Eating behavior and body image in athletes

Abstracts

The sports environment can enhance social and cultural pressure towards a lean body. The close relationship between corporal image and performance makes the female athletes a group particularly vulnerable to eating disorders and weight-control practices. The aim of the present work is to evaluate the presence of behaviors suggestive of eating disorders, shifts in body perception, and menstrual dysfunctions in female athletes of the Federal University of Rio de Janeiro School of Physical Education (EEFD-UFRJ). Twelve female athletes (20 ± 2 years), who had been training for 4.6 ± 2.3 years, a total of 13.8 ± 2.9 hours per week, were examined. The results of the athletes and of 32 young non-athletes (15 ± 1.4 years) were compared. The Eating Attitudes Test (EAT-26) was used to evaluate weight-control and eating attitudes and behaviors. The Bulimic Investigatory Test Edimburgh (BITE) was used to identify suggestive behaviors of bulimia nervosa, and also to obtain data about cognitive and behavioral aspects of this disorder. The Body Shape Questionnaire (BSQ) was used to verify dissatisfaction with body image. The three self-applied questionnaires were used in their Portuguese versions. The athletes presented body mass of 59 ± 7.3 kg, height of 1.65 ± 0.03 m, percentage of body fat (%F) of 23.1 ± 4.1% and body mass index (BMI) of 21.6 ± 2.3 kg/m². The entire sample had a %F compatible for age and sex, although 33% of the athletes presented some body image dissatisfaction. The BITE evidenced that 16.6% of the athletes presented a non-usual dietary pattern according to symptoms, but with no significance in terms of severity. These athletes are included among those who are not happy with their body. No athlete had a positive EAT-26 results. The findings of this study suggest the presence of symptoms that do not actually characterize bulimia or anorexia nervosa, but they point to the need of a careful surveillance and assessment of eating disorders and their precursors among athletes, as a way of prevention.

Eating disorders; Athletes; Amenorrhea; Body composition

O ambiente esportivo pode ser um meio ampliador de pressões socioculturais motivadas pelo ideal de corpo magro. A presente investigação teve o objetivo de verificar a presença de comportamentos sugestivos de transtornos do comportamento alimentar (TCAs), de alterações na imagem corporal e de disfunções menstruais em atletas da EEFD-UFRJ. Foram avaliadas 12 atletas do sexo feminino (20 ± 2,0 anos) com 4,6 ± 2,3 anos de treinamento (13,8 ± 2,9h/semana) e seus resultados foram comparados com os de 32 jovens não-atletas (15,0 ± 1,4 ano). A avaliação de atitudes e comportamentos relacionados com a alimentação e o controle de peso foi realizada pelo Eating Attitudes Test (EAT-26), a existência de comportamentos sugestivos de bulimia nervosa pelo Bulimic Investigatory Test Edimburgh (BITE) e o grau de insatisfação com a imagem corporal pelo Body Shape Questionnaire (BSQ). Esses são instrumentos de auto-aplicação e foram usados na versão em português. As atletas apresentavam massa corporal total de 59 ± 7,3kg, estatura de 1,65 ± 0,03m, percentual de gordura corporal (%G) 23,1 ± 4,1% e índice de massa corporal (IMC) de 21,6 ± 2,3kg/m². O BSQ apontou que 33% das atletas apresentam leve distorção da imagem corporal mesmo estando com valores para gordura corporal dentro de padrões esperados para a idade e sexo. O BITE apontou 16,6% das atletas com padrão alimentar não-usual na escala de sintomas, sem que estas apresentassem expressão na escala de gravidade. As atletas com padrão alimentar não-usual estão incluídas no total sugestivo de distorção de imagem corporal. O EAT-26 exibiu resultado negativo para todas as atletas. Os resultados encontrados na análise sugerem a presença de sintomas que não caracterizam, precisamente, a bulimia nervosa ou a anorexia nervosa, mas ressaltam a necessidade de avaliação e vigilância mais criteriosas da existência de TCA e seus precursores no ambiente esportivo, como forma de prevenção.

Transtorno do comportamento alimentar; Atletas; Amenorréia; Composição corporal

El ambiente deportivo puede ser un medio amplificador de presiones socioculturales motivadas por el ideal del cuerpo magro. La presente investigación tiene por objeto verificar la presencia de comportamientos sugestivos de trastornos del comportamiento alimenticio (TCAs), de alteraciones de la imagen corporal y de disfunciones menstruales en atletas de la EEFD-UFRJ. Fueron evaluadas 12 atletas de sexo femenino (20 ± 2,0 años) con 4,6 ± 2,3 años de entrenamiento (13,8 ± 2,9 horas/semana) y sus resultados serán comparados con 32 jovenes no atletas (15,0 ± 1,4 años). La evaluación de actitudes y comportamientos relacionados con la nutrición y el control del peso fue evaluado un test, Eating Attitudes Test (Test de Actitudes Nutricionales EAT-26), la existencia de los comportamientos sugestivos de bulimia nerviosa por el test, Bulimic Investigatory Test Edimburgh (Test de Investigación de Bulimia de Edimburgo, BITE) y el grado de insatisfacción con la imagen corporal por el test; Body Shape Questionnaire (Cuestionario de la forma Corporal BSQ). Estos son instrumentos de autoaplicación usados en una versión portuguesa. Las atletas presentaron una masa corporal total de 59 ± 7,0kg, una estatura de 1,65 ± 0,03m, un porcentual de grasa corporal (%G) 23 ± 4,0% y un índice de massa corporal (IMC) de 21,6 ± 2,0kg/m². El test BSQ apuntó que un 33% de las atletas presentaron una leve distorción de la imagen corporal mismo estando con valores para grasa corporal dentro de los patrones esperados para la edad y el sexo. El test BITE apuntó que el 16,6% de las atletas con un padrón alimenticio no usual en la escala de síntomas, sin que estas presentasen expresiones en la escala de gravedad. Las atletas con patrón alimenticio no usual están incluídas en el total sugestivo de distorsión de la imagen corporal. El test EAT-26 mostró un resultado negativo para todas las atletas. Los resultados encontrados en el análisis sugieren la presencia de síntomas que no se caracterizan precisamente, por una bulimia nerviosa o una anorexia nerviosa, relatan la necesidad de una evaluación y una vigilancia mas criteriosas de la existencia de TCA y de sus precursores en el ambiente deportivo, como forma de prevención.

