2019 98 342 28.7 [23.9; 33.8] Abdul Manaf , 2016 13 206 6.3 [ 3.4; 10.5]
Alkazemi, 2018 532 1147 46.4 [43.5; 49.3]
Albrahim, 2019 145 396 36.6 [31.9; 41.6]
Alwosaifer, 2016 179 656 27.3 [23.9; 30.9]
Bizri, 2020 Q1 22 131 16.8 [10.8; 24.3] Alcaraz-Ibáñez, 2019 223 545 40.9 [36.8; 45.2]
Bosi, 2016 20 202 9.9 [ 6.2; 14.9] Alhazmi, 2019 98 342 28.7 [23.9; 33.8]
Chan, 2020 141 1017 13.9 [11.8; 16.1] Alkazemi, 2018 532 1147 46.4 [43.5; 49.3]
Compte, 2015 18 472 3.8 [ 2.3; 6.0]
Damiri, 2021 Q2 221 1047 21.1 [18.7; 23.7]
Alwosaifer, 2016 179 656 27.3 [23.9; 30.9]
Din, 2019 103 672 15.3 [12.7; 18.3] Azzouzi, 2019 233 710 32.8 [29.4; 36.4]
Ebrahim, 2019 185 400 46.2 [41.3; 51.3] Badrasawi, 2019 77 154 50.0 [41.8; 58.2]
Farchakh, 2019 Q2 189 627 30.1 [26.6; 33.9] Barayan, 2018 45 319 14.1 [10.5; 18.4]
Fatima, 2018 32 120 26.7 [19.0; 35.5]
Gramaglia, 2019 Q1 56 664 8.4 [ 6.4; 10.8]
Benítez, 2019 231 600 38.5 [34.6; 42.5]
Havemann, 2011 14 26 53.8 [33.4; 73.4] Bosi, 2016 20 202 9.9 [ 6.2; 14.9]
Iyer, 2021 50 332 15.1 [11.4; 19.4] Brumboiu, 2018 57 222 25.7 [20.1; 31.9]
Jamali, 2020 Q1 146 407 35.9 [31.2; 40.7] Chammas, 2017 146 457 31.9 [27.7; 36.4]
Jennings, 2006 26 240 10.8 [ 7.2; 15.5]
Liao, 2006 T1 15 487 3.1 [ 1.7; 5.0] Chaudhari, 2017 27 193 14.0 [ 9.4; 19.7]
Liao, 2008 T2 17 486 3.5 [ 2.1; 5.5] Compte, 2015 18 472 3.8 [ 2.3; 6.0]
Marciano, 1988 147 994 14.8 [12.6; 17.1] Din, 2019 103 672 15.3 [12.7; 18.3]
Mazzaia, 2018 30 120 25.0 [17.5; 33.7] Ebrahim, 2019 185 400 46.2 [41.3; 51.3]
Mealha, 2013 Q1 8 189 4.2 [ 1.8; 8.2]
Momeni, 2020 73 385 19.0 [15.2; 23.2] Erol, 2019 33 298 11.1 [ 7.7; 15.2]
Ngan, 2017 29 263 11.0 [ 7.5; 15.5] Farchakh, 2019 Q1 467 627 74.5 [70.9; 77.9]
Nichols, 2009 16 383 4.2 [ 2.4; 6.7] Farchakh, 2019 Q2 189 627 30.1 [26.6; 33.9]
Padmanabhan, 2017 73 156 46.8 [38.8; 54.9]
Fatima, 2018 32 120 26.7 [19.0; 35.5]
Pereira, 2011 48 214 22.4 [17.0; 28.6]
Pitanupong, 2017 141 885 15.9 [13.6; 18.5] Gramaglia, 2019 Q1 56 664 8.4 [ 6.4; 10.8]
Polanco, 2020 7 90 7.8 [ 3.2; 15.4] Gramaglia, 2019 Q2 246 664 37.0 [33.4; 40.8]
Radwan, 2018 220 662 33.2 [29.7; 37.0] Kiss-Toth, 2018 519 1965 26.4 [24.5; 28.4]
Ramaiah, 2015 29 172 16.9 [11.6; 23.3]
Reyes-Rodríguez, 2011 36 709 5.1 [ 3.6; 7.0]
Ko, 2015 99 203 48.8 [41.7; 55.9]
Roshandel, 2012 86 400 21.5 [17.6; 25.9] Koushiou, 2019 Q1 102 334 30.5 [25.6; 35.8]
Saleh, 2018 Q1 573 2001 28.6 [26.7; 30.7] Koushiou, 2019 Q2 60 340 17.6 [13.7; 22.1]
Sharma, 2019 78 370 21.1 [17.0; 25.6] Ladner, 2019 732 3076 23.8 [22.3; 25.3]
Shashank, 2016 Q1 39 134 29.1 [21.6; 37.6]
Simona Bo, 2014 Q1 40 440 9.1 [ 6.6; 12.2]
Lee, 2015 31 199 15.6 [10.8; 21.4]
Taha , 2018 Q1 424 1200 35.3 [32.6; 38.1] Mazzaia, 2018 30 120 25.0 [17.5; 33.7]
Taha, 2018 Q2 131 1200 10.9 [ 9.2; 12.8] Ngan, 2017 29 263 11.0 [ 7.5; 15.5]
Yoneda, 2020 41 469 8.7 [ 6.3; 11.7] Padmanabhan, 2017 73 156 46.8 [38.8; 54.9]
Yu, 2015 60 1328 4.5 [ 3.5; 5.8]
Random effects model 23821 16.9 [13.9; 20.3] Parra-Fernández, 2018 76 454 16.7 [13.4; 20.5]
Heterogeneity: I 2 = 97.69%,  2 = 0.56, p = 0 Pitanupong, 2017 141 885 15.9 [13.6; 18.5]
Plichta, 2019 317 1120 28.3 [25.7; 31.0]
Measure = EAT-40
Radwan, 2018 220 662 33.2 [29.7; 37.0]
Akdevelioglu, 2010 34 577 5.9 [ 4.1; 8.1]
Erol, 2019 33 298 11.1 [ 7.7; 15.2] Ramaiah, 2015 29 172 16.9 [11.6; 23.3]
Grange, 1998 153 1402 10.9 [ 9.3; 12.7] Rasman, 2018 119 279 42.7 [36.8; 48.7]
Kutlu, 2013 13 262 5.0 [ 2.7; 8.3] Saleh, 2018 Q1 573 2001 28.6 [26.7; 30.7]
Mancilla-Diaz, 2007 246 1402 17.5 [15.6; 19.6]
Uzun, 2006 71 414 17.1 [13.6; 21.1]
Saleh, 2018 Q2 767 2001 38.3 [36.2; 40.5]
Random effects model 4355 10.6 [ 7.4; 14.9] Sharma, 2019 78 370 21.1 [17.0; 25.6]
Heterogeneity: I 2 = 93.35%,  2 = 0.21, p < 0.001 Shashank, 2016 Q1 39 134 29.1 [21.6; 37.6]
Shashank, 2016 Q2 26 134 19.4 [13.1; 27.1]
Measure = SCOFF
Alcaraz-Ibáñez, 2019 223 545 40.9 [36.8; 45.2]
Spillebout, 2019 122 731 16.7 [14.1; 19.6]
Azzouzi, 2019 233 710 32.8 [29.4; 36.4] Taha , 2018 Q1 424 1200 35.3 [32.6; 38.1]
Barry, 2021 233 804 29.0 [25.9; 32.3] Taha, 2018 Q2 131 1200 10.9 [ 9.2; 12.8]
Bizri, 2020 Q2 24 124 19.4 [12.8; 27.4] Tavolacci, 2015 717 3457 20.7 [19.4; 22.1]
Brumboiu, 2018 57 222 25.7 [20.1; 31.9]
Chammas, 2017 146 457 31.9 [27.7; 36.4] Tavolacci, 2018 231 1225 18.9 [16.7; 21.2]
Damiri, 2021 Q1 329 1047 31.4 [28.6; 34.3] Weigel, 2016 32 304 10.5 [ 7.3; 14.5]
Jamali, 2020 Q2 199 407 48.9 [43.9; 53.9] Yu, 2015 60 1328 4.5 [ 3.5; 5.8]
Kiss-Toth, 2018 519 1965 26.4 [24.5; 28.4] Random effects model 35006 23.8 [20.7; 27.2]
Ko, 2015 99 203 48.8 [41.7; 55.9]
Ladner, 2019 732 3076 23.8 [22.3; 25.3] Heterogeneity: I 2 = 97.94%,  2 = 0.39, p = 0
Parreño-Madrigal, 2020 136 481 28.3 [24.3; 32.5]
Rasman, 2018 119 279 42.7 [36.8; 48.7] YearCat = 2020 Onwards
Safer, 2020 341 974 35.0 [32.0; 38.1]
Abo Ali, 2020 203 615 33.0 [29.3; 36.9]
Saleh, 2018 Q2 767 2001 38.3 [36.2; 40.5]
Sarah, 2021 15671 77193 20.3 [20.0; 20.6] Al Banna, 2021 84 365 23.0 [18.8; 27.7]
Sharifian, 2021 286 3110 9.2 [ 8.2; 10.3] AlShebali, 2020 35 503 7.0 [ 4.9; 9.5]
Shashank, 2016 Q2 26 134 19.4 [13.1; 27.1] Barry, 2021 233 804 29.0 [25.9; 32.3]
Spillebout, 2019 122 731 16.7 [14.1; 19.6]
Tavolacci, 2015 717 3457 20.7 [19.4; 22.1]
Bizri, 2020 Q1 22 131 16.8 [10.8; 24.3]
Tavolacci, 2018 231 1225 18.9 [16.7; 21.2] Bizri, 2020 Q2 24 124 19.4 [12.8; 27.4]
