Frequency of disordered eating habits among fashion models (2024)

Journal List
Wiley Open Access Collection
PMC9790310

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

Eur Eat Disord Rev. 2022 Nov; 30(6): 823–829.

Published online 2022 Apr 26. doi:10.1002/erv.2912

PMCID: PMC9790310

PMID: 35474384

Nikolett Bogár,¹ Szilvia Dukay‐Szabó,¹ Dávid Simon,² Ferenc Túry,¹ and Bea Pászthy³

Author information Article notes Copyright and License information PMC Disclaimer

Associated Data

Data Availability Statement

Abstract

Objective

Sociocultural influences, including an increasing pressure for fashion models to maintain a thin body frame may be crucial in the development of eating disorders. The present study aimed to establish whether fashion models are more likely than non‐models to develop eating disorders.

Methods

Female fashion models were selected by snowball sampling (n=179, mean age: 25.9 SD=4.70years). They were compared with an age adjusted control group (n=261, mean age: 25.0 SD=4.97years). Participants completed an online questionnaire containing the Eating Disorder Inventory.

Results

The average BMI of the fashion models was in the underweight range (mean BMI=18.1 SD=1.68). The BMI of the control group was significantly higher (mean=22.1 SD=4.23, p<0.001). The frequency of simulated anorexia nervosa was 3.9% among the fashion models and 1.1% in the control group (p=0.057). 14.6% of the models showed subclinical anorexia nervosa symptoms versus 2.7% in the control group (p<0.001). The ratio of bulimia nervosa and subclinical bulimia nervosa showed no significant difference between the two groups.

Conclusion

Female fashion models showed no significant difference from the control group in the frequency of anorexia nervosa and bulimia nervosa but had a significantly higher frequency of the subclinical form of anorexia nervosa.

Keywords: anorexia nervosa, bulimia nervosa, fashion, models, thin beauty ideal

Highlights

There is an intensive pressure for fashion models to maintain a thin body frame, increasing the risk of eating disorders.
The frequency of full‐blown anorexia nervosa and bulimia is slightly, but not significantly higher among fashion models as compared to control subjects.
Subclinical anorexia nervosa is significantly more frequent among fashion models than in the control group.

Abbreviations

AN: anorexia nervosa
BMI: Body Mass Index
BN: bulimia nervosa
CI: confidence interval
DSM‐5: Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition
EBSS: Eating Behaviour Severity Scale
ED: eating disorder
EDI: Eating Disorder Inventory
EDI‐DT: Eating Disorder Inventory Drive for Thinness subscale

1. INTRODUCTION AND AIMS

Eating disorders (EDs) are multifactorial psychiatric disorders. Besides biological and environmental factors, background sociocultural influences play an important role in their development. Members of certain professions that require low body weight for aesthetic or performance purposes (such as ballet dancers, flight attendants, jockeys) are at a higher risk for the development of EDs (Abraham,¹⁹⁹⁶). Fashion models face strong pressure to maintain a thin body frame and to meet certain measurement values, potentially causing the development of ED symptoms (Bogár & Túry,²⁰¹⁹).

The average BMI of female professional fashion models is well under the lower healthy limit (18.5), and around two thirds of fashion models report that they intentionally use weight controlling methods to get in shape (Rodgers etal.,²⁰¹⁷). This can lead to the development of partial ED symptoms in otherwise healthy individuals. Models typically come under pressure from model agents or fashion designers (Bogár & Túry,²⁰¹⁹). Among a group of models in one study, 62% had been advised by industry professionals to lose weight or to change their body size or shape if they wanted to be moresuccessful in their profession. This external pressure ledto unhealthy weight controlling behaviours suchas dieting, exercising, or skipping meals (Rodgers etal.,²⁰¹⁷). The highly competitive nature of the industry (Johnson,²⁰¹¹) pressures models to conform to these shape controlling demands, due to the availability of new young fashion models, especially from East‐European countries, who can easily replace almost any of them (Scully,²⁰¹⁶).

One model, Victoire Dauxerre, writes in her memoirs that models are required to maintain an almost asexual physique and usually appear clone‐like on the runway. Fashion Week is the greatest challenge for aspiring models, demanding an almost skeletal aesthetic: hips should not exceed 90cm in circumference (Dauxerre,²⁰¹⁷). Models not meeting the criteria often have to endure body shaming by their agents, which can be regarded as a form of psychological abuse (Bogár & Túry,²⁰¹⁷).