Transtorno del comportamiento alimenticio; Atletas; Amenorrea; Composición corporal

ORIGINAL ARTICLE

Eating behavior and body image in athletes^*

Comportamiento alimentario e imagen corporal de los atletas

Fátima Palha de Oliveira^I; Maria Lúcia Magalhães Bosi^II; Patrícia dos Santos Vigário^III; Renata da Silva Vieira^III

^IProfessor Adjunto IV – EEFD-UFRJ

^IIProfessor Adjunto IV – NESC-HU-UFRJ

^IIIAcadêmica de Educação Física – UFRJ

^{Correspondence}

ABSTRACT

The sports environment can enhance social and cultural pressure towards a lean body. The close relationship between corporal image and performance makes the female athletes a group particularly vulnerable to eating disorders and weight-control practices. The aim of the present work is to evaluate the presence of behaviors suggestive of eating disorders, shifts in body perception, and menstrual dysfunctions in female athletes of the Federal University of Rio de Janeiro School of Physical Education (EEFD-UFRJ). Twelve female athletes (20 ± 2 years), who had been training for 4.6 ± 2.3 years, a total of 13.8 ± 2.9 hours per week, were examined. The results of the athletes and of 32 young non-athletes (15 ± 1.4 years) were compared. The Eating Attitudes Test (EAT-26) was used to evaluate weight-control and eating attitudes and behaviors. The Bulimic Investigatory Test Edimburgh (BITE) was used to identify suggestive behaviors of bulimia nervosa, and also to obtain data about cognitive and behavioral aspects of this disorder. The Body Shape Questionnaire (BSQ) was used to verify dissatisfaction with body image. The three self-applied questionnaires were used in their Portuguese versions. The athletes presented body mass of 59 ± 7.3 kg, height of 1.65 ± 0.03 m, percentage of body fat (%F) of 23.1 ± 4.1% and body mass index (BMI) of 21.6 ± 2.3 kg/m². The entire sample had a %F compatible for age and sex, although 33% of the athletes presented some body image dissatisfaction. The BITE evidenced that 16.6% of the athletes presented a non-usual dietary pattern according to symptoms, but with no significance in terms of severity. These athletes are included among those who are not happy with their body. No athlete had a positive EAT-26 results. The findings of this study suggest the presence of symptoms that do not actually characterize bulimia or anorexia nervosa, but they point to the need of a careful surveillance and assessment of eating disorders and their precursors among athletes, as a way of prevention.

Key words: Eating disorders. Athletes. Amenorrhea. Body composition.

RESUMEN

El ambiente deportivo puede ser un medio amplificador de presiones socioculturales motivadas por el ideal del cuerpo magro. La presente investigación tiene por objeto verificar la presencia de comportamientos sugestivos de trastornos del comportamiento alimenticio (TCAs), de alteraciones de la imagen corporal y de disfunciones menstruales en atletas de la EEFD-UFRJ. Fueron evaluadas 12 atletas de sexo femenino (20 ± 2,0 años) con 4,6 ± 2,3 años de entrenamiento (13,8 ± 2,9 horas/semana) y sus resultados serán comparados con 32 jovenes no atletas (15,0 ± 1,4 años). La evaluación de actitudes y comportamientos relacionados con la nutrición y el control del peso fue evaluado un test, Eating Attitudes Test (Test de Actitudes Nutricionales EAT-26), la existencia de los comportamientos sugestivos de bulimia nerviosa por el test, Bulimic Investigatory Test Edimburgh (Test de Investigación de Bulimia de Edimburgo, BITE) y el grado de insatisfacción con la imagen corporal por el test; Body Shape Questionnaire (Cuestionario de la forma Corporal BSQ). Estos son instrumentos de autoaplicación usados en una versión portuguesa. Las atletas presentaron una masa corporal total de 59 ± 7,0kg, una estatura de 1,65 ± 0,03m, un porcentual de grasa corporal (%G) 23 ± 4,0% y un índice de massa corporal (IMC) de 21,6 ± 2,0kg/m². El test BSQ apuntó que un 33% de las atletas presentaron una leve distorción de la imagen corporal mismo estando con valores para grasa corporal dentro de los patrones esperados para la edad y el sexo. El test BITE apuntó que el 16,6% de las atletas con un padrón alimenticio no usual en la escala de síntomas, sin que estas presentasen expresiones en la escala de gravedad. Las atletas con patrón alimenticio no usual están incluídas en el total sugestivo de distorsión de la imagen corporal. El test EAT-26 mostró un resultado negativo para todas las atletas. Los resultados encontrados en el análisis sugieren la presencia de síntomas que no se caracterizan precisamente, por una bulimia nerviosa o una anorexia nerviosa, relatan la necesidad de una evaluación y una vigilancia mas criteriosas de la existencia de TCA y de sus precursores en el ambiente deportivo, como forma de prevención.