Tavolacci, 2020 370 1493 24.8 [22.6; 27.1] Carriedo, 2020 109 911 12.0 [ 9.9; 14.3]
Random effects model 100638 27.6 [24.1; 31.5] Castejón, 2020 335 604 55.5 [51.4; 59.5]
Heterogeneity: I 2 = 98.5%,  2 = 0.19, p < 0.001
Chan, 2020 141 1017 13.9 [11.8; 16.1]
Measure = EDE-Q Damiri, 2021 Q1 329 1047 31.4 [28.6; 34.3]
AlShebali, 2020 35 503 7.0 [ 4.9; 9.5] Damiri, 2021 Q2 221 1047 21.1 [18.7; 23.7]
Barayan, 2018 45 319 14.1 [10.5; 18.4] Iyer, 2021 50 332 15.1 [11.4; 19.4]
Carriedo, 2020 109 911 12.0 [ 9.9; 14.3]
Chaudhari, 2017 27 193 14.0 [ 9.4; 19.7] Jamali, 2020 Q1 146 407 35.9 [31.2; 40.7]
Thangaraju, 2020 27 199 13.6 [ 9.1; 19.1] Jamali, 2020 Q2 199 407 48.9 [43.9; 53.9]
Zhou, 2020 96 130 73.8 [65.4; 81.2] Kara, 2021 228 579 39.4 [35.4; 43.5]
Random effects model 2255 18.1 [ 8.3; 35.0]
Momeni, 2020 73 385 19.0 [15.2; 23.2]
Heterogeneity: I 2 = 97.77%,  2 = 1.20, p < 0.001
Nancy, 2021 68 150 45.3 [37.2; 53.7]
Measure = BEDS-7 Parreño-Madrigal, 2020 136 481 28.3 [24.3; 32.5]
Badrasawi, 2019 77 154 50.0 [41.8; 58.2] Polanco, 2020 7 90 7.8 [ 3.2; 15.4]
Random effects model 154 50.0 [42.2; 57.8]
Heterogeneity: not applicable
Rostad, 2021 191 1044 18.3 [16.0; 20.8]
Safer, 2020 341 974 35.0 [32.0; 38.1]
Measure = EDI Sarah, 2021 15671 77193 20.3 [20.0; 20.6]
Benítez, 2019 231 600 38.5 [34.6; 42.5] Sharifian, 2021 286 3110 9.2 [ 8.2; 10.3]
Castejón, 2020 335 604 55.5 [51.4; 59.5]
Joja, 2012 17 110 15.5 [ 9.3; 23.6]
Tavolacci, 2020 370 1493 24.8 [22.6; 27.1]
Mealha, 2013 Q2 31 189 16.4 [11.4; 22.5] Thangaraju, 2020 27 199 13.6 [ 9.1; 19.1]
Nancy, 2021 68 150 45.3 [37.2; 53.7] Tury, 2020 Q1T1 42 538 7.8 [ 5.7; 10.4]
Rathner, 1994 Q1 18 379 4.7 [ 2.8; 7.4] Tury, 2020 Q1T2 134 969 13.8 [11.7; 16.2]
Sepulveda, 2007 480 2551 18.8 [17.3; 20.4]
Tury, 2020 Q2T1 24 538 4.5 [ 2.9; 6.6] Tury, 2020 Q2T1 24 538 4.5 [ 2.9; 6.6]
Tury, 2020 Q2T2 68 969 7.0 [ 5.5; 8.8] Tury, 2020 Q2T2 68 969 7.0 [ 5.5; 8.8]
Weigel, 2016 32 304 10.5 [ 7.3; 14.5] Yoneda, 2020 41 469 8.7 [ 6.3; 11.7]
Random effects model 6394 16.9 [ 9.6; 28.2] Zhou, 2020 96 130 73.8 [65.4; 81.2]
Heterogeneity: I 2 = 98.77%,  2 = 1.09, p < 0.001
Random effects model 97625 20.8 [17.6; 24.5]
Measure = ORTO-15 Heterogeneity: I 2 = 98.4%,  2 = 0.34, p = 0
Farchakh, 2019 Q1 467 627 74.5 [70.9; 77.9]
Gramaglia, 2019 Q2 246 664 37.0 [33.4; 40.8]
YearCat = 2010-2014
Plichta, 2019 317 1120 28.3 [25.7; 31.0]
Simona Bo, 2014 Q2 114 440 25.9 [21.9; 30.3] Akdevelioglu, 2010 34 577 5.9 [ 4.1; 8.1]
Random effects model 2851 41.1 [22.0; 63.3] Havemann, 2011 14 26 53.8 [33.4; 73.4]
Heterogeneity: I 2 = 99.17%,  2 = 0.85, p < 0.001 Joja, 2012 17 110 15.5 [ 9.3; 23.6]
Measure = QEDD
Kutlu, 2013 13 262 5.0 [ 2.7; 8.3]
Greenleaf, 2009 56 204 27.5 [21.5; 34.1] Mealha, 2013 Q1 8 189 4.2 [ 1.8; 8.2]
Random effects model 204 27.5 [21.8; 34.0] Mealha, 2013 Q2 31 189 16.4 [11.4; 22.5]
Heterogeneity: not applicable Pereira, 2011 48 214 22.4 [17.0; 28.6]
Measure = SD
Reyes-Rodríguez, 2011 36 709 5.1 [ 3.6; 7.0]
Herzog, 1985 18 121 14.9 [ 9.1; 22.5] Roshandel, 2012 86 400 21.5 [17.6; 25.9]
Random effects model 121 14.9 [ 9.6; 22.4] Simona Bo, 2014 Q1 40 440 9.1 [ 6.6; 12.2]
Heterogeneity: not applicable
Simona Bo, 2014 Q2 114 440 25.9 [21.9; 30.3]
Measure = EDDS Random effects model 3556 13.0 [ 8.3; 19.8]
Kara, 2021 228 579 39.4 [35.4; 43.5] Heterogeneity: I 2 = 95.3%,  2 = 0.66, p < 0.001
Koushiou, 2019 Q2 60 340 17.6 [13.7; 22.1]
Random effects model 919 27.3 [11.2; 52.7]
2 2 YearCat = 1990-1994
Heterogeneity: I = 97.77%,  = 0.60, p < 0.001
Grange, 1998 153 1402 10.9 [ 9.3; 12.7]
Measure = WCS Rathner, 1994 Q1 18 379 4.7 [ 2.8; 7.4]
Koushiou, 2019 Q1 102 334 30.5 [25.6; 35.8] Rathner, 1994 Q2 38 379 10.0 [ 7.2; 13.5]
Random effects model 334 30.5 [25.8; 35.7]
Heterogeneity: not applicable
Random effects model 2160 8.4 [ 5.6; 12.6]
Heterogeneity: I 2 = 83.75%,  2 = 0.13, p = 0.002
Measure = DEBQ
Lee, 2015 31 199 15.6 [10.8; 21.4] YearCat = 2005-2009
Random effects model 199 15.6 [11.2; 21.3]
Heterogeneity: not applicable
Greenleaf, 2009 56 204 27.5 [21.5; 34.1]
Jennings, 2006 26 240 10.8 [ 7.2; 15.5]
Measure = ORTO-11 Liao, 2006 T1 15 487 3.1 [ 1.7; 5.0]
Parra-Fernández, 2018 76 454 16.7 [13.4; 20.5] Liao, 2008 T2 17 486 3.5 [ 2.1; 5.5]
Random effects model 454 16.7 [13.6; 20.5]
Heterogeneity: not applicable Mancilla-Diaz, 2007 246 1402 17.5 [15.6; 19.6]
Nichols, 2009 16 383 4.2 [ 2.4; 6.7]
Measure = ANIS Sepulveda, 2007 480 2551 18.8 [17.3; 20.4]
Rathner, 1994 Q2 38 379 10.0 [ 7.2; 13.5] Uzun, 2006 71 414 17.1 [13.6; 21.1]
Tury, 2020 Q1T1 42 538 7.8 [ 5.7; 10.4]
Tury, 2020 Q1T2 134 969 13.8 [11.7; 16.2] Random effects model 6167 10.6 [ 7.3; 15.1]
Random effects model 1886 10.5 [ 7.2; 14.9] Heterogeneity: I 2 = 95.74%,  2 = 0.31, p < 0.001
2 2
Heterogeneity: I = 84.64%,  = 0.11, p = 0.001
YearCat = 1985-1989
Measure = EDS
Rostad, 2021 191 1044 18.3 [16.0; 20.8] Herzog, 1985 18 121 14.9 [ 9.1; 22.5]
Random effects model 1044 18.3 [16.1; 20.8] Marciano, 1988 147 994 14.8 [12.6; 17.1]
Heterogeneity: not applicable Random effects model 1115 14.8 [12.8; 17.0]
Random effects model 145629 19.7 [17.9; 21.6]
Heterogeneity: I 2 = 0%,  2 = 0, p = 0.980
Heterogeneity: I 2 = 98.17%,  2 = 0.34, p = 0
0 20 40 60 80 100 Random effects model 145629 19.7 [17.9; 21.6]
Prevalence (%) Heterogeneity: I 2 = 98.17%,  2 = 0.34, p = 0
0 20 40 60 80 100
Prevalence (%)