Data on the frequency of EDs among fashion models is scarce, and sample sizes have been small in all thepublished studies. A multicultural study assessing ED risk factors in a large group of fashion models using ED specific questionnaires would thus be of great importance. There have been several studies proving the negative effects that the media's portrayal of the beauty ideal has on the general population (Derenne & Beresin,²⁰⁰⁶; Park,²⁰⁰⁵), but the effects of the fashion industry's beauty ideal on its own employees, and how that has changed over the years, have not yet been investigated in sufficient depth.

Previous studies suggest that there is little difference between fashion models and non‐models as regards clinically significant EDs (Zancu & Enea,²⁰¹⁷). However, models have significantly lower BMI, putting them into the underweight category, and they are taller than the general population (Brenner & Cunningham,¹⁹⁹²). Despite being slightly underweight, models experience a stronger drive for thinness than non‐models (Swami & Szmigielska,²⁰¹³). Of more concern is that models display subclinical ED symptoms more frequently than the average population (Preti etal.,²⁰⁰⁸). It is important to keep in mind that some models are genetically more slender than the calculated healthy weight for their height, resulting in a lower BMI without necessarily involving unhealthy eating behaviours (Brenner & Cunningham,¹⁹⁹²).

The present study aims to provide data on the frequency of EDs among female fashion models compared to a control group and to reveal the risk factors for partial ED symptoms. It may be the most thorough multicultural empirical study to date of whether international fashion models have higher frequencies of anorexia nervosa (AN) and bulimia nervosa (BN) than non‐models.

Our hypothesis was that models are more likely to develop full or partial AN or BN than the control group.

2. METHODS

An online survey was conducted among models and non‐models using snowball sampling. Since no sampling frame for the models population is available or can feasibly be constructed, we used snowball sampling, aiming to access as broad and heterogenous a part of the population as possible. The online survey was distributed by various non‐profit organisations concerned with the well‐being and safety of models, and it was also circulated via international fashion model networks and social media platforms. As for the control group, snowball sampling allowed us to collect an international sample from a similar age group with our limited resources. The control group survey was distributed in several universities in several countries. 196 models and 305 non‐models responded to the questionnaire.

Model group participants were selected among fashion models who had been models for at least 1 year, by age (16–37years), gender (female only), height (≥170cm) and calculated BMI (BMI≤25). Overweight and obese subjects were excluded because they form a special population as they show different psychopathological features (Rajan & Menon,²⁰¹⁷), furthermore the present study aims to investigate straight size models rather than plus size models. A control group of internationally heterogeneous female individuals pursuing careers other than modelling was selected and adjusted to the same age group (16–37years). Participants who did not give any one of these details were omitted. Participants who gave scanty answers to the EDI questionnaire were also omitted. The final sample comprised 179 respondents in the model group and 261 respondents in the control group.

2.1. Measures

The survey consisted of the Eating Behaviour Severity Scale (EBSS) and the Eating Disorder Inventory (EDI) along with general sociodemographic and anthropometric questions. The EBSS assesses the frequency of disturbed eating and purging behaviour for the previous 4weeks and the previous 6months (Yager etal.,¹⁹⁸⁷). The EDI is one of the most used self‐rating instruments for the assessment of disturbed eating attitudes and behaviour and the main psychopathological symptoms found in patients with EDs (Garner etal.,¹⁹⁸³). It contains eight subscales and has been used in clinical trials and epidemiological screening surveys. The three diagnostic subscales were used in the research, and the internal consistencies were appropriate in both the model and control subsamples: in the Drive for Thinness subscale, Cronbach's alpha was 0.938/0.924 (model/control subsamples); in the Bulimia subscale, 0.921/0.861; and in the Body Dissatisfaction subscale, 0.929/0.914.

