You are here

  • Home
  • Archive
  • Volume 8, Issue 2
  • Behind the athletic body: a clinical interview study of identification of eating disorder symptoms and diagnoses in elite athletes

Article Text

Article menu
Qualitative research
Behind the athletic body: a clinical interview study of identification of eating disorder symptoms and diagnoses in elite athletes
  1. Mia Beck Lichtenstein1,
  2. Karen Krogh Johansen2,
  3. Eik Runge2,
  4. Marina Bohn Hansen3,
  5. Trine Theresa Holmberg2,
  6. Kristine Tarp2
  1. 1Department of Clinical Research, Syddansk Universitet Det Samfundsvidenskabelige Fakultet, Odense, Denmark
  2. 2Research Unit for Digital Psychiatry, Mental Health Services in the Region of Southern Denmark, Odense, Denmark
  3. 3Department of Psychology, University of Southern Denmark, Odense, Denmark
  1. Correspondence to Dr Mia Beck Lichtenstein; mlichtenstein{at}health.sdu.dk

https://doi.org/10.1136/bmjsem-2021-001265

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Key messages

    What is already known

    • Eating disorders are more prevalent in athletic samples than in the general population.

    • Previous studies have explored the prevalence of eating disorders using generic questionnaires but are missing diagnostic interviews verifying the results.

    What are the new findings

    • Despite strong athletic bodies, athletes suffer from emotional distress and physical health problems.

    • Eating disorders may be overlooked by trainers when athletes strive for top performances.

    • The most prevalent eating disorder among athletes in this sample is Eating Disorder Not Otherwise Specified, with restrictive eating, body dissatisfaction and fear of weight gain.

    • Eating Disorders Examination Questionnaire and SCOFF questionnaire are valid instruments to identify eating disorder symptoms in athletes.

    Introduction

    Eating disorders (EDs) in athletes are increasing interest and debate topics. EDs may be triggered by performance pressure, comments from coaches or teammates, public weigh-ins or societal expectations of an athletic or skinny body.1 EDs may have serious negative health consequences, low energy availability and reduced sports performance.2–6

    Symptoms of EDs are described in the diagnostic manuals: WHO’s ‘International Statistical Classification of Diseases and Related Health Problems, 10th revision’ (ICD-10)7 and ‘The Diagnostic and Statistical Manual of Mental Disorders, fifth edition’ (DSM-5).8

    Anorexia nervosa (AN) is characterised by restrictive eating, low body weight, fear of weight gain and disturbed body image. Atypical AN can be diagnosed if one symptom is not present (atypical anorexia nervosa (AAN)). It is noteworthy that, unlike AN, low body weight is not a criterion for AAN. Bulimia nervosa (BN) is characterised by binge eating and compensatory behaviours such as vomiting or excessive exercise. EDs with lower severity or frequency of symptoms are classified as Eating Disorders Not Otherwise Specified (EDNOS) in ICD-10 and as Other Specified Feeding and Eating Disorder (OSFED) in DSM-5.

    Subclinical conditions are often labelled as disordered Eating (DE), comprising a range of irregular eating behaviours and negative body image not fitting the diagnostic criteria.

    The prevalence of ED/DE has been investigated across countries and athlete populations, including Danish,9 Spanish10 and Slovenian.11 Previous studies found that 10%–20% reported symptoms of EDs.12–18 Athletes have higher prevalence rates than non-athletes, especially in weight-sensitive sports such as endurance, antigravity and aesthetic sports.17 19 Female athletes have a higher risk than male athletes, with studies finding 20% of females compared with 8% of males in a Norwegian study.17

    Different instruments have been validated and used to measure ED symptoms in the general population or clinical samples. The SCOFF questionnaire (SCOFF) is widely used to screen larger populations. SCOFF is an acronym for the five assessed symptoms (Sick, Control, One stone, Fat and Food).20 With only five items, SCOFF is an easy and simple tool used to measure symptoms of an ED. The Eating Disorder Examination Questionnaire (EDE-Q) is a 28-item self-reported questionnaire designed to assess the range and severity of features associated with a diagnosis of an ED.21 It has been used in athletic samples to identify ED symptoms.16 22–24

    However, these tools cannot assess the specific ED diagnosis nor the severity of symptoms. For that purpose, conducting a clinical interview addressing eating patterns, body image and exercise pattern is imperative. Using diagnostic manuals to assess athletes’ ED pathology can provide information about the screening instruments’ symptom load and diagnostic validity.

    If EDE-Q and SCOFF can identify EDs in athlete samples, they may be useful for regular screening of athletes that need support and treatment. Regular screening of athletes has been requested in previous research investigating EDs in athletes.1 12–15 Still, there is little evidence of the validity of using generic ED tools in athletic samples, and widespread or routine screening has not been adopted. In addition, clinical interviews have been requested,16 but only very few studies6 17 18 have conducted clinical interviewing, and none of them has used ICD-10 or DSM-5.

    Aims

    This study aimed to use a diagnostic interview to assess the external validity of generic ED questionnaires (EDE-Q and SCOFF) to identify athletes with EDs and to explore the subjective experience of symptomology in a sample of athletes competing at the national or international level.

    Methods

    Design

    The present study’s design was mixed methods, employing a triangulation design to validate a quantitative model and interpret quantitative and qualitative data in conjunction.25 The first part was a psychometric study validating the diagnoses identified by generic EDs questionnaires; the second part was based on 28 semi-structured interviews using a ‘general inductive approach (GIA) for analysing’. The study followed the Consolidated Criteria for Reporting Qualitative Studies.26

    Participants

    Elite athletes (competing at the national or international level) were recruited from a Danish survey study published in 202027 using EDE-Q and SCOFF to assess the prevalence of ED symptoms. A sample of 417 Danish male and female national team athletes, including national team recruiting squads, completed an online questionnaire. All participants were invited by email in 2019 and 2020 (mean age 20 years, 51% females) from the following sports categories: running, cycling, orienteering, triathlon, swimming, track and field, martial arts, rowing, sports dance, ice skating, gymnastics, badminton, handball and football. The questionnaires were repeated in January–March 2021 to obtain a current score useful for the interview study. We invited 75 athletes with a high score and 75 with a low score on SCOFF and EDE-Q. We had 59 responses (17 with a high score and 42 with a low score), of whom 28 (17 with a high score and 11 with a low score) were invited to participate in the interview and 18 were accepted (10 with a high score and 8 with a low score).