Fig. 14  Subgroup meta-analysis by disordered eating measure

Fig. 15  Subgroup meta-analysis by Timeframe/Year

13
3236 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

Fig. 16  Meta-regression
between year and disordered

1
eating in university students

0
Proportion of ED Symptoms

-1
-2
-3

1985 1990 1995 2000 2005 2010 2015 2020

Year

10–12%, or at least 1 in 9, of university undergraduates meet There has been a strong sense among clinicians in some
our criteria for disordered eating. parts of the world (e.g., the USA and Canada) that “the
The statistic of 1 in 9, let alone 1 in 5—or possibly 1 in problem of eating disorders and disordered eating” has been
4, based on the Bayesian estimate—undergraduates scor- worsened by the COVID-19 pandemic. In a recent study of
ing above the at-risk cut-off on various validated measures five cross-sectional samples of French university students,
that screen for eating disorders and represent disordered surveyed at 3-year intervals between 2009 and 2021, Tavo-
eating deserves the attention of researchers, clinicians, lacci et al. (2021) found that the prevalence of screen-based
public health officials, and mental health advocates for two disordered eating for the female and male samples remained
reasons. First the rationale for, and validity data in support fairly stable at 26%-31% and 7–13%, respectively, between
of, the instruments used to measure screen-based (at-risk) 2009 and 2018. However, for both females and males the
disordered eating strongly suggests that there are a mean- prevalence increased dramatically (+ 20% and + 18%,
ingful number of university undergraduates who have an respectively) between 2018 and 2021, perhaps reflecting
eating disorder that is currently undiagnosed and presum- the stressors of the COVID-19 pandemic.
ably untreated. Although this has been pointed out by many Our global data, which incorporates a set of studies of
researchers and clinicians, it remains alarming [32] because French undergraduates by Tavolacci et al. (2015, 2018,
early detection and treatment have been shown to decrease 2020), paints a different picture. We also found evidence
symptoms to a greater extent and improve the chance of of stability—in our study, between 2005 and 2014—but the
recovery [165]. increase we noted was in the period 2015–2022. Moreo-
This aspect of our findings is supported by a very recent ver, we found a slight decrease in the weighted pooled
study that sought to determine whether established dispari- prevalence of screen-based disordered eating from the pre-
ties in ED prevalence and receiving mental healthcare for COVID period of 2015–2019 (23.8%) to the post-COVID
marginalized groups within this population have widened period (2020 onwards; 20.8%). Our data are consistent with
or narrowed over time for different cognitive and behavioral several other recent reviews [167, 168] in suggesting that a
ED symptoms, current probable EDs, lifetime ED diagnoses, COVID-19 effect is accurate in some places and for some
and mental healthcare use among college students across the vulnerable people—and, in particular, people who already
United States [166]. This study found that between 2013 and have an eating disorder—but not in general. Future research
2020 there were non-linear increases in ED symptoms and and meta-analytic reviews are needed to clarify the moderat-
mental health care among young adults in the United States. ing and mediating variables at work here.
Moreover, consistent with the data from this meta-analysis, Over the past 10 years the prevalence and seriousness of
young adults in the United States with higher BMIs had disordered eating and eating disorders in males has received
more ED burden with time, as did bisexuals, homosexual, considerable attention [169]. Nevertheless, our confounder
lesbian, or queer people [166]. analysis indicated that, as the preponderance of females in

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3237

a sample of university undergraduates increased, to a small It is possible that the truncated range of the mean sample
but significant degree so did the prevalence of screen-based ages accounts for this null finding. Further research on the
disordered eating. This finding, in the context of the very relationship between age in undergraduates and postbacca-
significant sex difference seen in, for example, the Tavol- laureate (e.g., graduate, medical, law students) and screen-
acci et al. (2021) samples, is a reminder that, while males based disordered eating is needed. Preliminary epidemio-
certainly have disordered eating and eating disorders, as do logical data from a community sample in Cyprus indicates
those who do not identify as either female or female, there is that a higher percentage of both men (12.3%) and women
still something about the construction, policing, and socio- (23.2%) in the 25–45 age range met or exceeded the EAT-
political status of femininity that places females, including 26 cut-off score of 20 than their counterparts in the 12–18
female undergraduate students, across the globe at greater and 46–60 ranges [178]. Of course, age is embedded in the
risk for screen-based disordered eating, as well as eating transition from older adolescence to emerging adulthood that
disorders [170]. Further research, including meta-analysis, is a foundation for the interest in screen-based disordered
should continue to investigate risk and protective factors that eating in undergraduates, so longitudinal designs beginning
moderate and mediate this disparity. in early or mid-adolescence (see, e.g., Project EAT; Yoon
As noted previously, our confounder analyses also found et al., 2020) are needed.
that, even though the range was narrow, the greater the mean Although our confounder analysis did not find a gen-
BMI of the sample, the higher the prevalence of screen- eral difference in screen-based disordered eating between
based disordered eating. Burnette et al. (2018) reported a Western and non-Western countries, broad sociocultural
similar finding for a sample of U.S. female undergraduate factors are likely to be relevant to developmental phenom-
students, but not male undergraduates [28]. Our finding is ena, so cross-cultural replications will also be necessary.
also consistent with cross-sectional and longitudinal epide- In this regard a cross-sectional study of over 3,200 males
miological studies of U.S. adolescents and emerging adults and females in South Korea, using the same EAT-26 cut-off
showing that greater levels of disordered eating and poor score we did in our meta-analysis, found that the prevalence
psychosocial health among overweight youth [171–173]. In of disordered eating varied only between 6.7% and 7.2%
this regard Yoon et al. (2020) reported that BMI and disor- for age categories 10–12, 13–14, 15–17, and older in 900
dered eating behaviors rose in a correlated fashion across undergraduates ages 18 through 24 [179]. Yet, a previous
15 years of 4 follow-ups of a community sample that was 11 meta-analysis by our team [33] of over 3200 pre-medical
through 18 at baseline [174]. undergraduate students from Brazil, China, India, Malaysia,
In some cultures at least, it is likely that there is a recipro- Pakistan, and the UK found that the prevalence of EAT-26-
cal relationship between disordered eating and body mass based disordered eating in females was moderated by higher
gain which is mediated by internalized weight stigma and BMI and older age, whereas this was not the case for males.
a dieting mentality, and shaped by other sociocultural fac- One strength of our meta-analysis with undergraduates
tors that promote the well-established risk factors of body in general is that nearly two thirds of the individual stud-
dissatisfaction and weight and shape concerns [174]. How- ies used the EAT-26 and SCOFF screening measures. The
ever, further cross-cultural research on the confounders and EAT-26 measure is well validated in a variety of clinical
mediators of the relationship between BMI and screen-based and non-clinical populations from different cultural back-
disordered eating is necessary to test particular path models grounds [180]. Likewise, the SCOFF measure appears to be
that acknowledge cultural variability. For example, a recent a very practical, highly effective screening tool for detecting
survey of Chinese female undergraduates found that the risk for eating disorders [181]. In contrast, we recommend
relationship between body shame and scores on a dietary against further use of the EAT-40 in studies of screen-based
restraint measure was stronger for those with lower BMI disordered eating, due to a very low sensitivity that results
scores [175]. Moreover, a recent meta-analysis of the rela- in a large rate of false-negatives [182], which probably
tionship between disordered eating and use of social net- accounts for the fact that in our meta-analysis the six stud-
working sites found that there was a small positive relation- ies (published between 1998 and 2019) using the EAT-40
ship for university students, but regardless of sample that yielded a prevalence of disordered eating (10.62%), half that
relationship was weaker for those with higher BMIs [176]. of the remaining studies.
We did not find the age was associated with disordered
eating. This is in line with the results of a study of 680 U.S.
female undergraduate students who were screened to elimi- Study strengths and limitations
nate those who would probably qualify to an eating disorder
diagnosis. This study found a trivial, nonsignificant associa- To our knowledge this is the first meta-analysis of the preva-
tion of 0.05 between age and scores on a semi-structured lence of screen-based disordered eating in the population of
interview assessing screen-based disordered eating [177]. university undergraduate students across the world. Other