Palabras clave: Transtorno del comportamiento alimenticio. Atletas. Amenorrea. Composición corporal.

INTRODUCTION

The model of beauty society imposes a lean body, without considering health-related aspects and the different physical features of the population. Such distorted beauty standard makes an increasing number of women to go on weight-control diets, to over-practice physical exercises, and to a wide use of laxatives, diuretics and anorectic drugs¹. These are considered precursory behaviors of eating disorders – which include anorexia and bulimia nervosa, and the so-called non-specific eating disorders, among which the periodic eating compulsion disorder stands out. The detection of partial syndromes of such diseases is of high importance for prevention, as they currently occur more often than the disease itself². Young women, because they are more vulnerable to pressures of society, culture and economics on esthetic standards, are the group at higher risk for eating disorders³. Eating disorders are from multiple causes, including genetic, environmental and behavioral factors. People with anorexia nervosa or bulimia nervosa have, in common, an excessive concern with weight and diet, are dissatisfied and have a distorted perception of their body image, and typically are resistant to treatment. Lethality rate from anorexia nervosa ranges between 5% and 20%, which justifies the need for prevention.

Anorexia nervosa is characterized by a self-imposed dietary restriction, in order to achieve a marked weight loss. The goal of having a thin body is guided by behaviors that include a reduction of "fattening" food intake and/or the use of self-induced purgation methods, or the compulsive practice of physical activities. As the most common symptoms, the patients keep their body weight lower than 85% of what is considered suitable for their height and age, they present a morbid fear of getting fat, changes in their perception of their body, menstrual disorders, bone demineralization, loss of muscle mass and body fat, digestive disorders, cardiac arrhythmia, dehydration, intolerance to cold, thin and weak hair, among other problems⁵.

A significant energy restriction, with a short-time loss of weight, lead to a decrease of the basal metabolic rate, and impairment of musculoskeletal, cardiovascular, endocrine, and thermoregulatory functions, and difficulties in attention and concentration^6,7.

Among the psychological symptoms, even though there is not a definite profile, the patient often presents anxiety, depression, low self-esteem, irritability, intolerance to frustrations, and mood unbalance, with an impact in his/her social, affective and professional life.

The main characteristic of bulimia nervosa is a compulsive overeating, followed by some sort of purgation. In this disease, the patient eats an amount of food much higher than typically would be eaten by most people over the same period of time and in similar conditions, associated to a loss-of-control sensation. This sensation leads the patient to extreme attitudes, such as self-inducing vomiting, or laxative and/or diuretic abuse, excessive workout, extreme diets or even fasting, to prevent consequences of the over-intake of food. The main effects from such practices are loss of fluids and electrolytes during purgation (dehydration), acid-base unbalance, electrolyte unbalance, cardiac arrhythmias, decrease of serum potassium levels (the main cation accountable for muscle contraction), gastrointestinal disorders, increase of the parotid gland, erosion and loss of teeth enamel^1,8. Bulimia nervosa is also related with an excessive concern about weight and body shape, and the patient tends to judge him/herself from these variables. It is the excessive fear of getting fat that differentiates the bulimic from anyone who may eventually overeat. Differently from the anorectic individual, whose extreme leanness makes the problem evident, the bulimic person has normal weight or is slightly overweight, and because of that the disease may be hidden for many years^9,10. The prevalence of bulimia nervosa ranges between 1% and 4% in the adolescent and young adult female population².

The compulsive-eating disorder is characterize by the binge eating, by which the patient intakes huge amounts of food over a short period of time, until he/she feels uncomfortable. The patient typically eats alone, and afterward feels guilty and depressed¹¹.

In athletes, even though the prevalence of anorexia nervosa and bulimia nervosa is not really known, particularly in Brazil, studies have found an increased frequency (from 15 to 62%), particularly in some sports^5,12. The close relationship between corporal image and physical performance makes athletes a group particularly vulnerable for such disorders, considering the importance given to weight control^13-15. Recent studies mention the influence of coaches, sponsors and families, through their comments on the athletes' weight and shape, as a powerful element for triggering abnormal eating disorders. It is also known that the adoption of a restrictive diet at an early age, particularly if without professional supervision, may trigger an eating disorder, causing major health hazards and thus, impairing athletic performance^16,17.

Sports that require low body weight and overestimate aesthetics, using it as a criterion for obtaining good results in competitions – such as in artistic gymnastics, synchronized swimming, running, and dance – have been pointed out by investigations carried out in this area as presenting the highest incidence of eating disorders and of precursory eating disorder behaviors^18-23. West²⁴, investigating 42 university gymnastics athletes, found that 62% of them used, at least, one type of inappropriate dietary practice. Dietary restriction, which is not recommended to non-athlete young people, is extremely dangerous to athletes, as it may lead to a caloric deficit and electrolytic unbalancing, causing malnutrition and compromising physical fitness. A low-calorie diet is also associated to low pulse and frequency of the luteinizing hormone and triiodothyronine^25-28. These hormones act on the menstrual cycle control, and, if altered, they cause secondary amenorrhea and oligomenorrhea, typical eating-disorder symptoms.

The interrelationship and, in many cases, coexistence of eating disorders, amenorrhea and osteoporosis form the female athlete triad^29,30. The development of abnormal eating patterns may lead to menstrual dysfunction and to osteoporosis; therefore, the symptom control from such abnormalities is mandatory. Each component of this triad may interfere in physical performance, and be a morbidity or mortality cause.