Simulated (i.e. questionnaire‐based) diagnoses for EDs were generated using BMI values, the suggested cut‐off scores for the EDI Drive for Thinness subscale, and the frequency of bulimic behaviours defined in the DSM‐5 (American Psychiatric Association,²⁰¹³). Diagnosis of AN was simulated by the criteria BMI<17.0 and EDI‐DT≥14. The BN diagnostic criteria were at least onebinge episode per week, at least one compensatory behaviour (vomiting, use of laxatives/diuretics/diet pills, dieting) after binging and an above cut‐off EDI Drive for Thinness (EDI‐DT≥14) and Bulimia (EDI‐Bulimia≥14) subscales. Criteria for subclinical AN were BMI below 18.5 and sub‐threshold test score (EDI‐DT: 10–13). Subclinical BN was diagnosed if binge eating episodes occurred no more than three times a month, there was at least one compensatory behaviour (vomiting, use of laxatives/diuretics/diet pills, dieting) after binging and there were sub‐threshold scores (EDI‐DT: 10–13 and EDI‐bulimia: 6–13). In accordance with the DSM‐5, we did not include the presence of amenorrhoea as a criterion.

2.2. Statistical analysis

The SPSS 23 statistical software was used for the analysis. Descriptive statistics were carried out on the data, and we calculated averages and standard deviations. In view of the relatively large sample size, group means were compared using the t‐test. We used Chi‐square tests to compare proportions. Agresti‐Coull interval estimates were given for the proportions as recommended by Brown, Caj, & DasGupta(²⁰⁰¹).

The survey was kept confidential and anonymous, and participation was voluntary.

The research is in accordance with the Helsinki Declaration and was approved by the Regional Research Ethical Board of the Semmelweis University Budapest (No. 3/2020).

3. RESULTS

The mean age of the model group was 25.9years (SD=4.7, range 16–37years; Table1). Mean height for the model group was 177.3cm (SD=3.58, range 170.0–188.0cm). The control group had a significantly lower mean height (M=167.4, SD=6.59cm, range 150.0–188.0cm). Mean BMI for the model group was 18.1 (SD=1.68, range=14.0–24.8). Mean BMI for the control group was significantly higher: 22.1 (SD=4.23, p<0.001, range 14.7–43.3). It is remarkable that 44.7% of the models reported BMI of between 18.5 and 17.0, and a further 21.2% reported it as under 17.0, that is, severely underweight. Only 12.3% of the non‐models control group had BMI of between 18.5 and 17.0, and for another 4.2%, it was under 17.0.

TABLE 1

Demographic data and frequency of simulated diagnoses of anorexia nervosa and bulimia nervosa among fashion models and in the control group

	Fashion models n=179	Control group
		Total	BMI ≤ 25	Height ≥170	Height ≥170 and BMI ≤ 25
		n=261	n=214	n=102	n=83
Age (year) mean (SD) range	25.9 (4.70)16–37	25.0 (4.97)16–37 p=0.077	24.8 (4.89)*16–37 p=0.022	26.0 (4.94)16–35 p=0.829	25.4 (4.67)16–35 p=0.471
Height (cm)	177.3 (3.58)170.0–188.0	167.4 (6.59)***	167.4 (6.55)***	174.0 (3.49)***	174.0 (3.56)***
		150.0–188.0	152.0–188.0	170.0–188.0	170.0–188.0
		p<0.001	p<0.001	p<0.001	p<0.001
Weight (kg)	56.9 (5.61)43.5–75.0	61.9 (12.69)***	57.7 (7.53)40.0–75.0	66.8 (13.52)***	61.8 (6.99)***
		40.0–125.0 p<0.001	p=0.259	45.8–125.0 p<0.001	45.8–75.0 p<0.001
BMI	18.1 (1.68)14.0–24.8	22.1 (4.23)***	20.6 (2.24)***	22.1 (4.52)***	20.4 (2.22)***
		14.7–43.3 p<0.001	14.7–24.8 p<0.001	15.7–43.3 p<0.001	15.7–24.6 p<0.001
Simulated anorexia nervosa	3.9% CI [1.8–8.1]	1.1% CI [0.2–3.5] p=0.057	1.4% CI [0.3–4.3] p=0.104	0.0% CI [0.0–4.5]* p=0.040	0.0% CI [0.0–5.4] p=0.066
Simulated subclinical anorexia nervosa	14.6% CI [10.1–20.6]	2.7% CI [1.2–5.6]*** p<0.001	3.3% CI [1.5–6.8]*** p<0.001	2.9% CI [0.7–8.7]*** p<0.001	3.6% CI [0.8–10.6]** p=0.005
Simulated bulimia nervosa	1.7% CI [0.4–5.1]	0.4% CI [0.0–2.4] p=0.186	0.5% CI [0.0–3.0] p=0.248	0.0% CI [0.0–4.5] p=0.257	0.0% CI [0.0–5.4] p=0.317
Simulated subclinical bulimia nervosa	6.2% CI [3.4–10.8]	10.0% CI [6.9–14.3] p=0.104	7.0% CI [4.2–11.3] p=0.442	7.9% CI [3.9–15.1] p=0.368	6.1% CI [2.4–13.9] p=0.614

Open in a separate window

*p<0.05, **p<0.01, ***p<0.001.