    Furthermore, we recruited participants from a similar study conducted on subelite athletes (competing at a national level but not on a national team). A total of 206 subelite athletes (29 with a high score and 177 with a low score) (mean age 21 years; females 52%) completed the questionnaires in February–April 2021 from the following sport disciplines: running, cycling, orienteering, triathlon, swimming, track and field, martial arts, rowing, sports dance, ice skating, gymnastics, badminton, handball and football. We invited 16 athletes (10 with a high score and 6 with a low score) to participate in the interview, and 10 accepted (8 with a high score and 2 with a low score).

    The final sample consisted of 28 athletes. Of these, 18 had high scores on EDE-Q (cut-off 2.3) and SCOFF (cut-off 2), while 10 had low scores. The age span was 16–47 years (mean age 23 years) with 25 females (89%). The athletes came from the following sports: track and field (n=10), rowing (n=6), cycling (n=4), triathlon (n=2), swimming (n=2), orienteering (n=2), martial arts (n=1) and ice skating (n=1).

    Instruments

    The Eating Disorder Examination (EDE-Q) is a gold standard for assessing the severity and type of ED pathology.21 The scale contains 28 items and 5 subscales: restraint, eating concern, shape concern and weight concern. A total score called ‘global score’ is an average of the subscales. We used a Danish version of the EDE-Q validated in athletes.27 A global score of >2.3 has been used to categorise female athletes at risk of EDs,28 29 while >1.7 has been suggested for male athletes.30 SCOFF has five items designed to assess ED symptoms.20 The responses are scored dichotomous as ‘yes’ or ‘no’, with two or more positive answers indicating the risk of an ED. A meta-analysis of the diagnostic accuracy of the SCOFF scale showed excellent performance.31 SCOFF has been translated and validated in Danish, showing reasonable sensibility and sensitivity.32

    In this study, we used the EDE-Q as the main instrument to identify ED symptoms as the interview guide was developed on the EDE interview.

    We also included basic sociodemographic data such as age, gender, height, weight, athletic level and type of sport in the survey.

    Interviews

    The interviews followed an interview guide based on the EDE interview adapted to assess ED symptoms from DSM-5 and ICD-10. The present study developed an interview guide to diagnose EDs in athletes. The guide was called the Eating Disorder Interview for Athletes (EDIA). The EDIA is attached in online supplemental appendix 1.

    Supplemental material

    Five interviewers conducted all 28 interviews between March and June 2021. All interviewers were certified in using the EDE interview. None of the interviewers had previous relations with the athletes.

    The interviews were conducted on video due to restrictions from the COVID-19 pandemic. All interviews were transcribed based on audio recordings that lasted 31–114 min in high-risk athletes and 20–98 min in low-risk athletes. After the interview, all participants were offered oral feedback and evaluation with information about diagnosis and treatment options.

    Qualitative analyses

    Qualitative data from the semi-structured interviews were analysed through a GIA.33 GIA is frequently used within the health area as it offers a straightforward method of systematic analysis of qualitative data to derive findings linked to evaluation questions. GIA is generic and, as such, not guided by explicit philosophical assumptions. Thus, findings are to emerge inductively from the themes in the raw data without restraints imposed by structured methodologies.

    The interviewers summarised the main themes of the transcriptions based on the EDIA. The main themes were the central ED symptoms: eating concerns, weight concerns, shape concerns, excessive exercise and health problems. Data analysis was inductive and carried out through multiple readings of raw text regarding subthemes embedded in the content under these categories. All authors conducted the coding, and no software was used. Raw data were interpreted into a condensed summary format. The interpreters decided what data to use and what not to. Thus, one text segment may have been coded into more than one subtheme, and some may not have been coded if not relevant to the categories. Essential quotations were added. The authors assessed trustworthiness by checking each other’s coding.

    Results

    First, we present the ED diagnoses that athletes’ symptom descriptions indicated in the interviews. Next, we present the symptoms derived from the inductive approach for analysing interview data.

    Eating disorder diagnoses in the athletes based on the clinical interview

    An overview of the participants, the scores on EDE-Q and SCOFF and the suggested diagnoses are presented in table 1 (the double line distinguish confirmed diagnosis from no diagnosis). The most common diagnoses, based on ICD-10, were EDNOS with restrictive eating patterns (EDNOS-r) (n=12) and atypical anorexia (AAN) (n=6).

    View this table:
    Table 1

    Participant’s sport, level, gender, BMI, EDE-Q, SCOFF and diagnoses based on ICD-10 and DSM-5

    EDE-Q was the main instrument for classification in this study. It identified 18 cases with scores above EDE-Q cut-off 2.3 and 9 cases with low scores on EDE-Q. However, EDE-Q failed to identify two athletes (ID 20 and ID 23) diagnosed with an ED in the interview.

    SCOFF identified 18 cases with scores above the cut-off 2 and 8 with low scores. However, SCOFF failed to identify one athlete with an ED diagnosis (ID 18), while it found an ED not confirmed by the interview (ID 21).

    Table 2 contains the counts on true and false positive and negative results on EDE-Q and SCOFF, respectively. EDE-Q has a sensitivity of 90% and a specificity of 100%, while SCOFF has a sensitivity of 94% and a specificity of 88%.