The general population, and particularly the population focused in our study (athletes and their coaches), must be informed about risk behaviors that preceded the development of eating disorders, for an early diagnosis and control to prevent severe eating disorder cases.

The scope of this investigation was to assess the occurrence of behaviors suggestive of eating disorders, changes in body image, and presence of menstrual dysfunctions in female athletes of the Federal University of Rio de Janeiro School of Physical Education (EEFD-UFRJ).

MATERIAL AND METHODS

A cross-sectional study was carried out with a target-population considered to be of higher risk for eating disorders: female young athletes.

The sample included 12 young women (20 ± 2 years) at childbearing age (age at menarche: 12.2 ± 1.5 years), affiliated to clubs of different sports modalities in Rio de Janeiro, and students of EEFD-UFRJ. The assessed group of athletes included six from collective sports (handball, soccer, volleyball), three from individual sports (triathlon, swimming, dance), and three from fighting sports (karate, jiu-jitsu and kick-boxing), who reported and average training period of 4.6 ± 2.3 years, and a total of 13.8 ± 2.9 hours of practice a week. A control group of 32 non-athlete post-menarche adolescents (15.0 ± 1.4 years) was evaluated according to the same protocol of the athletes.

The presence of menstrual dysfunctions, diagnosis of osteoporosis, and recent bone injuries, and the training routine were self-reported in a questionnaire supervised by the authors of this study. This questionnaire was especially developed for this investigation, being previously validated by eight UFRJ experts. The four adopted questionnaires were applied at EEFD-UFRJ's Exercise Physiology Lab.

After responding to the four questionnaires, the athletes were submitted to anthropometrical evaluation³¹ for assessment of body composition, and the following measurements were made: skinfold (chest, triceps, biceps, subscapular, suprailiac, abdominal, medium axillary, thigh and leg – CESCORF, 0,1 mm); body girth (relaxed and contracted biceps, thigh, leg, abdomen and hip flexible metallic tape – CARDIOMED, 1 cm); bone breadth (wrist, knee and elbow caliper, 1 mm); height (stadiometer, 1 mm) and total body mass (electronic scale, 50 g). From the measures found, the different parameters were calculated, and the present analysis includes the proportion of body fat (%F)³², obtained from body density estimates³³, lean body mass [total body mass fat body mass, kg], and the body mass index [BMI = total body mass/height², kg/m²]³⁴. Body fat estimate for the control group was made from body density estimates calculated from Jackson et al. model³⁵.

To assess typical attitudes and behaviors of patients with anorexia nervosa and to rank the severity of the eating disorders, the Eating Attitudes Test (EAT-26)³⁶ was adopted. EAT-26 is a self-report questionnaire internationally used to identify and assess abnormal eating standards, and is also useful to follow up progression of medical cases. The questionnaire includes 26 questions, with six response options: always, very often, often, sometimes, seldom, and never, and three aspects are analyzed:

Aspect 1: Diet – the first three items address pathologic high-calorie food refusal and concerns about fitness;

Aspect 2: Bulimia nervosa – the six questions that follow are to reflect respondent's thoughts on food and bulimic attitudes;

Aspect 3: Oral control – the last seven items address self-control regarding food and acknowledge social pressure in the environment towards weight gaining.

The evaluation of EAT-26 responses is done by rating three to the most extreme responses ("always" or "never"), two to the second most extreme responses, and one to the third most extreme responses; the other responses are not scored. The scores from each EAT-26 question are added and computed for each person assessed. If the total score is higher than 21, EAT-26 is considered to be positive (EAT-26⁺), thus confirming the presence of abnormal dietary attitudes, and risk for the development of eating disorders.

The authors who developed EAT-26 used a 20-score cut-off point. In our study we used a cut-off of 21 points, as its sensitivity and specificity are highly satisfactory^37,38, and it is being used by other Brazilian authors^4,39, allowing results to be compared.

To assess the presence and intensity of bulimic behaviors, the Bulimic Investigatory Test Edinburgh – BITE⁴⁰ was used. It is a self-response questionnaire with 33 questions and two sub-scales, one for SYMPTOMS and the other for their SEVERITY. These scales are described below, according to Cordás⁴¹.

There are three scoring groups for the SYMPTOM scale:

Ø High (a score of 20 or higher): this is considered a high scoring, with presence of compulsive eating behavior and likely to meet diagnostic criteria for bulimia nervosa according to DSM-III-R².

Ø Medium (scores between 10 and 19): suggests a non-usual eating pattern, and not all bulimia criteria are typically present. People with scores ranging from 15 to 19 may be a subclinical group of compulsive eaters, early-stage bulimics, or bulimics in remission - they should undergo a diagnostic interview with specialists.

Ø Low (a score below 10): within normal range.

BITE's SEVERITY scale includes questions 6, 7 and 27 of the questionnaire, and measures the severity of the compulsive behavior according to the frequency of attitudes; it is to be analyzed in the cases that score 10 or higher in the symptom scale, and presents three stages:

Ø High (a score > 10): high degree of severity, and may point to the presence of psychogenic vomiting or laxative abuse without a compulsive behavior, and the patient should be referred to a diagnostic interview with specialists.

Ø Moderate (a score between 5 and 9): are considered clinically significant, and patients should be referred to a diagnostic interview with specialists.

Ø Low (score up to 5): a clinically non-significant result.