Regarding the ethnic diversity of the participants, the majority of fashion models identified as white (56.4% vs. 92.3% in controls), 2.8% were Asian (vs. 2.7% in controls), 3.4% were black (vs. 1.5 in controls), 7.8% of fashion models fall into other racial identification (vs. 3.4% of controls) while 29.6% of the fashion models did not provide data (vs. 0% of the controls; Table2). Participants were gathered from 36 different countries, including Austria, Belgium, Botswana, Canada, Ecuador, France, Iran, Korea, Norway, Pakistan, Russia, Tonga, and Vietnam. Most participants were from Hungary, France, Russia, the Netherlands, and the United States of America.

TABLE 2

Distribution of fashion models and control group by race and ethnicity

	Fashion models n=179	Control group
		Total	BMI ≤ 25	Height ≥170	Height ≥170 and BMI ≤ 25
		n=261	n=214	n=102	n=83
Asian	2.8%CI [1.1–6.6]	2.7% CI [1.2–5.6]	2.3% CI [0.8–5.5]	1.0% CI [0.0–6.0]	1.2% CI [0.0–7.3]
Black	3.4% CI [1.4–7.4]	1.5% CI [0.5–4.0]	1.4% CI [0.3–4.3]	2.0% CI [0.2–7.5]	1.2% CI [0.0–7.3]
White	56.4% CI [49.1–63.4]	92.3%*** CI [88.3–95.0]	93.5%*** CI [89.2–96.2]	97.1%*** CI [91.3–99.3]	97.6%*** CI [91.0–99.8]
Other	7.8% CI [4.6–12.8]	3.4%* CI [1.7–6.5]	2.8%* CI [1.2–6.2]	0.0% CI [0.0–4.5]	0.0% CI [0.0–5.4]
No data	29.6% CI [23.4–36.7]	0.0% CI [0.0–1.8]	0.0% CI [0.0–2.2]	0.0% CI [0.0–4.5]	0.0% CI [0.0–5.4]
Total	100%	100%	100%	100%	100%

Open in a separate window

*p<0.05, **p<0.01, ***p<0.001.

A simulated diagnosis of AN was made for seven models (3.9% CI [1.8, 8.1]) and three control subjects (1.1% CI [0.2, 3.5]; p=0.057). The fashion models had a significantly higher frequency of subclinical AN symptoms (26 subjects, 14.6% CI [10.1, 20.6]) than the control group (seven cases, 2.7% CI [1.2, 5.6]). There were threesimulated BN cases among the models (1.7% CI [0.4, 5.1]) and one (0.4% CI [0.0, 2.4]) among the controls.Bycontrast, subclinical BN was present in 26 non‐models (10% CI [6.9, 14.3]) and 11 models (6.2% CI [3.4, 10.8]).

The subgroup of non‐models adjusted for height (≥170cm) was still significantly shorter (mean height=174.0cm, SD=3.49cm) and had significantly higher BMI (22.1, SD=4.52) than the models. This group (n=102, mean age: 26.0years, SD=4.94, range: 16–35years) had significantly fewer simulated diagnoses of AN and subclinical AN (0.0% CI [0.0, 4.5] p=0.040 and 2.9% CI [0.7, 8.7] p<0.001 respectively). No case of simulated BN was found, and neither did the frequency of subclinical BN appear to be significantly higher among non‐models (7.9% CI [3.9, 15.1] p=0.368]).

Members of the control group adjusted for BMI (n=214, mean age: 24.8 (SD=4.89) years, range: 16–37years) were significantly shorter (M: 167.4cm SD=6.55, range: 152.0–188.0cm, p<0.001) and had significantly higher BMI (M=20.6, SD=2.24, p<0.001) than the models. The proportion of subclinical AN symptoms was significantly lower (3.3% CI [1.5, 6.8]) than among the models. Simulated BN was less frequent (0.5% CI [0.0, 3.0]) than among the models, but the difference was not significant. The frequency of subclinical BN showed no significant difference (7.0% CI [4.2, 11.3]) from the fashion models group.