    View this table:
    Table 2

    Sensitivity and specificity of EDE-Q (n=27) and SCOFF (n=26)

    A qualitative analysis of the eating disorder symptoms presented in the interviews

    Based on the 28 interviews, the following categories were identified within the 22 subjects who displayed ED symptoms: eating concerns (restrictive or irregular eating, distress and guilt related to eating, binge eating and compensatory behaviours), weight concerns, shape concerns, excessive exercise and health problems. Table 3 presents an overview of the ED symptoms derived from the EDIA. None of the above-mentioned themes emerged in subjects with no diagnosis (n=6). Thus, the qualitative results, including citations, are based only on subjects with confirmed ED diagnoses (n=20) and DE (n=2). We will examine the most prominent aspects of each category individually below: eating concerns, weight concerns, shape concerns, excessive exercise, and health problems.

    View this table:
    Table 3

    Psychological, behavioural and physical signs/symptoms of eating disorders in interviewed athletes

    Eating concerns

    Restrictive or irregular eating

    Athletes with ED symptoms reported having chaotic, irregular or restrictive eating patterns. All athletes indicated that they recently had restrained or did restrain the amount of food consumed, periodically or constantly. Twelve out of 22 athletes reported that they had devalued or skipped meals throughout the day to lose weight or influence body shape. Ten athletes counted calories or followed a 1300–1500 kcal diet per day. One athlete reported that she drank 8 L of water a day to suppress hunger. The motives were related to concerns about weight, body shape and sports performance. ID 1 explained it:

    If I had gone a whole day without eating, I was satisfied with myself. And my training sessions were often placed at dinner time, so when I came home, I said to myself: I’m not that hungry today, and then maybe I wouldn’t eat throughout the whole day. I just don’t think that I ate that much. I mean, I was trying to skip all the meals that I could get away with.

    Twenty-two athletes reported avoiding specific foods, which they may otherwise like, typically high in fat, sugar, carbohydrates, calories or processed foods. Four athletes explicitly stated this was with the desire to lose weight. ID 18 elaborated:

    So, I’ve cut everything possible, whatever I could strip. I eat cheese with the lowest fat percentage, and I only drink skimmed milk. I like bread but haven’t eaten it for many years. So, yes, that’s it, and of course sweets and all that stuff, I also have no interest in eating that.

    Distress and guilt related to eating

    Twenty athletes reported that thoughts about food, eating or calories were distressing, disturbing, distracting, irritating, burdensome or took up much mental space. Seventeen athletes often experienced emotions of guilt and shame concerning eating. ID 16 had experienced it like this: “If I feel I have been eating too much or too sugary… Well, I get the feeling and self-perception that I weigh—I don’t know—20 kg more than I really do.”

    Many statements linked these thoughts to poor self-esteem and self-image: ID 1 had felt it like this:

    I just think I had a very poor self-image, and I didn’t feel that I was good enough. I wasn’t good enough at my sport, I wasn’t good enough at looking good, and I wasn’t good at anything at all. So, I tried to take control of my eating because perhaps that could help me.

    Binge eating and compensatory behaviours

    Nine athletes were concerned, anxious or fearful of losing control of their eating or experiencing a binge-eating episode. Eleven athletes had experienced a subjective or objective binge-eating episode. Six athletes reported compensatory behaviour after binge eating or eating ‘unhealthy foods’ such as sweets, cake, ice cream, etc, or carbohydrate or fat-dense foods, like pizza, pasta and cheese.

    Additional training was also used as a compensatory behaviour to avoid feeling guilty about eating. Eight athletes reported previous purging behaviour to compensate for (subjective) binge eating or eating calorie-dense meals. Although most of these athletes did not purge anymore, four still had thoughts about doing it.

    Changes in the ability to train normally due to injury, disease, vacation or lack of training accessibility could trigger compensatory behaviours such as restrictive energy intake (reducing carbohydrate intake and increasing intake of vegetables and protein-dense foods).

    This pattern is also common on days off from training. Athletes reported feeling anxious, depressed, frustrated or fearful of becoming fat or gaining weight. While injured from running, ID 23 mentioned that she chose not to participate in a family vacation because she experienced distress at the thought of not having access to a gym: “As we were heading to Malta, I was about to die not knowing if there was a gym because I was unable to run… it’s alright that I am not running now, but then I do excessive bike sessions on my home trainer.”

    Weight concerns

    Twenty athletes reported being unsatisfied with their weight, and 15 athletes had a desire to lose weight. ID 4 explained: “I want to do everything to become the best, but if I am to be the best, I have to be sickly thin. All cyclists are constantly dieting, and there are no limits to how thin you can be.”

    Reasons for weight concerns included body image, sports performance and feeling uncomfortable after weight gain. Eighteen athletes were anxious about gaining weight during an injury. They felt triggered by comments from coaches or teammates, and ID 4 experienced pressure from her family: “My father says that cyclists starve themselves. He says I shouldn’t weigh more than 60 kg.”

    Five athletes were ambivalent about their weight. On the one hand, they wanted to lose weight, while on the other hand, they worried about losing muscle mass negatively impacting sports performance: ID 8 explained: “If I lose weight, I will think of my thighs—my legs look tiny as if I am losing muscle mass.”

    The athletes had varying attitudes towards weighing themselves. Some strived for control by weighing themselves regularly, while others avoided the scale because they felt emotional distress from seeing their weight.

    Shape concerns

    Twenty-one athletes experienced dissatisfaction with their bodies, especially their legs, waist, hips, arms, ribs and stomach. Three athletes reported a feeling of being fat on the face, shoulders or back. ID 5 expressed: “There are some days where I just don’t want to be in my body.”

    The athletes reporting shape concerns used terms like embarrassment (shame), distress, feeling fat, chubby, sad and insecure. The feeling of shame caused some athletes to cover or hide their bodies in oversized clothing. ID 15 elaborated: “Well, I think that as a swimmer… you are almost undressed when you do your sport. I look down at my body every single day. I don’t think a day passes where I don’t think that I want my legs to be skinnier.”