To assess the degree of dissatisfaction with body image, the Body Shape Questionnaire – BSQ (validated by Cooper et al.)⁴² was used, which measures the concern one has with the shape of the body, self-depreciation due to physical aspect and to the feeling of being fat. According to Cordás⁴³, the questionnaire differentiates two aspects of body image: the accuracy of body size estimates, and the feelings towards the body (dissatisfaction or depreciation of one's fitness). The questionnaire includes 34 questions, with six response alternatives: 1) never, 2) seldom, 3) sometimes, 4) often, 5) very often, 6) always. According to the response marked, the considered score for the question is the number of that response (for instance: the score for never is 1). The scores are summed and the total is considered for each individual. The rating of results is according to the total number of points, and reflects the level of concern for body image. A total score lower or equal to 80 is considered normal, i.e., no distortion of body image. A score between 81 and 110 means mild body image distortion; between 111 and 140, moderate distortion; and over 140, severe body image distortion.

EAT-26, BITE and BSQ were used in their Portuguese versions⁴³. These are self-report tools used in internal medicine for triage and identification of eating disorders and other psychiatric co-morbidities, and study participants did not go to a diagnostic interview (APA)².

The Ethics Committee of the Collective Health Study Center, Clementino Fraga Filho Hospital, approved this study and all participants signed an informed consent from agreeing to the procedures to be adopted and authorizing the scientific use of the study results. Participants' anonymity and privacy were kept in the study.

Calculation of parameters, descriptive data statistics and comparison (non-paired Student's t test) of the variables and eating disorder questionnaire responses between the groups were performed with the use of software Statistica for Windows (5.1, 1998), with a significance level of p < 0.05.

RESULTS

Menstrual dysfunction

Primary amenorrhea, which is the absence of menarche at age 16, may be related to the practice of exhaustible physical exercises¹². The athletes of our investigation, even though they did not present primary amenorrhea, had their menarche at a significantly older age then the control group, and both groups were within the average menarche standard of Brazilian women⁴⁴ (table 1).

There were a significant number of athletes (50%) and controls (25%) with dysmenorrhea (menstruation accompanied by severe colic). The relationship between physical exercises and dysmenorrhea is not well established in the literature. In some athletes, dysmenorrhea symptoms remain with training, in others, the symptoms fade^45,46.

In our investigation, 75% of the athletes mentioned a change in performance when they compete at the menstrual period. This figure includes all athletes with dysmenorrhea and others without it. There is no consensus among investigators as to the possibility of performance impairment of athletes at competitions during the menstrual period. Some authors^45,46 did not find effects from menses on motor performance, while others report that performance improves in the post-menstrual phase and worsens during menses or on the three previous days^37-49.

Only one athlete (soccer, 8.3% of the sample) informed having had secondary amenorrhea (interruption of menses for three or more months)¹¹ over the last two years of training. In the control group, a case of secondary amenorrhea (3.0% of the sample) was also identified. In the literature, a prevalence of 2 to 5% of sedentary women with amenorrhea has been registered, whereas for athletes, depending on their sports, it may range from 10 to 66%⁵⁰.

Osteoporosis: The athletes stated in the questionnaire that they did not have a recent diagnosis of osteoporosis, nor suffered bone injuries (over the past two years). These statements may indicate absence of osteoporosis in this sample. We stress, however, that for a definite diagnosis a bone densitometry test is required. To overcome this limitation of our investigation, the clinic of orthopedics of UFRJ's University Hospital was involved in the study and is preparing the necessary equipment for the test and diagnosis, which will be done in further investigations with athletes of national teams.

Body composition

The mean total body mass of the athletes, in kg, was higher than of controls, but this was not statistically significant (table 1). This means that the control group, in spite of being younger, had a body mass similar to athletes'. The difference between the groups was evident when the analysis of the total body mass was broke down in its different components. Athletes presented lower body fat (%F) and higher lean mass compared to non-athletes (table 1). These results were expected as response to the training athletes are submitted to.

Some athletes had body fat proportion (%F) higher than expected for athletes of their category, but below the proportion of non-athlete women (25.5%)^5,51. This may be due to the fact that the sample, in spite of not including high-level athletes, comprised young women who have been training for 4.6 ± 2.3 years, and thus have their organic response from energy lipid metabolism adapted to training. Such adjustment has a strong impact on the athletes' body composition, and might have determined a significantly lower %F for the group of athletes from the school of physical education EEFD compared to the control group, which includes sedentary young women⁵² (table 1).

EAT-26 results

Table 2 presents the mean results and standard deviation of both athlete and control groups for the EAT-26, BITE and BSQ, and a comparison among them. One sees from the mean results of both groups on the three tools evaluated, that they are within normal standards for eating disorders. In spite of athletes presenting mean values a little higher than controls, the difference was not statistically significant.

The analysis of individual EAT-26 results showed lack of pathological or at-risk attitudes for the development of anorexia nervosa among the investigated athletes, differently of what the literature points out to athletes^11-12,53,54. Such good result for anorexia nervosa was not replicated in the control group, which presented 15.6% positive results for EAT-26 (table 3). Nunes et al.⁵⁵ found similar results for non-athlete adolescents from Rio Grande do Sul (positive EAT-26 in 13.7%).

BITE results

In our study, 16.7% of athletes and 21.8% of non-athlete adolescents scored between 10 and 19 on the BITE symptom scale. These results express a non-typical eating pattern, showing an at-risk eating behavior (table 3). The BITE severity scale showed that, in spite of athletes presenting a non-typical eating pattern, the result is not clinically significant for bulimia. Among the controls who presented non-typical eating standard on the BITE symptom scale, 42.8% had clinically significant results for bulimia nervosa (table 3).

BSQ results

The analysis of BSQ showed a mean average result higher for athletes compared to controls, but it was not statistically significant (table 2). In a relative analysis, distortion of body image is more worrisome in non-athletes than in athletes (table 3).