13
3238 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

strengths are the large sample size (135,454 participants general population, indicating that more research is
from 91 studies) and the categorization of event rates using needed.
cut-off scores from well-validated measures of FEDs. Never-
theless, the findings should be interpreted with consideration
of several limitations. First, the self-report nature of the data What does this study add?
from the studies included may be confounded by shame and/
or social desirability, and diluting the power of anonymity • About 20% of university students exhibit a high level of
and leading to underreporting of ED symptomatology. Sec- screen-based disordered eating behavior and can be clas-
ond, most of the studies considered by this meta-analysis had sified 'at risk' of developing a clinical eating disorder.
a cross-sectional design, so that the direction of the causality • Results of Bayesian analyses confirm the results of a
remains unclear. Third, we exclusively examined English prevalence-based approach. This is the first time Bayes-
Language articles, which may have led to omission of some ian statistics are used to compute odds of disordered eat-
relevant non-English articles. Finally, another limitation is ing.
inherent in one of the strengths: the inevitable heterogeneity • Strong evidence suggests that screen- based disordered
of the numerous studies selected. eating is increasing among university students in recent
years; thus, planning access for preventive interventions
and for supporting those who need outreach and treat-
Conclusion and implications ment is urgently needed.

The pooled prevalence of screen-based disordered eating in

university undergraduate students in 40 countries appears
to be 20%. As a number of studies eliminated people with
Funding The authors have not disclosed any funding.
ongoing eating disorders from their sample, this finding sup-
ports many previous studies indicating that far too many stu- Declaration
dents have an eating disorder and are not accessing accurate
diagnosis and available treatment (see, e.g., Falvey et al., Conflict of interest The authors have not disclosed any competing in-
2021). Moreover, as probably only a 10–15% of that 20% terests.
have a diagnosable eating disorder, our data indicate that
a large percentage of undergraduates are struggling with
disordered eating as a biopsychosocial health problem. It References
is important to develop means of identifying these students
and offering them original or culturally appropriate versions 1. Frank GKW, Shott ME, DeGuzman MC (2019) The neurobiol-
ogy of eating disorders. Child Adolesc Psychiatr Clin N Am
of the Body Project [183] or other effective preventive inter- 28(4):629–640
ventions for high-risk undergraduate students [12]. Finally, 2. Pop-Jordanova N, Zorcec T, Demerdzieva A (2017) anoreXia:
our review indicates that prospective studies, using sensitive anorMaL PHoBia of norMaL weiGHt. Prilozi 38(2):45–53
and specific screening measures such as the EAT-26 and 3. WHO (2007) International classification of diseases and related
health problems, 10th revision. World Health Organization
the SCOFF, are urgently needed to illuminate the interac- 4. APA, Diagnostic and statistical manual of mental disorders:
tions between the risk factors and to use the information to DSM-5. Vol. 5. 2013: American psychiatric association Wash-
construct or improve prevention programs for students with ington, DC
disordered eating attitudes and behaviors. 5. Santomauro DF et al (2021) The hidden burden of eating disor-
ders: an extension of estimates from the Global Burden of Dis-
ease Study 2019. Lancet Psychiatry 8(4):320–328
6. Kutz AM et al (2020) Eating disorder screening: a systematic
review and meta-analysis of diagnostic test characteristics of the
What is already known on this subject? SCOFF. J Gen Intern Med 35(3):885–893
7. Mealha V et al (2013) Students of dietetics & nutrition; a high
• Individuals with eating disorders have disordered eating risk group for eating disorders? Nutr Hosp 28(5):1558–1566
8. Galmiche M et al (2019) Prevalence of eating disorders over the
attitudes and behaviors, but not everyone who engages 2000–2018 period: a systematic literature review. Am J Clin Nutr
in disordered eating has, or will be diagnosed with, an 109(5):1402–1413
eating disorder. The difference is in the frequency and 9. Qian J et al (2022) An update on the prevalence of eating dis-
severity of the behaviors, as well as the amount of dis- orders in the general population: a systematic review and meta-
analysis. Eat Weight Disord 27(2):415–428
tress they cause the person. 10. Levine MP, Smolak L (2006) The prevention of eating problems
• Previous research has reported that university students and eating disorders: Theory, research, and practice. Psychology
have a higher prevalence of disordered eating than the Press