The control group was further restricted by both height (≥170cm) and BMI (≤ 25), resulting in a subgroup more similar to the models in height and BMI (n=83, mean age: 25.4 (SD=4.67) years, range: 16–35years, M: 174.0cm SD=3.56, range: 170–188cm, p<0.001), although the differences remained statistically significant. There were no cases of simulated anorexia nervosa (0.0% CI [0.0–5.4] or simulated bulimia nervosa (0.0% CI [0.0–5.4], but simulated subclinical anorexia nervosa was significantly lower in this group (3.6% CI [0.8–10.6], p=0.005). Simulated subclinical bulimia nervosa was not significantly lower (6.1% CI [2.4–13.9]).

3.1. Discussion

Fashion models have a high risk of developing disordered eating habits. They experience pressure to keep a certain physique and many of them use weight controlling behaviours such as dieting, excessive exercising, use of laxatives, self‐induced vomiting and juice cleanses (Bogár & Túry,²⁰¹⁹; Rodgers etal.,²⁰¹⁷). Due to the size requirements of the fashion industry, models' BMI is below the lower healthy limit (Zancu & Enea,²⁰¹⁷). This is a long‐existing phenomenon in the fashion industry, and several regulations have been proposed to protect models from being pressured to reach unhealthy BMI values. These regulations have not been fully implemented (Bogár etal.,²⁰²¹). Runways are still dominated by models portraying an extremely thin beauty ideal (Pemberton,²⁰¹⁹).

Even though the extremely slim beauty ideal of the fashion industry has always been controversial, and the mass media raises concerns about skeletal models and their wellbeing, only a few studies have investigated the prevalence of EDs in this specific group. One study of dancers and models found 7% of them to show symptoms of AN, but half of those had an onset prior to the modelling course (Garner & Garfinkel,¹⁹⁸⁰). Another study found no difference between fashion models and the control group regarding disordered eating habits (Brenner & Cunningham,¹⁹⁹²). In an Italian study, fashion models, whose average BMI was 17.7, showed a significantly higher frequency of partial ED symptoms than the control group but only slightly greater prevalence of the full syndrome (Santonastaso etal.,²⁰⁰²). Another study reported a significantly higher frequency of EDs among fashion models than among non‐models (Preti etal.,²⁰⁰⁸). Professional fashion models reported higher scores on the EDI–DT subscale. Positive body appreciation increased with duration of modelling, as did drive for thinness, implying that already‐thin models have a strong desire to maintain their slim physique (Swami & Szmigielska,²⁰¹³).

To conclude, clinical ED syndromes were not found to be more frequent among models than among controls except for clinical AN compared to the height‐adjusted control group, but subclinical AN does appear to be more common. The findings of the present study are also partially in alignment with our previous study (Bogár & Túry,²⁰¹⁹).

Further selection by height was made. Since height is related to development of EDs (Favaro etal.,²⁰⁰⁷) and fashion models must meet a certain height to be employed, it seemed logical to adjust the control group according to stature. Although the participants in the fashion model group were significantly taller than the subjects from different professions, the models had higher frequencies of simulated AN and subclinical AN. This further emphasises the intense pressure on young models to be extremely thin. A survey conducted in 2016 found that 62.4% of fashion models had been asked to lose weight or change body size, and 69.4% to “tone up”, in the previous year (Rodgers etal.,²⁰¹⁷). This is evidence of intense pressure on fashion models to maintain a thin body even though the average BMI of models is well below the lower healthy limit (Preti etal.,²⁰⁰⁸; Rodgers etal.,²⁰¹⁷; Santonastaso etal.,²⁰⁰²).

Analogously to the selection criteria for fashion models, we filtered the control subgroup by adjusting for BMI (BMI<25). It seemed necessary to exclude the psychopathological traits typically displayed by overweight and obese individuals as they tend to experience higher body dissatisfaction and may internalise the thin ideal more. Dieting, binge eating or use of diet pills or laxatives can also be more frequent in such groups (Rajan & Menon,²⁰¹⁷). The results for the control group by comparison with the fashion model group remained similar after adjusting for BMI.