Our investigation revealed also an important practice of abnormal eating behaviors in both groups. Such habits may reflect the development of partial eating disorder syndromes, and are characterized as overtly abnormal eating and at-risk behaviors⁴. Table 4 presents some of BITE questions where one can see the concern with weight control and practice of dietary restrictions, particularly fasting, which are conflicting with physical activities.

DISCUSSION

According to the literature, the prevalence of the female athlete triad ranges from 10 to 66% among elite athletes, particularly those whose sports modalities require low body mass (runners, gymnasts, dancers)^11,30,56. In the sample of our investigation, we found only one case of amenorrhea (without a diagnosis of osteoporosis). This result may be due to the fact that athletes did not practice sport modalities where prevalence of this syndrome is higher, and our subjects were not elite athletes. Considering that recent studies have reported a strong correlation between amenorrhea and eating disorders and with low bone density^7,16,23, it is to be stressed the need for more detailed clinical investigations on the presence of amenorrhea in athletes, particularly elite ones.

The presence of eating disorders and inadequate weight-loss practices (impact diets, vomiting induction, fasting, use of laxatives, compulsive exercising, etc.) combined to dissatisfaction of body image has been pointed out in the literature. Studies have found that, even with body mass within normal population standards, it is significant the number of young women unhappy with their weight, and that do inappropriate practices to reduce body mass⁵⁷. In Porto Alegre, Nunes et al.⁵⁸, in a populational study with young women aged 12 to 19, found that perception of body weight (feeling fat) was determinant for abnormal eating behaviors, even among those whose with normal BMI. Our investigation confirms these findings, as, in spite of the healthy standards for total body mass, %F and BMI^5,51 in both the athlete and control groups, the presence of at-risk behaviors for eating disorders was found in EAT-26 (for non-athletes only), BITE, and dissatisfaction with body image in the BSQ (table 3). BITE scores of athletes and controls, even being of moderate intensity and severity, are worrisome, serving as a warning about the possibility of progression for severe behaviors and the development of eating disorders^59-61.

Estimates of eating disorder among athletes range between 15 and 62%, being more marked in esthetic sports, such as dance, bodybuilding, artistic skating and gymnastics^52,53. Sundgot-Borgen⁵³ observed a higher prevalence of eating disorder in athletes (18%) than in non-athletes (5%). Assessing such prevalence among different sports modalities, the author found that the presence of eating disorders and the use of inadequate weight control was significantly higher among bodybuilders (34%) and those who practice sports that require low body mass (27%). For technical (13%) or team (11%) sports, the prevalence found was lower, closer to the controls' (7%).

The results from our study showed that the frequency of at-risk eating behaviors pointed out at BITE and EAT-26 was more significant among controls than in EEFD's athlete students. Literature² points out that anorectic behaviors begin at age 15, and bulimic at age 19. Thus, not-surprisingly the control group, in spite of being younger, presents a higher proportion of abnormal results than the athletes group. From our point of view, these results are supported by the fact that the athletes under assessment practiced sports considered to be of lower risks for eating disorders. The fact that they are not elite athletes (members of national teams) may also add to the result, as they have less training and less stress level, factors that could be related to the presence of eating disorders. However, Weight and Noakes⁶⁴ did not find higher prevalence of eating disorders in athletes compared to controls either, in spite of having analyzed endurance runners, a sports modality where other authors did find a high prevalence of eating disorders. Thus, we believe it is necessary to epidemiological studies to be carried out, considering the different sports modalities, training levels, and including males, for one to have an overall perspective of the eating behavior impact among Brazilian athletes.

CONCLUSION

Even though, at the moment, there is no sign of anorexia or bulimia nervosa among the investigated athletes, the results from our study show the presence of worriment about body weight, combined with a mild distortion of body image and the practice of dietary restrictions. Such practices are called partial eating disorder syndromes, and when associated to systematic and exhaustible physical exercises, they become risk factors, and their early detection may prevent progression to severe eating disorders.

This was the first of a series of multidisciplinary investigations to be carried out at the Federal University of Rio de Janeiro-UFRJ, which will unfold to include other sports modalities and populations, with the purpose of understanding and guiding an issue of complex biological, psychological, epidemiological, anthropological, sociological, economic and historical aspects thus requiring an interdisciplinary focus to integrate these different dimensions.

ACKNOWLEDGMENTS

FAPERJ, FUJB e UFRJ.

All the authors declared there is not any potential conflict of interests regarding this article.