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3239

11. Levine M, Smolak L (2015) The role of protective factors in 31. Fekih-Romdhane F et al (2022) The prevalence of feeding
the prevention of negative body image and disordered eating. and eating disorders symptomology in medical students: an
Eat Disord 24:1–8 updated systematic review, meta-analysis, and meta-regression.
12. Levine MP, Smolak L (2021) The prevention of eating prob- Eat Weight Disord 27(6):1991–2010
lems and eating disorders: Theories, research, and applications, 32. Jahrami H et al (2019) Eating disorders risk among medical
2nd ed. In: The prevention of eating problems and eating disor- students: a global systematic review and meta-analysis. Eat
ders: Theories, research, and applications, 2nd edn. New York, Weight Disord 24(3):397–410
NY, US: Routledge/Taylor & Francis Group. xiii, 459-xiii, 459 33. Jahrami H et al (2019) The relationship between risk of eat-
13. Hoek HW, van Hoeken D (2003) Review of the prevalence and ing disorders, age, gender and body mass index in medical
incidence of eating disorders. Int J Eat Disord 34(4):383–396 students: a meta-regression. Eat Weight Disord 24(2):169–177
14. Alfalahi M et al (2021) Prevalence of eating disorders and 34. Page MJ, The PRISMA et al (2020) Statement: an updated
disordered eating in Western Asia: a systematic review and guideline for reporting systematic reviews. BMJ 2021:372
meta-Analysis. Eat Disord. https://​doi.​org/​10.​1080/​10640​266.​ 35. Stroup DF et al (2000) Meta-analysis of observational
2021.​19694​95 studies in epidemiology: a proposal for reporting. JAMA
15. Bornioli A et al (2019) Adolescent body dissatisfaction and 283(15):2008–2012
disordered eating: predictors of later risky health behaviours. 36. Conn VS et al (2003) Grey literature in meta-analyses. Nurs
Soc Sci Med 238:112458 Res 52(4):256–261
16. Crow S et al (2006) Psychosocial and behavioral correlates of 37. Ata RN, Schaefer LM, Thompson JK (2015) Sociocultural
dieting among overweight and non-overweight adolescents. J theories of eating disorders. In: The Wiley Handbook of Eat-
Adolesc Health 38(5):569–574 ing Disorders, Wiley, West Sussex, UK, pp 269–282. https://​
17. Rush CC, Curry JF, Looney JG (2016) Alcohol expectancies doi.​org/​10.​1002/​97811​18574​089.​ch21
and drinking behaviors among college students with disordered 38. Gordon RA (2000) Eating disorders: anatomy of a social epi-
eating. J Am Coll Health 64(3):195–204 demic, 2nd edn. Blackwell Publishing, Malden, p 281
18. Landi F et al (2016) Anorexia of aging: risk factors, conse- 39. Haddaway N, McGuinness L, Pr itchard C (2021)
quences, and potential treatments. Nutrients 8(2):69 PRISMA2020: R package and ShinyApp for producing
19. Latzer Y et al (2020) Childhood maltreatment in patients with PRISMA 2020 compliant flow diagrams (Version 0.0. 2)
binge eating disorder with and without night eating syndrome 40. van de Schoot R et al (2021) An open source machine learning
vs. control. Psychiatry Res 293:113451 framework for efficient and transparent systematic reviews. Nat
20. Solmi M et al (2021) Age at onset of mental disorders world- Mach Intell 3(2):125–133
wide: large-scale meta-analysis of 192 epidemiological studies. 41. Rathbone J, Hoffmann T, Glasziou P (2015) Faster title and
Mol Psychiatry 27(1):281–295 abstract screening? Evaluating Abstrackr, a semi-automated
21. Stice E, Maxfield J, Wells T (2003) Adverse effects of online screening program for systematic reviewers. Syst Rev
social pressure to be thin on young women: an experimen- 4(1):1–7
tal investigation of the effects of “fat talk.” Int J Eat Disord 42. PLOTCON. WebPlotDigitizer. 2017 [cited July 20, 2021;
34(1):108–117 Available from: https://​autom​eris.​io/​WebPl​otDig​itizer/
22. Aparicio-Martinez P et al (2019) Social media, thin-ideal, body 43. UN. 2022; Available from: https://​www.​un.​org/​dgacm/​en/​conte​
dissatisfaction and disordered eating attitudes: an exploratory nt/​regio​nal-​groups
analysis. Int J Environ Res Public Health 16(21):4177 44. Luchini C et al (2017) Assessing the quality of studies in meta-
23. Smolak L, Levine MP (1996) Adolescent transitions and the analyses: advantages and limitations of the Newcastle Ottawa
development of eating problems. In: The developmental psy- Scale. World J Meta-Anal 5:80–84
chopathology of eating disorders: Implications for research, 45. Shuster JJ (2010) Empirical vs natural weighting in random
prevention, and treatment. Lawrence Erlbaum Associates, Inc. effects meta-analysis. Stat Med 29(12):1259–1265
pp 207–233 46. Borenstein M et al (2010) A basic introduction to fixed-effect
24. Mahoney CR et al (2019) Intake of caffeine from all sources and and random-effects models for meta-analysis. Res Synth Meth-
reasons for use by college students. Clin Nutr 38(2):668–675 ods 1(2):97–111
25. Jahrami H et al (2020) Intake of caffeine and its association with 47. Jackson D, White IR, Thompson SG (2010) Extending Der-
physical and mental health status among university students in Simonian and Laird’s methodology to perform multivariate
Bahrain. Foods 9(4):473 random effects meta-analyses. Stat Med 29(12):1282–1297
26. Narayanan A et al (2021) Students’ use of caffeine, alcohol, die- 48. Lewis S, Clarke M (2001) Forest plots: trying to see the wood
tary supplements, and illegal substances for improving academic and the trees. BMJ 322(7300):1479–1480
performance in a New Zealand university. Health Psychol Behav 49. Rücker G, Schwarzer G (2021) Beyond the forest plot: the
Med 9(1):917–932 drapery plot. Res Synth Methods 12(1):13–19
27. Nagata JM et al (2021) Associations between ergogenic supple- 50. van Aert RC, Wicherts JM, van Assen MA (2016) Conducting
ment use and eating behaviors among university students. Eat meta-analyses based on p values: reservations and recommen-
Disord 29(6):599–615 dations for applying p-uniform and p-curve. Perspect Psychol
28. Burnette CB, Simpson CC, Mazzeo SE (2018) Relation of BMI Sci 11(5):713–729
and weight suppression to eating pathology in undergraduates. 51. Sutton AJ, Abrams KR (2001) Bayesian methods in meta-
Eat Behav 30:16–21 analysis and evidence synthesis. Stat Methods Med Res
29. Makino M, Tsuboi K, Dennerstein L (2004) Prevalence of eating 10(4):277–303
disorders: a comparison of Western and non-Western countries. 52. Higgins JP, Thompson SG (2002) Quantifying heterogeneity
MedGenMed 6(3):49 in a meta-analysis. Stat Med 21(11):1539–1558
30. Farchakh Y, Hallit S, Soufia M (2019) Association between 53. NAND. Eating Disorder Statistics. 2022; Available from:
orthorexia nervosa, eating attitudes and anxiety among medi- https://​anad.​org/​eating-​disor​ders-​stati​stics/
cal students in Lebanese universities: results of a cross-sectional 54. van Eeden AE, van Hoeken D, Hoek HW (2021) Incidence,
study. Eat Weight Disord 24(4):683–691 prevalence and mortality of anorexia nervosa and bulimia ner-
vosa. Curr Opin Psychiatry 34(6):515–524