    The reasons for being dissatisfied with the body were aesthetical, comments from coaches or teammates about their body, a great desire to have an athletic appearance (toned muscles and visible blood vessels), comparing bodies in sporty outfits, striving for weight optimisation and a fear of others’ thoughts.

    Twenty-one athletes tried to control their body and appearance by assessing body fat percentage, body weight and muscle mass and checking their bodies in the mirror or windows. Furthermore, 12 athletes used an app to monitor body size, assess fat around the stomach or thighs or check if the clothes still fit. However, eight athletes said that they tried to protect themselves from scales, mirrors and photographs to prevent triggering a negative body perception.

    Excessive exercise

    All athletes described increasing exercise volume and intensity to burn calories, prevent weight gain or modify body shape and tone muscles. The athletes practised various cardio exercises, usually long walks, cycling as transport, running and strength training. These extra workouts were usually beyond their coach’s recommendations. However, two athletes had been told by their coach to do extra cardio to lose weight. Twelve athletes reported that they had exercised or regulated their diet in periods where they dealt with injuries. ID 17 expressed how exercise satisfied the need for control: “There is so much control connected to training in minutes and distance. I don’t want to let that control go at all.”

    Health problems

    All athletes reported that body and health were influenced by their eating and exercise patterns. Nine athletes experienced amenorrhoea due to low body fat, ranging from a couple of days to >2 years. Eleven athletes explained that they often felt pain and tenderness during training. In addition, seven athletes reported increased cold sensitivity and freezing due to prolonged training and insufficient energy intake. Despite wearing more clothes than others, they still felt cold, especially around their fingers and toes. Other health issues were skin problems, hair loss, hunger, energy loss and dizziness.

    Twelve athletes experienced physical or mental fatigue in general and in connection to training. They explained that they felt exhausted and the body needed a longer recovery time. They also reported poor sleep, not being able to relax and more illness. For example, ID 9 explained how fatigue influenced physical health: “I clearly feel days where I am, like, have avoided eating or haven’t eaten, I clearly feel that I become much more tired and that I become tired faster. Also, my muscle recovery becomes a lot worse.”

    The athletes also experienced that their mental health was affected by obsessive exercise and weight loss. ID 9 further elaborated on fatigue and mental health: “I mean, when I become tired and such, I don’t perform how I am supposed to in training, and then I become sad or disappointed. I become disappointed about myself and my performance.”

    Discussion

    Eating disorder diagnoses in athletes

    Clinical interviews using EDIA found 20 athletes with ED diagnoses according to ICD-10 and DSM-5. Of these, 72% were diagnosed with EDNOS, and 28% had atypical anorexia. The EDIA found ‘no diagnoses’ in eight athletes, but two of them presented subclinical symptoms (DE). None of the athletes in the present study had AN because body mass index (BMI) was over 18.5 (probably due to high muscle mass), and none presented regular binge-purging behaviour. Two athletes reported that they had previously been diagnosed with AN but had increased body weight during the year before the interviews.

    Unspecified ED diagnosis was the most prevalent pathology in female athletes presenting for ED treatment in a study by Strandberg et al,34 finding 60% with OSFED, 33% with AN and 7% with BN. EDNOS or OSFED are conditions that may be less visible or perceived as normal in the sports culture. This reality contributes to the perception of not being ‘sick enough’ to warrant treatment. It serves as an obstacle to assessment and timely intervention, putting athletes’ health at risk.

    Sensitivity and specificity of EDE-Q and SCOFF

    This study confirmed that EDE-Q and SCOFF presented high external validity in identifying ED diagnoses. EDE-Q presented 90% sensitivity and 100% specificity, while SCOFF presented 94% sensitivity and 88% specificity.

    These findings indicate that EDE-Q and SCOFF can be used to screen athletes for EDs, but false negatives (under-reporting) may appear using EDE-Q. At the same time, over-reporting was not an issue in the current study. Using SCOFF, there is a small risk of false positives (one case) and false negatives (one case).

    Thus, previous prevalence studies on ED symptoms based on SCOFF and EDE-Q seem to be valid and reflect a true problem. However, a sport-specific screening instrument may increase specificity and increase the knowledge about sport-specific eating concerns and body dissatisfaction.

    A study on eating disorders in sport by Fewell et al35 found that athlete patients had significantly lower EDE-Q scores than ED in non-athlete patients, indicating under-reporting of athletes in self-report measures. The authors suggested using a clinical interview for more accurate accounts of ED symptoms, despite being less feasible due to time and expense. We recommend using a diagnostic interview such as EDIA to confirm the type and severity of the diagnosis in an athlete sample.

    Athletes’ eating concerns and patterns

    The athletes often described their eating patterns as restrictive and irregular, with different attempts to reduce the number of calories per day (counting calories, eating healthy foods, skipping meals, following a diet and avoiding sugar and fat or processed food). Feelings of shame, guilt, anxiety and distress were typical food-related emotions. Compensatory behaviours reported after binge eating, ‘eating unhealthy foods’, injuries or vacations were restrained eating, obsessive exercise or purging attempts. Regarding injuries or changes in exercise routines as risk factors, our findings are consistent with findings from a qualitative analysis of factors related to ED onset in female collegiate athletes and non-athletes by Arthur-Cameselle et al.1 They found that fears of weight gain because of less physical activity, for example, injuries or illness, could lead to overcompensation via restrictive dieting or purging.

    Body dissatisfaction, triggers and health consequences

    The current study found that motives for restricting eating were the desire for an athletic, muscular and aesthetic body in the athletic culture. Adding to this, Arthur-Cameselle et al1 found that more than two-thirds of the athletes reported that the sports environment was related to the onset of EDs. Performance pressure was a trigger like an attempt to improve performance or achieve perceived body norms.