REFERENCES

Received in 29/4/03

2^nd version received in 10/10/03

Approved in 4/11/03

• 1. Beatty D, Finn SC. Position of the American Dietetic Association and the Canadian Dietetic Association: women's health and nutrition. J Am Diet Assoc 1995;95:362-6.
• ²
  American Psychiatric Association. Diagnostic and statistical manual of mental disorder. 4^ed Washington DC: American Psychiatric Press, 1994.
• 3. Cordás TA, Castilho S. Imagem corporal nos transtornos alimentares. Instrumento de avaliação: Body Shape Questionnaire. Psiquiatria Biológica 1994;2:17-21.
• 4. Nunes MAA. Prevalência de comportamentos alimentares anormais e práticas inadequadas de controle de peso em mulheres de 12 a 21 anos em Porto Alegre [Dissertação de mestrado]. Porto Alegre, RS: Universidade Federal de Pelotas, 1997.
• 5. Katch FI, McArdle WD. Nutrição, exercício e saúde. 4^a ed. Rio de Janeiro, RJ: MEDSI; 1996.
• 6. Bachrach LK, Guido D, Katzman D, Marcus R. Decrease bone density in adolescent girls with anorexia nervosa. Pediatrics 1990;86:440-7.
• 7. Rigotti NA, Nussbaun SR, Herzog DB, Neer RM. Osteoporosis in women with anorexia nervosa. N Engl J Med 1984;311:1601-6.
• 8. Villardi TCC, Ribeiro BG, Soares EA. Distúrbios nutricionais em atletas femininas e suas inter-relações. Revista de Nutrição 2000;1:61-9.
• 9. Morandé G, Carreira M. Anorexias nerviosas y bulimias. Una epidemia actual en adolescentes? Algunos aspectos en discusión. Revista de Psiquiatría de Chile 1988; V:19-28.
• 10. Yeager J, Andersen A, Devlin M, Egger H, Herzog D, Mitchell J, et al. Practice guideline for the treatment of patients with eating disorders. 2^nd ed. In: American Psychiatric Association. Practice guidelines for treatment of psychiatric disorders: compendium 2000. 1^st ed. Washington, DC: American Psychiatric Press, 2000;121-6.
• 11. Fairburn CG, Wilson GT. Binge eating: definition and classification. In: Fairburn CG, Wilson GT, editors. Binge eating: nature, assessment and treatment. New York: The Guilford Press, 1993;3-14.
• 12. Ruud JS. Nutrition and the female athlete. New York, NY: CRC Press, 1996.
• 13. Rosen LW, Hough DO. Pathogenic weigh control behaviors among female gymnasts. Phys Sportsmed 1988;16:140-6.
• 14. Brownell KD, Steen SN. Weigh cycling in athletes: effects on behavior, physiology and health. In: Browell KD, Rodin J, Wilmore JH, editors. Eating, body weight and performance in athletes: disorders of modern society. Philadelphia: Lea and Febiger, 1992;159-71.
• 15. Tangjier F, Sundgot-Borgen J. Influence of body weight reduction on maximal oxygen uptake in female athletes. Scand J Med Sci Sports 1991; 1:141-6.
• 16. McLean JA, Barr SI, Prior JYC. Dietary restraint, exercise, and bone mineral density in young women: are they related? Med Sci Sports Exerc 2001;33:1292-6.
• 17. Cobb KL, Bachrach LK, Greendale G, Marcus R, Neer RM, Nieves J, et al. Disordered eating, menstrual irregularity, and bone mineral density in female runners. Med Sci Sports Exerc 2003;35:711-9.
• 18. Lopiano DA, Zotos C. Modern athletics, the pressure to perform. In: Brownell KD, Rodin J, Wilmore JH, editors. Eating, body weight and performance in athletes: disorders of modern society. Philadelphia: Lea and Febiger, 1992;275-92.
• 19. Rodin J, Larson L. Social factors and the ideal body weight and the ideal body shape. In: Brownell KD, Rodin J, Wilmore JH, editors. Eating, body weight and performance in athletes: disorders of modern society. Philadelphia: Lea and Febiger, 1992;146-58.
• 20. Sundgot-Borgen J. Risk and trigger factors for the development of eating disorders in female elite athletes. Med Sci Sports Exerc 1994;26: 414-9.
• 21. Wilmore JH. Eating and weight disorders in the female athlete. Int J Sports Nutr 1991;1:104-17.
• 22. Wilson GT, Eldredge KL. Pathology and development of eating disorders: implications for athletes. In: Browell KD, Rodin J, Wilmore JH, editors. Eating, body weight and performance in athletes: disorders of modern society. Philadelphia: Lea and Febiger, 1992;115-27.
• 23. Zucker P. Eating disorders in young athletes: a round table. Phys Sportsmed 1985; 13:89-106.
• 24. West RV. The female athlete. The triad of disordered eating, amenorrhea and osteoporosis. Sports Med 1998;26:63-71.
• 25. Loucks AB, Mortola JF, Girton L, Yen SSC. Alterations in the hypothalamic-pituitary-ovarian and hypothalamic-pituitary-adrenal axes in athletic women. J Clin Endocrinol Metab 1989;68:402-11.
• 26. Loucks AB. Effects of exercise training on the menstrual cycle: exercise and mechanism. Med Sci Sports Exerc 1990;22:275-80.
• 27. Loucks AB, Vaitukaitis J, Cameron JL. The reproductive system and exercise in women. Med Sci Sports Exerc 1992;24:S288-93.
• 28. Veldhuis JD, Evans WS, Demers LM, Thorner MO, Wakat D, Rogol AD. Altered neuroendocrine regulation of gonadotropin secretion in women distance runners. J Clin Endocrinol Metab 1985;61:557-63.
• 29. Yeager KK, Agostin R, Nattiv A, Drinkwater B. The female athlete triad. Med Sci Sports Exerc 1993;25:775-7.
• 30. Safran MR, McKeag DB, Camp SPV. Manual de medicina esportiva. São Paulo, SP: Manole; 2002.