13
3240 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

55. Spiegelhalter DJ (2004) Incorporating Bayesian ideas into 81. Albrahim T et al (2019) The spectrum of disordered eating atti-
health-care evaluation. Stat Sci 19(1):156–174 tudes among female university students: a cross-sectional study.
56. Lunn D et al (2013) Fully Bayesian hierarchical modelling in two Curr Res Nutr Food Sci J 7(3):698–707
stages, with application to meta-analysis. J Roy Stat Soc: Ser C 82. Alcaraz-Ibáñez M, Sicilia A, Paterna A (2019) Exploring the
(Appl Stat) 62(4):551–572 differentiated relationship between appearance and fitness-related
57. Bolstad WM, Curran JM (2016) Introduction to Bayesian statis- social anxiety and the risk of eating disorders and depression in
tics. John Wiley & Sons young adults. Scand J Psychol 60(6):569–576
58. Huedo-Medina TB et al (2006) Assessing heterogeneity in meta- 83. Alhazmi AH, Al Johani A (2019) Prevalence and associated
analysis: Q statistic or ­I2 index? Psychol Methods 11(2):193 factors of eating disorders among students in Taiba University,
59. Lin L, Chu H, Hodges JS (2017) Alternative measures of Saudi Arabia: a cross-sectional study. Malaysian J Public Health
between-study heterogeneity in meta-analysis: reducing the Med 19(1):172–176
impact of outlying studies. Biometrics 73(1):156–166 84. Alkazemi D et al (2018) Distorted weight perception correlates
60. Galbraith RF (1994) Some applications of radial plots. J Am Stat with disordered eating attitudes in Kuwaiti college women. Int J
Assoc 89(428):1232–1242 Eat Disord 51(5):449–458
61. Galbraith R (1988) Graphical display of estimates having differ- 85. AlShebali M, AlHadi A, Waller G (2021) The impact of ongoing
ing standard errors. Technometrics 30(3):271–281 westernization on eating disorders and body image dissatisfac-
62. Galbraith R (1988) A note on graphical presentation of estimated tion in a sample of undergraduate Saudi women. Eat Weight
odds ratios from several clinical trials. Stat Med 7(8):889–894 Disord 26(6):1835–1844
63. Viechtbauer W, Cheung MW (2010) Outlier and influence diag- 86. Alwosaifer AM et al (2018) Eating disorders and associated risk
nostics for meta-analysis. Res Synth Methods 1(2):112–125 factors among Imam Abdulrahman bin Faisal university prepara-
64. Patsopoulos NA, Evangelou E, Ioannidis JP (2008) Sensitivity of tory year female students in Kingdom of Saudi Arabia. Saudi
between-study heterogeneity in meta-analysis: proposed metrics Med J 39(9):910
and empirical evaluation. Int J Epidemiol 37(5):1148–1157 87. Azzouzi N et al (2019) Eating disorders among Moroccan medi-
65. Mathur MB, VanderWeele TJ (2020) Sensitivity analysis for pub- cal students: cognition and behavior. Psychol Res Behav Manag
lication bias in meta-analyses. J Roy Stat Soc: Ser C (Appl Stat) 12:129
69(5):1091–1119 88. Badrasawi MM, Zidan SJ (2019) Binge eating symptoms preva-
66. Egger M et al (1997) Bias in meta-analysis detected by a simple, lence and relationship with psychosocial factors among female
graphical test. BMJ 315(7109):629–634 undergraduate students at Palestine Polytechnic University: a
67. Duval S, Tweedie R (2000) A nonparametric, “Trim and Fill” cross-sectional study. J Eat Disord 7(1):1–8
method of accounting for publication bias in meta-analysis. J Am 89. Barayan S, Al-Yousif Z, Sabra A (2018) Prevalence of eating
Stat Assoc 95(449):89–98 disorders among female university medical students in Dammam,
68. Peters JL et al (2010) Assessing publication bias in meta-analyses Saudi Arabia. Int J Med Health Sci 7(3):107–112
in the presence of between-study heterogeneity. J R Stat Soc 90. Barry MR, Sonneville KR, Leung CW (2021) Students with food
173(3):575–591 insecurity are more likely to screen positive for an eating disorder
69. Sedgwick P (2013) Meta-analyses: heterogeneity and subgroup at a large, public university in the midwest. J Acad Nutr Diet
analysis. BMJ 346:f4040 121(6):1115–1124
70. Thompson SG, Higgins JP (2002) How should meta-regres- 91. Benítez A et al (2018) Analysis of the risk of suffering eating
sion analyses be undertaken and interpreted? Stat Med disorders in young student university of extremadura (Spain).
21(11):1559–1573 Enfermería Global 54:124–133
71. Open Science Collaboration (2015) Estimating the reproducibil- 92. Bizri M et al (2020) Prevalence of eating disorders among medi-
ity of psychological science. Science 349(6251):aac4716 cal students in a Lebanese medical school: a cross-sectional
72. R. 4.1.1. R Foundation for Statistical Computing, Vienna, Aus- study. Neuropsychiatr Dis Treat 16:1879
tria. 2021 2020 05/05/2020]; Available from: https://​www.R-​ 93. Bo S et al (2014) University courses, eating problems and muscle
proje​ct.​org/ dysmorphia: are there any associations? J Transl Med 12(1):1–8
73. Schwarzer G, Schwarzer MG (2012) Package ‘meta’. The R foun- 94. Bosi M et al (2016) Body image and eating behavior among
dation for statistical computing, p 9 medical students: eating disorders among medical students. Epi-
74. Viechtbauer W (2010) Conducting meta-analyses in R with the demiology (Sunnyvale) 6(4):256
metafor package. J Stat Softw 36(3):1–48 95. Brumboiu MI et al (2018) Nutritional status and eating disorders
75. Roever C, Friede T, Roever MC (2017) Package ‘bayesmeta’ among medical students from the Cluj-Napoca University centre.
76. McGuinness LA, Higgins JP (2021) Risk-of-bias VISualization Clujul Medical 91(4):414
(robvis): an R package and Shiny web app for visualizing risk- 96. Carriedo AP et al (2020) When clean eating isn’t as faultless:
of-bias assessments. Res Synth Methods 12(1):55–61 the dangerous obsession with healthy eating and the relation-
77. Abdul Manaf N, Saravanan C, Zuhrah B (2016) The prevalence ship between Orthorexia nervosa and eating disorders in Mexican
and inter-relationship of negative body image perception, depres- University students. J Eat Disord 8(1):1–12
sion and susceptibility to eating disorders among female medical 97. Castejón MÁ, Berengüí R (2020) Personality differences and
undergraduate students. J Clin Diagn Res 10(3):01 psychological variables related to the risk of eating disorders.
78. Abo Ali EA, Shehata WM (2020) Eating disorder risk among Anales de Psicología 36(1):64
medical students at Tanta University, Egypt. Egyptian J Commun 98. Chammas R et al (2017) Eating disorders among university stu-
Med 38(4):17 dents in a Middle Eastern urban setting: who is at risk? Clin Nutr
79. Akdevelioglu Y, Gümüs H (2010) Eating disorders and body 36:S113–S114
image perception among university students. Pak J Nutr 99. Chan YL et al (2020) Eating disorder among Malaysian Uni-
9(12):1187–1191 versity students and its associated factors. Asia Pacific J Public
80. Al Banna MH et al (2021) Prevalence and determinants of eating Health 32(6–7):334–339
disorder risk among Bangladeshi public university students: a 100. Chaudhari B et al (2017) The relationship of eating disorders
cross-sectional study. Health Psychol Res 9(1):24837 risk with body mass index, body image and self-esteem among
medical students. Ann Med Health Sci Res 7(3):144–149