    Moreover, athletes in our study reported embarrassment and dissatisfaction with their legs, stomach and hips, despite having normal BMI values and athletic bodies. A study from Iceland found high levels of body dissatisfaction in elite athletes across 20 different sports,16 pointing at the pressure to conform to an ideal body shape to improve performance. In that study, body dissatisfaction was a predictor of EDs. A Dutch study also found that negative body image was related to EDs in female elite athletes.36 Athletes with EDs felt fatter than athletes without EDs, despite no weight or body composition difference. The study also found that athletes often compare their bodies with competitors, triggering body dissatisfaction. These findings agree with the study by Arthur-Cameselle et al,1 who found that almost all athletes reported poor body image or body image concerns, resulting in negative comparisons to peers, even though they were not overweight. More than two-thirds reported feelings of inadequacy when they compared themselves with others. The authors speculated whether this could be an example of a double bind, where the athlete feels competent in sports but inadequate when it comes to attractiveness.1 37 38

    Furthermore, several athletes in our study reported ambivalence about weight and shape. They feared weight gain and were striving for low body weight, but at the same time, they were aware that energy and muscle mass is important for sports performance. In a qualitative analysis of cultural and environmental associations with body image, diet and well-being in female distance runners,39 a so-called ‘runner-body’ was described by one runner as a ‘series of contradictions’ where you were to be ‘firm but shapely, fit but sexy, strong but thin’. The runners noted a complex and competitive perspective from two different contexts: they wanted to fit society’s role of being ‘skinny’ but still look healthy and perform well within their sport. The nine athletes participating in a qualitative study about social pressure in elite sports and EDs by Stoyel et al40 noted this duality of the ideal athlete body.

    Some athletes in our study reported controlling or assessing their body dimensions daily. Others reported not weighing themselves or monitoring bodily appearance because it would lead to increased body control and restrictive behaviours. In the study by Arthur-Cameselle et al,1 almost half of the athletes described perfectionism and a desire for control. Regulating food intake allowed them to feel ‘perfect’, which led to over-reliance on ED behaviour.

    In addition, we found that comments from coaches, teammates or parents could trigger or maintain DE behaviours. These findings agree with Carson et al,39 who found that coaches’ comments about body weight and shape add further pressure on female athletes, regardless of whether the comments were targeted directly at them. Food-related and weight-related comments were reported to contribute to a harmful team environment and influence athletes’ eating and exercise behaviours. Similarly, the study by Arthur-Cameselle et al1 found that three-quarters of the athletes had experienced peer issues as triggers to the onset of EDs, and more than half of the athletes identified harmful influence from peers who modelled DE and exercise leading to an imitation of unhealthy habits or competitive dieting. Finally, a study by Byrne and McLean41 has shown that especially female athletes tend to experience sociocultural pressure to conform to a lean body shape, which is in concordance with our study. This constitutes a problem, as the same study indicated that thin-build athletes have a greater risk of having or developing EDs than normal-build athletes.41

    Finally, several athletes in the current study also reported health consequences. Examples were: amenorrhoea, pain, freezing, hair loss, dizziness, energy loss, fatigue, exhaustion, frequent colds, sleep problems, restlessness, stress, reduced performance and reduced recovery after exercise. Most of these symptoms are often seen in a well-described condition called relative energy deficiency in sport (RED-S), where the body lacks the energy to meet the demands of the performed amount of exercise. A strong association between EDs and RED-S has been reported.2–6

    Clinical implications and future recommendations

    To detect and address symptoms of EDs among athletes, we have summarised the symptoms covered in our study in table 3. Athletes, coaches and relatives can use the table to increase knowledge and awareness of ED symptoms in athletes. Systematic screening of athletes and referral to a multidisciplinary team was recommended in a review focusing on the assessment and management of EDs in athletes.42

    For further elaboration on how to detect, address and manage EDs in athletes, we refer to the guidelines by the National Athletic Trainers’ Association and the ‘Coach and Athletic Trainer Toolkit’ by the National Eating Disorder Association.43 44 These guidelines emphasise the unique position of coaches to play an active role in detection of ED symptoms and encouraging referrals to health professionals for full evaluation and treatment of ED symptoms in athletes.

    However, we discovered that sometimes coaches trigger and maintain these symptoms. Thus, it is of importance that athletes have a supporting environment, where especially coaches are educated in spotting ED symptoms and how to support athletes. This calls for structural changes in sports federations to ensure commitment from coaches. An issue in most sports is that coaches are often primarily volunteering with limited time for education.

    Limitations of the study

    This study did not include a control group of non-athletes completing the questionnaires and the interviews. Almost all participants were females, but future studies should include more males to explore potential gender differences in eating and body pathology.

    To strengthen the study, the quantitative analyses, based on the questionnaires, were supplemented by the qualitative analyses from the interviews.45 It is a strength that 28 athletes were interviewed, and data saturation was reached.

    Limitations include that the transcripts were not returned to the interviewees for feedback on the correctness of the transcriptions and that the codes were not stakeholder checked46 by letting the participants provide feedback on the findings.26 However, it may increase the reliability and internal validity of the findings47 and enhance the credibility of the analysis that five researchers have worked together on coding the data.33 48

    Conclusion

    This mixed-methods approach found that EDE-Q and SCOFF are valid instruments to identify athletes at risk of EDs with high sensitivity (>90% and >94%) and specificity (100% and >88%). Athletes do not over-report ED problems according to EDE-Q but may under-report. Sport-specific instruments are recommended to increase the sensitivity by identifying athletes with subclinical scores on generic tools. Still, despite SCOFF being a short, generic tool for large screening purposes, it picks up ED pathology.