• 31. Norton K, Olds T. Antropometrica. Rosário, Argentina: Biosystem Servicio Educativo, 2000.
• 32. Siri WE. Gross composition of the body. In: Lawrence J, Tobias CA, editors. Advanced in biological and medical physics. New York: Academic Press, 1956;v4.
• 33. Withers RT, Whittingham NO, Norton KL, La Fogia J, Ellis MW, Crockett A. Relative body fat and anthropometric prediction of body density of female athletes. Eur J Appl Physiol 1987;56:169-80.
• 34. Quetelet A. Antropométrie ou mesure des différentes facultés de l'homme. Bruxelles: C. Muquardt; 1870.
• 35. Jackson AS, Pollock ML, Warda A. Generalized equation for predicting body density of women. Med Sci Sports Exerc 1980;12:175-82.
• 36. Garner D, Garfinkel P. Eating attitudes test: an index of the symptoms of anorexia nervous. Psychol Med 1979;9:273-9.
• 37. Mann AH, Wakeling A, Wood K, Monck E, Dobbs R, Szmuller G. Screening of abnormal eating attitudes and psychiatric morbidity in unselected population of 15-year old schoolgirls. Psychol Med 1983;13:573-80.
• 38. Johnson-Sabine E, Wood K, Patton G, Mann A, Wakeling A. Abnormal eating attitudes in London schoolgirls. A prospective epidemiological study: factors associated with abnormal response on screening questionnaires. Psychol Med 1988;18:615-22.
• 39. Assunção SSM, Cordás TA, Araújo LASB. Atividade física e transtornos alimentares. Rev Psiq Clín 2002;29:4-13.
• 40. Hendrson M, Freeman CPL. A self-rating scale for bulimia The BITE. Br J Psychiatry 1987;50:18-24.
• 41. Cordás TA, Hpchgraf PB. O "BITE": instrumento para avaliação da bulimia nervosa versão para o português. Jornal Brasileiro de Psiquiatria 1993;42:141-4.
• 42. Cooper PJ, Taylor M, Cooper Z, Fairbun CG. The development and validation of the Body Shape Questionnaire. Int J Eat Disord 1987;6:485-94.
• 43. Cordás TA, Neves JEP. Escalas de avaliação de transtornos alimentares. Revista de Psiquiatria Clínica 1999;26:41-8.
• 44. Matsudo VKR. Menarca em desportistas brasileiras: estudo preliminar. Revista Brasileira de Ciências do Esporte 1982;4:2-6.
• 45. Giacomoni M, Bernard T, Gavarry O, Altare S, Falgairette G. Influence of the menstrual cycle and symptoms on maximal performance. Med Sci Sports Exerc 2000;32:486-90.
• 46. Lebrun CM. Effects of the menstrual cycle and oral contraceptives on sports performance. In: Women in sport (VIC). Blackwell Science, 2000; 37-61.
• 47. Kendrick ZV, Steffen CA, Rumsey WL. Effect of estradiol on tissue glycogen metabolism in exercised oophorectomized rats. J Appl Phys 1987;63:492-6.
• 48. Robergs RA, Roberts SO. Exercise physiology. St Louis: Mosby, 1997; 49-54.
• 49. Bonen A, Haynes F, Watson-Writght W. Effects of menstrual cycle on metabolic response to exercise. J Appl Phys 1983;55:1506-13.
• 50. Wolman RL, Harris MG. Menstrual abnormalities in elite athletes. Clin Sports Med 1989;1:95-9.
• 51. Guedes-Jr DP, Guedes JERP. Controle do peso corporal: composição corporal, atividade física e nutrição. Londrina, PR: Midiograf, 1998.
• 52. McArdle EWD, Katch FI, Katch VL. Essentials of exercise physiology. 2^nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2000.
• 53. Sundgot-Borgen J. Prevalence of eating disorders in elite female athletes. Int J Sport Nutr 1993;3:29-34.
• 54. Powers P, Johnson C. Targeting eating disorders in elite athletes. Part 1. Eat Disord Rev 1996;7:4-10.
• 55. Nunes MAA, Bagatini LF, Abuchaim ALG. O teste de atitudes alimentares (EAT-26) em adolescentes de Porto Alegre. Arquivos de Psiquiatria, Psicoterapia e Psicanálise 1994;1:132-7.
• 56. Yeager KK, Agostini R, Nattiv A, Drinkwater B. The female athlete triad disordered eating, amenorrhea and osteoporosis. Med Sci Sports Exerc 1993;25:775-7.
• 57. Grigg M. Disordered eating and unhealthy weight reduction practices among adolescent females. Prev Med 1996;25:745-8.
• 58. Nunes MA, Olinto MT, Barros FC, Camey S. Influência da percepção do peso e do índice de massa corporal nos comportamentos alimentares anormais. Rev Bras Psiquiatr 2001;23:21-7.
• 59. King MA. Eating disorders in a general practice population: prevalence, characteristics and follow-up at 12 to 18 months. Psychol Med (Suppl) 1989;14:211-8.
• 60. Patton GC. The spectrum of eating disorders in adolescence. J Psychosom Res 1988;32:579-84.
• 61. Sundgot-Borgen J. Risk and trigger factors for the development of eating disorders in female elite athletes. Med Sci Sports Exerc 1994;26: 414-8.
• 62. Rosen LW, McKeag DB, Hough DO, Curley V. Pathogenic weight control behavior in female athletes. Phys Sportsmed 1989;14:79-84.
• 63. O'Conner PJ, Lews RD, Kirchner EM. Eating disorder symptoms in female college gymnasts. Med Sci Sports Exerc 1995;27:550-5.
• 64. Weight L, Noakes TI. Is running an analog of anorexia? A survey of the incidence of eating disorders in female distance runners. Med Sci Sports Exerc 1987;14:213-7.

*

Universidade Federal do Rio de Janeiro, Rio de Janeiro – Brasil.

Publication Dates

• Publication in this collection
  05 Apr 2004
• Date of issue
  Nov 2003

History

• Received
  29 Apr 2003
• Reviewed
  10 Oct 2003
• Accepted
  04 Nov 2003