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3241

101. Christensen KA et al (2021) Food insecurity associated with ele- 122. Le Grange D, Telch CF, Tibbs J (1998) Eating attitudes and
vated eating disorder symptoms, impairment, and eating disorder behaviors in 1,435 South African Caucasian and non-Caucasian
diagnoses in an American University student sample before and college students. Am J Psychiatry 155(2):250–254
during the beginning of the COVID-19 pandemic. Int J Eating 123. Lee SJ, Cloninger CR, Chae H (2015) Cloninger’s temperament
Disord 54(7):1213–1223 and character traits in medical students of Korea with problem
102. Compte EJ, Sepulveda AR, Torrente F (2015) A two-stage epi- eating behaviors. Compr Psychiatry 59:98–106
demiological study of eating disorders and muscle dysmorphia 124. Liao Y et al (2013) Changes in eating attitudes, eating disor-
in male university students in Buenos Aires. Int J Eat Disord ders and body weight in Chinese medical university students.
48(8):1092–1101 Int J Soc Psychiatry 59(6):578–585
103. Damiri B et al (2021) Eating disorders and the use of cognitive 125. Mancilla-Diaz JM et al (2007) A two-stage epidemiologic
enhancers and psychostimulants among university students: a study on prevalence of eating disorders in female university
cross-sectional study. Neuropsychiatr Dis Treat 17:1633 students in Mexico. Eur Eating Disord Rev 15(6):463–470
104. Din ZU et al (2019) Tendency towards eating disorders and asso- 126. Marciano D, McSherry J, Kraus A (1988) Abnormal eating
ciated sex-specific risk factors among university students. Arch attitudes: prevalance at a Canadian University. Can Fam Physi-
Neuropsychiatry 56(4):258 cian 34:75
105. Ebrahim M et al (2019) Disordered eating attitudes correlate with 127. Mazzaia MC, Santos RMC (2018) Risk factors for eating dis-
body dissatisfaction among Kuwaiti male college students. J Eat orders among undergraduate nursing students. Acta Paulista
Disord 7(1):1–13 de Enfermagem 31:456–462
106. Erol Ö, Özer A (2019) Determination of orthorexia nervosa 128. Momeni M, Ghorbani A, Arjeini Z (2020) Disordered eat-
symptoms and eating attitudes in medicine students. Eur J Public ing attitudes among Iranian university students of medical
Health 29(Suppl 4):280 sciences: the role of body image perception. Nutr Health
107. Falvey SE et al (2021) Diagnosis of eating disorders among col- 26(2):127–133
lege students: a comparison of military and civilian students. Mil 129. Ngan SW et al (2017) The relationship between eating disorders
Med 186:975–983 and stress among medical undergraduate: a cross-sectional study.
108. Fatima W, Fatima R, Anwar NS (2018) Prevalence of eating Open J Epidemiol 7(02):85
disorders among female college students of Northern Broader 130. Nichols S et al (2009) Body image perception and the risk of
University, Arar, Kingdom of Saudi Arabia. Int J Child Health unhealthy behaviours among university students. West Indian
Nutr 7(3):115–121 Med J 58(5):465–471
109. Gramaglia C et al (2019) Orthorexia nervosa, eating patterns and 131. Padmanabhan V, Rahman MM, Fatima S (2018) Prevalence of
personality traits: a cross-cultural comparison of Italian, Polish eating disorders in dental students of a Young University in UAE.
and Spanish university students. BMC Psychiatry 19(1):1–11 Int J Sci Res 7(4):432–434
110. Greenleaf C et al (2009) Female collegiate athletes: prevalence 132. Parra-Fernández ML et al (2019) Adaptation and validation of
of eating disorders and disordered eating behaviors. J Am Coll the Spanish version of the DOS questionnaire for the detection
Health 57(5):489–496 of orthorexic nervosa behavior. PLoS ONE 14(5):e0216583
111. Havemann L et al (2011) Disordered eating and menstrual pat- 133. Parreño-Madrigal IM et al (2020) Prevalence of risk of eating
terns in female university netball players. South African J Sports disorders and its association with obesity and fitness. Int J Sports
Med 23(3):68–72 Med 41(10):669–676
112. Herzog DB et al (1985) Eating disorders and social malad- 134. Pereira LdNG et al (2011) Eating disorders among health science
justment in female medical students. J Nervous Mental Dis students at a university in southern Brazil. Revista de Psiquiatria
173(12):734–737 do Rio Grande do Sul 33:14–19
113. Iyer S, Shriraam V (2021) Prevalence of eating disorders and its 135. Pitanupong J, Jatchavala C (2017) Atypical eating attitudes and
associated risk factors in students of a Medical College Hospital behaviors in Thai medical students. Siriraj Med J 69(1):5–10
in South India. Cureus 13(1):e12926 136. Plichta M, Jezewska-Zychowicz M (2019) Eating behaviors,
114. Jamali YA et al (2020) Prevalence of eating disorders among attitudes toward health and eating, and symptoms of orthorexia
students of Quest University, Nawabshah, Pakistan. Merit Res J nervosa among students. Appetite 137:114–123
Med Med Sci 8(4):81–84 137. Polanco PP, Zetina LMM. Study of the risk of having eating
115. Jennings PS et al (2006) Acculturation and eating disorders in disorder habits in undergraduate and graduate students in Mexi-
Asian and Caucasian Australian university students. Eat Behav can institutions. ESPACIO I+ D, INNOVACIÓN MÁS DESAR-
7(3):214–219 ROLLO: p 54
116. Joja O, von Wietersheim J (2012) A cross-cultural comparison 138. Radwan H et al (2018) Eating disorders and body image concerns
between EDI results of Romanian and German students. Procedia as influenced by family and media among university students in
Soc Behav Sci 33:1037–1041 Sharjah, UAE. Asia Pac J Clin Nutr 27(3):695–700
117. Kiss-Tóth E et al (2018) Eating disorder in university students: 139. Ramaiah RR (2015) Eating disorders among medical students
an international multi-institutional study. Eur J Public Health. of a rural teaching hospital: a cross-sectional study. Int J Com-
https://​doi.​org/​10.​1093/​eurpub/​cky214.​010 munity Med Public Health 2(1):25–28
118. Ko N et al (2015) Disordered eating behaviors in university stu- 140. Rasman NSB et al (2018) Prevalence of eating disorders among
dents in Hanoi, Vietnam. J Eat Disord 3(1):1–7 medical students in Ipoh. Perak, Malaysia
119. Koushiou M, Nikolaou P, Karekla M (2020) Prevalence and cor- 141. Rathner G, Rumpold G (1994) Convergent validity of the eating
relates of eating disorders in greek-cypriot adolescents and young disorder inventory and the anorexia nervosa inventory for self-
adults. Eur J Counsel Psychol 8(1):3–18 rating in an Austrian nonclinical population. Int J Eat Disord
120. Kutlu R, Civi S (2013) Evaluation of eating habits, body percep- 16(4):381–393
tion and depression status of university students. Gülhane Tip 142. Reyes-Rodríguez ML et al (2011) A description of disordered eat-
Dergisi 55(3):196 ing behaviors in Latino males. J Am Coll Health 59(4):266–272
121. Ladner J et al (2019) Eating disorders among university stu- 143. Roshandel A, Safavi M, Ghasemi I (2012) Prevalence of eating
dents: a public health challenge. An European study. Eur J Public disorders among female students of university (Tehran-Iran). Life
Health. https://​doi.​org/​10.​1093/​eurpub/​ckz186.​486 Sci J 9(4):2822–2828

13
3242 Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243

144. Rostad IS, Tyssen R, Løvseth LT (2021) Symptoms of disturbed 164. Wu XY et al (2019) The association between disordered eating
eating behavior risk: Gender and study factors in a cross-sec- and health-related quality of life among children and adolescents:
tional study of two Norwegian medical schools. Eating Behav a systematic review of population-based studies. PLoS ONE
43:101565 14(10):e0222777
145. Safer M et al (2020) Eating disorders: prevalence and asso- 165. Treasure J, Russell G (2011) The case for early intervention in
ciated factors among health occupation students in Monastir anorexia nervosa: theoretical exploration of maintaining fac-
University (Tunisia). Tunis Med 98(12):895–912 tors. Br J Psychiatry 199(1):5–7
146. Saleh RN et al (2018) Disordered eating attitudes in female 166. Romano KA et al (2022) Changes in the prevalence and soci-
students of An-Najah National University: a cross-sectional odemographic correlates of eating disorder symptoms from
study. J Eat Disord 6(1):1–6 2013 to 2020 among a large national sample of U.S. young
147. Sepulveda A et al (2007) Prevention program for disturbed eat- adults: a repeated cross-sectional study. Int J Eat Disord
ing and body dissatisfaction in a Spanish university population: 55(6):776–789
a pilot study. Body Image 4(3):317–328 167. Devoe J, D., et al (2022) The impact of the COVID-19 pan-
148. Sharifian MJ et al (2021) Association between dental fear and demic on eating disorders: a systematic review. Int J Eating
eating disorders and Body Mass Index among Finnish univer- Disord. https://​doi.​org/​10.​1002/​eat.​23704
sity students: a national survey. BMC Oral Health 21(1):1–9 168. Schneider J et al (2022) A mixed-studies systematic review of
149. Sharma M et al (2019) Body image perception, eating attitude the experiences of body image, disordered eating, and eating
and influence of media among undergraduate students of medi- disorders during the COVID-19 pandemic. Int J Eat Disord.
cal college in Delhi: a cross sectional study. Int J Res Med Sci https://​doi.​org/​10.​1002/​eat.​23706
7(12):4627 169. Nagata JM et al (2021) Eating disorders in boys and men.
150. Shashank K, Gowda P, Chethan T (2016) A cross-sectional Springer
study to asses the eating disorder among female medical stu- 170. Murnen SK, Smolak L (2015) Gender and eating disorders.
dents in a rural medical college of Karnataka State. Natl J In: The Wiley handbook of eating disorders, Wiley, Uk, pp
Community Med 7:524–527 352–366. https://​doi.​org/​10.​1002/​97811​18574​089.​ch27
151. Spillebout A et al (2019) Mental health among university stu- 171. Eaton DK et al (2012) Youth risk behavior surveillance—
dents with eating disorders and irritable bowel syndrome in United States, 2011. Morb Mortal Wkly Rep Recomm Rep
France. Rev Epidemiol Sante Publique 67(5):295–301 61(4):1–162
152. Taha AAAE-A, Abu-Zaid HA, Desouky DE-S (2018) Eating 172. Neumark-Sztainer D et al (2002) Weight-related concerns and
disorders among female students of Taif University, Saudi Ara- behaviors among overweight and nonoverweight adolescents:
bia. Arch Iranian Med 21(3):111–117 implications for preventing weight-related disorders. Arch Pedi-
153. Tavolacci MP et al (2015) Eating disorders and associated atr Adolesc Med 156(2):171–178
health risks among university students. J Nutr Educ Behav 173. Neumark-Sztainer DR et al (2007) Shared risk and protective
47(5):412–420 factors for overweight and disordered eating in adolescents. Am
154. Tavolacci M et al (2018) Changes and specificities in health J Preventive Med 33(5):359–369
behaviors among healthcare students over an 8-year period. 174. Yoon C et al (2020) Disordered eating behaviors and 15-year
PLoS ONE 13(3):e0194188 trajectories in body mass index: findings from project eating
155. Tavolacci M-P, Déchelotte P, Ladner J (2020) Eating disorders and activity in teens and young adults (EAT). J Adolesc Health
among college students in France: characteristics, help-and 66(2):181–188
care-seeking. Int J Environ Res Public Health 17(16):5914 175. Yao L, Niu G, Sun X (2021) Body image comparisons on social
156. Thangaraju SI et al (2020) A cross-sectional study on preva- networking sites and chinese female college students’ restrained
lence of eating disorder and body image disturbance among eating: the roles of body shame, body appreciation, and body
female undergraduate medical students. J Ment Health Hum mass index. Sex Roles 84:465–476
Behav 25(1):53 176. Zhang J et al (2021) The relationship between SNS usage and
157. Túry F et al (2021) Eating disorder characteristics among Hun- disordered eating behaviors: a meta-analysis. Front Psychol
garian medical students: Changes between 1989 and 2011. J 12:641919
Behav Addict 9(4):1079–1087 177. Rohde P et al (2017) Age effects in eating disorder baseline
158. Uriegas NA et al (2021) Examination of eating disorder risk risk factors and prevention intervention effects. Int J Eat Disord
among university marching band artists. J Eat Disord 9(1):1–10 50(11):1273–1280
159. Uzun Ö et al (2006) Screening disordered eating attitudes and 178. Hadjigeorgiou C et al (2018) Disordered eating in three different
eating disorders in a sample of Turkish female college students. age groups in Cyprus: a comparative cross-sectional study. Public
Compr Psychiatry 47(2):123–126 Health 162:104–110
160. Weigel A et al (2016) Eating pathology in medical students 179. Hong S-C et al (2015) Prevalence of distorted body image in
in Eastern Germany: comparison with general population and young Koreans and its association with age, sex, body weight
a sample at the time of the German reunification. Eat Weight status, and disordered eating behaviors. Neuropsychiatr Dis Treat
Disord 21(3):445–451 11:1043
161. Yoneda R et al (2020) Reliability and validity of the Japanese 180. Garfinkel PE, Newman A (2001) The eating attitudes test:
translation of the Eating Disorders Quality of Life (ED-QOL) twenty-five years later. Eat Weight Disord 6(1):1–24
scale for Japanese healthy female university undergraduate 181. Morgan JF, Reid F, Lacey JH (1999) The SCOFF questionnaire:
students and patients with eating disorders. BioPsychoSocial assessment of a new screening tool for eating disorders. BMJ
Med 14(1):1–7 319(7223):1467–1468
162. Yu J et al (2015) Prevalence of disordered eating attitudes 182. Vetrone G et al (2006) Screening for eating disorders: false nega-
among university students in Wuhu, China. Nutr Hosp tives and eating disorders not otherwise specified. Eur J Psychia-
32(4):1752–1757 try 20(1):13–20
163. Zhou Y, Pennesi JL, Wade TD (2020) Online imagery rescripting 183. Becker CB, Stice E (2017) From efficacy to effectiveness to broad
among young women at risk of developing an eating disorder: a implementation: Evolution of the Body Project. J Consult Clin
randomized controlled trial. Int J Eat Disord 53(12):1906–1917 Psychol 85(8):767–782