After constraining for both height and BMI, in spite of the lower sample size, we found a significantly higher frequency of subclinical AN among fashion models than in the control group. This result, obtained in the most comparable research setting, underlines the probable importance in the aetiology of ED of being a fashion model.

Our assessment of ED risk factors involved a higher number of multicultural female fashion models than any previous comparable study. Moreover, we used different validated questionnaires for simulated diagnoses of AN and BN in both clinical and subclinical forms. Our study group consists of internationally heterogenous participants. The survey was performed among models working at a high level of their profession.

3.2. Limitations

The study displays certain limitations. The survey was open for submission during the COVID‐19 pandemic, when fashion jobs were limited. This could mean that models were under less pressure to maintain their slim physique, so that their reported current weight – which we used to calculate their BMI – was higher than usual. Secondly, some models, especially those in their thirties, may be less active in their profession than they used to be in their teenage years, causing both the BMI and theanswers given to the EDI questions to differ from those they would have given at their most active and presumably skinniest period of their career. EDI questions are evaluated according to the current state of mind and not one the respondent might have had in the past. Thirdly, the possibly higher likelihood for models dissatisfied with the current measurement requirements to participate in a survey like ours might have distorted the overall view. The self‐reported height and weight values used to calculate BMI may differ from the actual values (Giacchi,¹⁹⁹⁸). Questions concerning previous eating disorders were not included in the questionnaire. According to a previous interview‐based qualitative research, 3 out of 53 fashion models were diagnosed with anorexia nervosa prior to their modelling career (Bogár & Túry,²⁰¹⁹).

Furthermore, the control group was not adjusted to the study group for socioeconomic and cultural background. Further studies are required to investigate the risk factors and the frequency of EDs among professional fashion models, and they should incorporate longitudinal design. The symptoms assessed should go beyond AN and BN and include those of other disturbed eating and exercise behaviours, such as orthorexia nervosa or exercise addiction. The findings of such studies could be of great importance in the development of health regulations in the fashion industry to prevent the spread of EDs among models and persons exposed to model images. The fashion industry must make changes that reflect the utmost importance of fashion models' physical and mental wellbeing. These should include putting a stop to the health‐damaging behaviours that agents and designers enforce on models and introducing regular medical consultations.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

ACKNOWLEDGEMENT

No financial support was received for this study.

Notes

Bogár, N., Dukay‐Szabó, S., Simon, D., Túry, F., & Pászthy, B. (2022). Frequency of disordered eating habits among fashion models. European Eating Disorders Review, 30(6), 823–829. 10.1002/erv.2912 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