    The most prevalent symptoms of the athletes were restrictive eating patterns, weight and shape concerns related to athletic appearance, comments from coaches/teammates, fear of weight gain, body shame, excessive exercise, especially in periods with injuries and health problems (pain and fatigue). These findings indicate that athletes with ED symptoms are highly distressed about eating/foods, exercise and bodily appearance triggered by the sporting environment despite having athletic bodies.

    Increased information about ED symptoms, routine screening with validated instruments and clinical interviews (such as EDIA) may help identify EDs and thus prevent severe cases.

    Data availability statement

    Data are available on reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Clinical interviews without intervention do not need ethical approval, as confirmed by The Regional Committees on Health Research Ethics for Southern Denmark (case number 20192000-177). All participants gave oral and written consent to participate in the study.

    Acknowledgments

    Thanks to Team Denmark and The Danish Sport Federation (DIF). Thanks to Anna Melin for the project idea and collaboration. Thanks to Jacob Bindzus for conducting four clinical interviews.

    References

    1. ↵
      1. Arthur-Cameselle J,
      2. Sossin K,
      3. Quatromoni P
      . A qualitative analysis of factors related to eating disorder onset in female collegiate athletes and non-athletes. Eat Disord 2017;25:199–215.doi:10.1080/10640266.2016.1258940pmid:http://www.ncbi.nlm.nih.gov/pubmed/27897463
      OpenUrlPubMed
    2. ↵
      1. Mountjoy M,
      2. Sundgot-Borgen J,
      3. Burke L, et al
      . The IOC consensus statement: beyond the Female Athlete Triad--Relative Energy Deficiency in Sport (RED-S). Br J Sports Med 2014;48:491–7.doi:10.1136/bjsports-2014-093502pmid:http://www.ncbi.nlm.nih.gov/pubmed/24620037
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Mountjoy M,
      2. Sundgot-Borgen JK,
      3. Burke LM, et al
      . IOC consensus statement on relative energy deficiency in sport (RED-S): 2018 update. Br J Sports Med 2018;52:687–97.doi:10.1136/bjsports-2018-099193pmid:http://www.ncbi.nlm.nih.gov/pubmed/29773536
      OpenUrlFREE Full Text
    4. ↵
      1. Coelho AR,
      2. Cardoso G,
      3. Brito ME, et al
      . The female athlete Triad/Relative energy deficiency in sports (RED-S). Rev Bras Ginecol Obstet 2021;43:395–402.doi:10.1055/s-0041-1730289pmid:http://www.ncbi.nlm.nih.gov/pubmed/34077990
      OpenUrlPubMed
    5. ↵
      1. Lane AR,
      2. Hackney AC,
      3. Smith-Ryan A, et al
      . Prevalence of low energy availability in competitively trained male endurance athletes. Medicina 2019;55. doi:doi:10.3390/medicina55100665. [Epub ahead of print: 01 Oct 2019].pmid:http://www.ncbi.nlm.nih.gov/pubmed/31581498
    6. ↵
      1. Melin A,
      2. Tornberg Å B,
      3. Skouby S, et al
      . Energy availability and the female athlete triad in elite endurance athletes. Scand J Med Sci Sports 2015;25:610–22.doi:10.1111/sms.12261pmid:http://www.ncbi.nlm.nih.gov/pubmed/24888644
      OpenUrlCrossRefPubMed
    7. ↵
      1. World Health Organization (WHO)
      . The ICD-10 classification of mental and behavioural disorders: clinical descriptions and diagnostic guidelines. Geneva: WHO, 1992.
    8. ↵
      1. APA
      . Diagnostic and statistical manual of mental disorders: DSM-5. Washington, DC: American Psychiatric Association, 2013.
    9. ↵
      1. Lichtenstein MB,
      2. Melin A
      . SPISEFORSTYRRELSER I DANSK ELITEIDRÆT – en afdækning AF risikoen for spiseforstyrrelse, LAV energitilgængelighed, træningsafhængighed OG depression I 15 idrætsgrene. København: Team Danmark, Dansk Idrætsforbund, 2020.
    10. ↵
      1. Muros JJ,
      2. Ávila-Alche Ángela,
      3. Knox E, et al
      . Likelihood of suffering from an eating disorder in a sample of Spanish cyclists and triathletes. J Eat Disord 2020;8:70.doi:10.1186/s40337-020-00350-zpmid:http://www.ncbi.nlm.nih.gov/pubmed/33292590
      OpenUrlPubMed
    11. ↵
      1. Pustivšek S,
      2. Hadžić V,
      3. Dervišević E
      . Risk factors for eating disorders among male adolescent athletes. Zdr Varst 2015;54:58–65.doi:10.1515/sjph-2015-0008pmid:http://www.ncbi.nlm.nih.gov/pubmed/27646623
      OpenUrlPubMed
    12. ↵
      1. Galmiche M,
      2. Déchelotte P,
      3. Lambert G, et al
      . Prevalence of eating disorders over the 2000-2018 period: a systematic literature review. Am J Clin Nutr 2019;109:1402–13.doi:10.1093/ajcn/nqy342pmid:http://www.ncbi.nlm.nih.gov/pubmed/31051507
      OpenUrlCrossRefPubMed
    13. ↵
      1. Hoek HW
      . Review of the worldwide epidemiology of eating disorders. Curr Opin Psychiatry 2016;29:336–9.doi:10.1097/YCO.0000000000000282pmid:http://www.ncbi.nlm.nih.gov/pubmed/27608181
      OpenUrlPubMed
    14. ↵
      1. Keski-Rahkonen A,
      2. Mustelin L
      . Epidemiology of eating disorders in Europe: prevalence, incidence, comorbidity, course, consequences, and risk factors. Curr Opin Psychiatry 2016;29:340–5.doi:10.1097/YCO.0000000000000278pmid:http://www.ncbi.nlm.nih.gov/pubmed/27662598