13
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity (2022) 27:3215–3243 3243

184. Fichter MM et al (1988) Anorexia nervosa in Greek and Turkish 196. Stice E, Telch CF, Rizvi SL (2000) Development and valida-
adolescents. Eur Arch Psychiatry Neurol Sci 237(4):200–208 tion of the Eating Disorder Diagnostic Scale: a brief self-report
185. Giuffrida JM (2020) Improved Screening for Binge Eating measure of anorexia, bulimia, and binge-eating disorder. Psychol
Disorder in Primary Care. The University of North Carolina at Assess 12(2):123–131
Charlotte 197. Fairburn CG, Beglin SJ (2008) Eating disorder examination ques-
186. Herman BK et al (2016) Development of the 7-Item Binge-Eating tionnaire. Cognitive behavior therapy and eating disorders, pp
Disorder Screener (BEDS-7). Prim Care Companion CNS Dis- 309–313
ord. https://​doi.​org/​10.​4088/​PCC.​15m01​896 198. Garner DM, Olmstead MP, Polivy J (1983) Development and
187. Van Strien T et al (1986) The Dutch Eating Behavior Question- validation of a multidimensional eating disorder inventory for
naire (DEBQ) for assessment of restrained, emotional, and exter- anorexia nervosa and bulimia. Int J Eat Disord 2(2):15–34
nal eating behavior. Int J Eat Disord 5(2):295–315 199. Garner DM et al (1982) The eating attitudes test: psychometric
188. Mond JM et al (2006) Eating Disorder Examination Question- features and clinical correlates. Psychol Med 12(4):871–878
naire (EDE-Q): norms for young adult women. Behav Res Ther 200. Rosenvinge JH et al (2001) A new instrument measuring dis-
44(1):53–62 turbed eating patterns in community populations: development
189. Mond JM et al (2004) Validity of the Eating Disorder Examina- and initial validation of a five-item scale (EDS-5). Eur Eating
tion Questionnaire (EDE-Q) in screening for eating disorders in Disord Rev 9(2):123–132
community samples. Behav Res Ther 42(5):551–567 201. Donini LM et al (2004) Orthorexia nervosa: a preliminary study
190. Parra-Fernandez ML et al (2018) Adaptation and validation of with a proposal for diagnosis and an attempt to measure the
the Spanish version of the ORTO-15 questionnaire for the diag- dimension of the phenomenon. Eat Weight Disord 9(2):151–157
nosis of orthorexia nervosa. PLoS ONE 13(1):e0190722 202. Mintz LB et al (1997) Questionnaire for eating disorder diagno-
191. Roncero M, Barrada JR, Perpiñá C (2017) Measuring orthorexia ses: reliability and validity of operationalizing DSM—IV criteria
nervosa: psychometric limitations of the ORTO-15. Span J Psy- into a self-report format. J Couns Psychol 44(1):63
chol 20:E41 203. Morgan JF, Reid F, Lacey JH (2000) The SCOFF questionnaire.
192. Davison KK, Markey CN, Birch LL (2003) A longitudinal exami- Western J Med 172(3):164
nation of patterns in girls’ weight concerns and body dissatisfac- 204. Killen JD et al (1994) Pursuit of thinness and onset of eating
tion from ages 5 to 9 years. Int J Eat Disord 33(3):320–332 disorder symptoms in a community sample of adolescent girls:
193. Silva WRD et al (2017) Body weight concerns: cross-national a three-year prospective analysis. Int J Eat Disord 16(3):227–238
study and identification of factors related to eating disorders.
PLoS ONE 12(7):e0180125 Publisher's Note Springer Nature remains neutral with regard to
194. Fichter MM, Keeser W (1980) Das Anorexia-nervosa-Inven- jurisdictional claims in published maps and institutional affiliations.
tar zur Selbstbeurteilung (ANIS). Arch Psychiatr Nervenkr
228(1):67–89
195. Garner DM, Garfinkel PE (1979) The Eating Attitudes Test:
an index of the symptoms of anorexia nervosa. Psychol Med
9(2):273–279

Authors and Affiliations

Omar A. Alhaj1 · Feten Fekih‑Romdhane2,3 · Dima H. Sweidan1 · Zahra Saif4 · Mina F. Khudhair5 ·
Hadeel Ghazzawi6 · Mohammed Sh. Nadar7 · Saad S. Alhajeri8 · Michael P. Levine9 · Haitham Jahrami4,5
1
Omar A. Alhaj Department of Nutrition, Faculty of Pharmacy and Medical
[email protected] Sciences, University of Petra, Amman, Jordan
2
Feten Fekih‑Romdhane Faculty of Medicine of Tunis, Tunis El Manar University,
[email protected] Tunis, Tunisia
3
Dima H. Sweidan The Tunisian Center of Early Intervention Is Psychiatry,
[email protected] Department of Psychiatry “Ibn Omrane”, Razi Hospital, rue
des orangers, Manouba, Tunisia
Zahra Saif
4
[email protected] Ministry of Health, Manama, Bahrain
5
Mina F. Khudhair Department of Psychiatry, College of Medicine and Medical
[email protected] Sciences, Arabian Gulf University, Manama, Bahrain
6
Hadeel Ghazzawi Nutrition and Food Science Department, Agriculture School,
[email protected] The University of Jordan, P.O.Box 11942, Amman, Jordan
7
Mohammed Sh. Nadar Occasional Therapy Department, Faculty of Allied Health
[email protected] Sciences, Kuwait University, Kuwait City, Kuwait
8
Saad S. Alhajeri Ministry of Health, Kuwait City, Kuwait
[email protected] 9
Emeritus Professor, Department of Psychology, Kenyon
Michael P. Levine College, Gambier, OH 43022, USA
[email protected]

13