Abraham, S. (1996). Eating and weight controlling behaviours of young ballet dancers. Psychopathology, 29(4), 218‐222. 10.1159/0002849962 [PubMed] [CrossRef] [Google Scholar]
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM‐5) (5th ed.). American Psychiatric Publishing. 10.1176/appi.books.9780890425596.dsm10 [CrossRef] [Google Scholar]
Bogár, N., Pászthy, B., & Túry, F. (2021). A divatipar egészségvédelmi szabályozása [Health protective regulation of the fashion industry – Hungarian]. Orvosi Hetilap, 162(23), 905–910. 10.1556/650.2021.32135 [PubMed] [CrossRef] [Google Scholar]
Bogár, N., & Túry, F. (2017). Evészavarok kockázati tényezői a divatszakmában dolgozók narratívumaiban [Risk factors of eating disorders in the narratives of members of the fashion industry – Hungarian]. Psychiatria Hungarica, 32(1), 41–53. [PubMed] [Google Scholar]
Bogár, N., & Túry, F. (2019). The fashion industry and eating disorders – The dangers of the catwalk. Cambridge Scholars Publishing. [Google Scholar]
Brenner, J. B., & Cunningham, J. G. (1992). Gender differences in eating attitudes, body concept, and self‐esteem among models. Sex Roles, 27(7):413‐437. 10.1007/BF00289949 [CrossRef] [Google Scholar]
Brown, L. D., Cai, T., & DasGupta, A. (2001). Interval estimation for a binomial proportion. Statistical Science, 16(2), 101–117. 10.1214/ss/1009213286 [CrossRef] [Google Scholar]
Dauxerre, V. (2017). Size zero: My life as a disappearing model. HarperCollins Publishers. [Google Scholar]
Derenne, J. L., & Beresin, E. V. (2006). Body image, media, and eating disorders. Academic Psychiatry, 30(3), 257‐261. 10.1176/appi.ap.30.3.257 [PubMed] [CrossRef] [Google Scholar]
Favaro, A., Tenconi, E., Degortes, D., Soave, M., Zanetti, T., Nardi, M. T., Caregaro, L., & Santonastaso, P. (2007). Association between low height and eating disorders: Cause or effect?International Journal of Eating Disorders, 40(6), 549‐553. 10.1002/eat.20409 [PubMed] [CrossRef] [Google Scholar]
Garner, D. M., & Garfinkel, P. E. (1980). Socio‐cultural factors in the development of anorexia nervosa. Psychological Medicine, 10(4), 647‐656. 10.1017/s0033291700054945 [PubMed] [CrossRef] [Google Scholar]
Garner, D. M., Olmstead, M. P., & Polivy, J. (1983). Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders, 2(2), 15‐34. 10.1002/1098-108X(198321)2:2<15::AID-EAT2260020203>3.0.CO;2-6 [CrossRef] [Google Scholar]
Giacchi, M., Mattei, R., & Rossi, S. (1998). Correction of the self‐reported BMI in a teenage population. International Journal of Obesity, 22(7), 673‐677. 10.1038/sj.ijo.0800646 [PubMed] [CrossRef] [Google Scholar]
Johnson, K. (2011). Importing the flawless girl. Nevada Law Journal, 12, 831–866. [Google Scholar]
Park, S.‐Y. (2005). The influence of presumed media influence on women’s desire to be thin. Communication Research, 32(5), 594–614. 10.1177/0093650205279350 [CrossRef] [Google Scholar]
Pemberton, B. (2019). Waif of shame. Skeletal models including Kaia Gerber storm YSL runaway despite fashion giant’s vow to end use of painfully thin women. Retrieved February 28, 2019, from www.thesun.co.uk/fabulous/8531185/skeletal‐models‐kaia‐gerber‐ysl‐runaway‐painfully‐thin [Google Scholar]
Preti, A., Usai, A., Miotto, P., Petretto, D. R., & Masala, C. (2008). Eating disorders among professional fashion models. Psychiatry Research, 159(1–2), 86‐94. 10.1016/j.psychres.2005.07.040 [PubMed] [CrossRef] [Google Scholar]
Rajan, T. M., & Menon, V. (2017). Psychiatric disorders and obesity: A review of association studies. Journal of Postgraduate Medicine, 63(3), 182‐190. 10.4103/jpgm.JPGM_712_16 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
Rodgers, R. F., Ziff, S., Lowy, A. S., Yu, K., & Austin, S. B. (2017). Results of a strategic science study to inform policies targeting extreme thinness standards in the fashion industry. International Journal of Eating Disorders, 50(3), 284‐292. 10.1002/eat.22682 [PubMed] [CrossRef] [Google Scholar]
Santonastaso, P., Mondini, S., & Favaro, A. (2002). Are fashion models a group at risk for eating disorders and substance abuse?Psychotherapy and Psychosomatics, 71(3), 168‐172. 10.1159/000056285 [PubMed] [CrossRef] [Google Scholar]
Scully, J. (2016, December 5). How fashion became about power, and lost its ability to dream. http://www.youtube.com/watch?v=5L5uYojEJTw [Google Scholar]
Swami, V., & Szmigielska, E. (2013). Body image concerns in professional fashion models: Are they really an at‐risk group?Psychiatry Research, 207(1–2), 113‐117. doi.org/ 10.1016/j.psychres.2012.09.009 [PubMed] [CrossRef] [Google Scholar]
Yager, J., Landsverk, J., & Edelstein, C. K. (1987). A 20‐month follow‐up study of 628 women with eating disorders, I: Course and severity. American Journal of Psychiatry, 144(9), 1172‐1177. 10.1176/ajp.144.9.1172 [PubMed] [CrossRef] [Google Scholar]
Zancu, S. A., & Enea, V. (2017). Eating disorders among fashion models: A systematic review of the literature. Eating and Weight Disorders, 22(3), 395‐405. 10.1007/s40519-016-0293-5 [PubMed] [CrossRef] [Google Scholar]

Articles from European Eating Disorders Review are provided here courtesy of Wiley