      OpenUrlPubMed
    15. ↵
      1. Martinsen M,
      2. Sundgot-Borgen J
      . Higher prevalence of eating disorders among adolescent elite athletes than controls. Med Sci Sports Exerc 2013;45:1188–97.doi:10.1249/MSS.0b013e318281a939pmid:http://www.ncbi.nlm.nih.gov/pubmed/23274604
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Kristjánsdóttir H,
      2. Sigurðardóttir P,
      3. Jónsdóttir S, et al
      . Body image concern and eating disorder symptoms among elite Icelandic athletes. Int J Environ Res Public Health 2019;16. doi:doi:10.3390/ijerph16152728. [Epub ahead of print: 31 07 2019].pmid:http://www.ncbi.nlm.nih.gov/pubmed/31370175
    17. ↵
      1. Sundgot-Borgen J,
      2. Torstveit MK
      . Prevalence of eating disorders in elite athletes is higher than in the general population. Clin J Sport Med 2004;14:25–32.doi:10.1097/00042752-200401000-00005pmid:http://www.ncbi.nlm.nih.gov/pubmed/14712163
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Torstveit MK,
      2. Rosenvinge JH,
      3. Sundgot-Borgen J
      . Prevalence of eating disorders and the predictive power of risk models in female elite athletes: a controlled study. Scand J Med Sci Sports 2008;18:108–18.doi:10.1111/j.1600-0838.2007.00657.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/17490455
      OpenUrlPubMedWeb of Science
    19. ↵
      1. Smolak L,
      2. Murnen SK,
      3. Ruble AE
      . Female athletes and eating problems: a meta-analysis. Int J Eat Disord 2000;27:371–80.doi:10.1002/(SICI)1098-108X(200005)27:4<371::AID-EAT1>3.0.CO;2-Ypmid:http://www.ncbi.nlm.nih.gov/pubmed/10744843
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Morgan JF,
      2. Reid F,
      3. Lacey JH
      . The SCOFF questionnaire: assessment of a new screening tool for eating disorders. BMJ 1999;319:1467–8.doi:10.1136/bmj.319.7223.1467pmid:http://www.ncbi.nlm.nih.gov/pubmed/10582927
      OpenUrlFREE Full Text
    21. ↵
      1. Fairburn CG,
      2. Beglin SJ
      . Assessment of eating disorders: interview or self-report questionnaire? Int J Eat Disord 1994;16:363–70.doi:10.1002/1098-108X(199412)16:4<363::AID-EAT2260160405>3.0.CO;2-%23pmid:http://www.ncbi.nlm.nih.gov/pubmed/7866415
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Arnow KD,
      2. Feldman T,
      3. Fichtel E, et al
      . A qualitative analysis of male eating disorder symptoms. Eat Disord 2017;25:297–309.doi:10.1080/10640266.2017.1308729pmid:http://www.ncbi.nlm.nih.gov/pubmed/28394743
      OpenUrlPubMed
    23. ↵
      1. Darcy AM,
      2. Hardy KK,
      3. Crosby RD, et al
      . Factor structure of the eating disorder examination questionnaire (EDE-Q) in male and female college athletes. Body Image 2013;10:399–405.doi:10.1016/j.bodyim.2013.01.008pmid:http://www.ncbi.nlm.nih.gov/pubmed/23453695
      OpenUrlPubMed
    24. ↵
      1. Kampouri D,
      2. Kotopoulea-Nikolaidi M,
      3. Daskou S, et al
      . Prevalence of disordered eating in elite female athletes in team sports in Greece. Eur J Sport Sci 2019;19:1267–75.doi:10.1080/17461391.2019.1587520pmid:http://www.ncbi.nlm.nih.gov/pubmed/30880593
      OpenUrlPubMed
    25. ↵
      1. Creswell J,
      2. Clark P
      . Designing and conducting mixed methods research. 3rd ed. Thousand Oaks, CA: SAGE Publications, Inc, 2018.
    26. ↵
      1. Tong A,
      2. Sainsbury P,
      3. Craig J
      . Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care 2007;19:349–57.doi:10.1093/intqhc/mzm042pmid:http://www.ncbi.nlm.nih.gov/pubmed/17872937
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Lichtenstein MB,
      2. Melin AK,
      3. Szabo A, et al
      . The prevalence of exercise addiction symptoms in a sample of national level elite athletes. Front Sports Act Living 2021;3:635418.doi:10.3389/fspor.2021.635418pmid:http://www.ncbi.nlm.nih.gov/pubmed/34179773
      OpenUrlPubMed
    28. ↵
      1. Mond JM,
      2. Hay PJ,
      3. Rodgers B, et al
      . Validity of the eating disorder examination questionnaire (EDE-Q) in screening for eating disorders in community samples. Behav Res Ther 2004;42:551–67.doi:10.1016/S0005-7967(03)00161-Xpmid:http://www.ncbi.nlm.nih.gov/pubmed/15033501
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Müller A,
      2. Loeber S,
      3. Söchtig J, et al
      . Risk for exercise dependence, eating disorder pathology, alcohol use disorder and addictive behaviors among clients of fitness centers. J Behav Addict 2015;4:273–80.doi:10.1556/2006.4.2015.044pmid:http://www.ncbi.nlm.nih.gov/pubmed/26690622
      OpenUrlPubMed
    30. ↵
      1. Schaefer LM,
      2. Smith KE,
      3. Leonard R, et al
      . Identifying a male clinical cutoff on the eating disorder Examination-Questionnaire (EDE-Q). Int J Eat Disord 2018;51:1357–60.doi:10.1002/eat.22972pmid:http://www.ncbi.nlm.nih.gov/pubmed/30480321
      OpenUrlPubMed
    31. ↵
      1. Botella J,
      2. Sepúlveda AR,
      3. Huang H, et al
      . A meta-analysis of the diagnostic accuracy of the SCOFF. Span J Psychol 2013;16:E92.doi:10.1017/sjp.2013.92pmid:http://www.ncbi.nlm.nih.gov/pubmed/24230955
      OpenUrlPubMed
    32. ↵
      1. Lichtenstein MB,
      2. Hemmingsen SD,
      3. Støving RK
      . Identification of eating disorder symptoms in Danish adolescents with the SCOFF questionnaire. Nord J Psychiatry 2017;71:340–7.doi:10.1080/08039488.2017.1300322pmid:http://www.ncbi.nlm.nih.gov/pubmed/28290749
      OpenUrlPubMed
    33. ↵
      1. Thomas DR
      . A general inductive approach for analyzing qualitative evaluation data. Am J Eval 2006;27:237–46.doi:10.1177/1098214005283748
      OpenUrl
    34. ↵
      1. Stranberg M,
      2. Slager E,
      3. Spital D, et al
      . Athlete-Specific treatment for eating disorders: initial findings from the Walden goals program. J Acad Nutr Diet 2020;120:183–92.doi:10.1016/j.jand.2019.07.019pmid:http://www.ncbi.nlm.nih.gov/pubmed/31590997
      OpenUrlPubMed
    35. ↵
      1. Fewell LK,
      2. Nickols R,
      3. Schlitzer Tierney A, et al
      . Eating disorders in sport: comparing eating disorder symptomatology in athletes and non-athletes during intensive eating disorder treatment. J Clin Sport Psychol 2018;12:578–94.doi:10.1123/jcsp.2018-0046
      OpenUrl
    36. ↵
      1. de Bruin APK,
      2. Oudejans RRD,
      3. Bakker FC, et al
      . Contextual body image and athletes' disordered eating: the contribution of athletic body image to disordered eating in high performance women athletes. Eur Eat Disord Rev 2011;19:201–15.doi:10.1002/erv.1112pmid:http://www.ncbi.nlm.nih.gov/pubmed/21584913
      OpenUrlPubMed
    37. ↵
      1. Krane V,
      2. Choi PYL,
      3. Baird SM, et al
      . Living the paradox: female athletes Negotiate femininity and muscularity. Sex Roles 2004;50:315–29.doi:10.1023/B:SERS.0000018888.48437.4f
      OpenUrlCrossRefWeb of Science
    38. ↵
      1. Mosewich AD,
      2. Vangool AB,
      3. Kowalski KC
      . And field athletes' meanings of muscularity. Journal of Applied Sport Psychology 2009;21:99–115.
      OpenUrlCrossRef
    39. ↵
      1. Carson TL,
      2. Tournat T,
      3. Sonneville K, et al
      . Cultural and environmental associations with body image, diet and well-being in NCAA di female distance runners: a qualitative analysis. Br J Sports Med 2021;55:433–7.doi:10.1136/bjsports-2020-102559pmid:http://www.ncbi.nlm.nih.gov/pubmed/33139255
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. Stoyel H,
      2. Delderfield R,
      3. Shanmuganathan-Felton V, et al
      . A qualitative exploration of sport and social pressures on elite athletes in relation to disordered eating. Front Psychol 2021;12:633490.doi:10.3389/fpsyg.2021.633490pmid:http://www.ncbi.nlm.nih.gov/pubmed/33967900
      OpenUrlPubMed
    41. ↵
      1. Byrne S,
      2. McLean N
      . Elite athletes: effects of the pressure to be thin. J Sci Med Sport 2002;5:80–94.doi:10.1016/S1440-2440(02)80029-9pmid:http://www.ncbi.nlm.nih.gov/pubmed/12188089
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Joy E,
      2. Kussman A,
      3. Nattiv A
      . 2016 update on eating disorders in athletes: a comprehensive narrative review with a focus on clinical assessment and management. Br J Sports Med 2016;50:154–62.doi:10.1136/bjsports-2015-095735pmid:http://www.ncbi.nlm.nih.gov/pubmed/26782763
      OpenUrlAbstract/FREE Full Text
    43. ↵
      1. Bonci CM,
      2. Bonci LJ,
      3. Granger LR, et al
      . National athletic trainers' association position statement: preventing, detecting, and managing disordered eating in athletes. J Athl Train 2008;43:80–108.doi:10.4085/1062-6050-43.1.80pmid:http://www.ncbi.nlm.nih.gov/pubmed/18335017
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. National Eating Disorder Association (NEDA)
      . Coach and athletic trainer toolkit.
    45. ↵
      1. Arfken CL,
      2. Balon R
      . Declining participation in research studies. Psychother Psychosom 2011;80:325–8.doi:10.1159/000324795pmid:http://www.ncbi.nlm.nih.gov/pubmed/21829043
      OpenUrlCrossRefPubMed
    46. ↵
      1. Lincoln YS,
      2. Guba EG,
      3. Pilotta JJ
      . Naturalistic inquiry. 9. Beverly Hills, CA: Sage, 1985: 438–9.doi:10.1016/0147-1767(85)90062-8
      OpenUrl
    47. ↵
      1. Malterud K
      . Qualitative research: standards, challenges, and guidelines. Lancet 2001;358:483–8.doi:10.1016/S0140-6736(01)05627-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/11513933
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      1. O'Brien BC,
      2. Harris IB,
      3. Beckman TJ, et al
      . Standards for reporting qualitative research: a synthesis of recommendations. Acad Med 2014;89:1245–51.doi:10.1097/ACM.0000000000000388pmid:http://www.ncbi.nlm.nih.gov/pubmed/24979285
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors The authorships of this manuscript follow the four criteria: substantial contributions to the conception or design of the work; or the acquisition, analysis or interpretation of data for the work; AND drafting the work or revising it critically for important intellectual content; AND final approval of the version to be published; AND agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. MBL is the guarantor.

    • Funding The project was partly funded by Team Denmark and The Danish Sport Federation (DIF).

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.