Introduction
Regular physical activity leads to remarkable benefits determining a better quality of life related to a higher psychophysical well-being (30). To reach an ideal weight, to shape the body, to prevent physical pathologies, to fight stress, or to avoid a sedentary life, an exponential increase of people who regularly go to the gymnasium has been observed in the last 2 decades (13). Nevertheless, active individuals are not free from risks. In fact, it is not uncommon to observe pathological behaviors among gym clients, both men and women, which can endanger health (15).
Similarly to patients with eating disorders (EDs), active individuals usually worry about their body shape, put special attention on their eating patterns, show exercise dependence, and have a perfectionism personality trait (7). Furthermore, considering themselves too frail and flaccid, male gym clients often spend many hours a day at the gymnasium engaging in strength training programs, undergo hyperproteic diets, assume food integrators and in worst cases anabolic drugs, and continuously monitor their muscle development and their body mass. Despite these attitudes that have not yet been defined and coded in psychiatric nosography, this body image disorder has been addressed as bigorexia, reverse anorexia, or muscle dysmorphia (MD) (17,24). In particular, this disorder shares many characteristics with ED (i.e., persistent and rigid eating patterns for weight control), Obsessive-compulsive disorder (i.e., compulsory body monitoring and exercise dependence) and body dysmorphic disorder (i.e., body image disorder), with the individuals, nonchalant of positive and negative comments, completely concentrating on themselves, fearing of being imperfect, feeling ashamed of showing themselves (20). Muscle dysmorphia is often masked by the demands of the sport (17), being elite bodybuilders at higher risk (14). Furthermore, this disorder is more often observed in men aged 18-35 years (18), although women are not exempt from this problem (16). In Italy, despite medical evaluation being mandatory before becoming gym clients, there is a lack of control for psychological profiles that hinders the identification of pathological ED and MD behaviors.
Thus, the aim of this study was to identify whether body weight control is a relevant aspect for exercise among gym clients and to evaluate whether their body concerns, eating concerns, and exercise dependence resemble ED or MD.
Methods
Experimental Approach to the Problem
The institutional review board for use of human subjects approved this study. According to the literature (15), to investigate the possible exercise-related body concerns, eating concerns, and exercise dependence on the risk for ED and MD, the inclusion criterion was that individuals spent more than 1.5 hours at least 3 times a week at the fitness room. In particular, ED and MD, respectively, follow a “body destruction” (i.e., individuals who overexercise to lose weight) and a “body construction” (i.e., individuals who overexercise to gain weight) patterns (Figure 1), with physical exercise playing a crucial role. Moreover, the following exclusion criteria were considered: (a) age < 18 years and (b) participation in sport competitions during the last 2 years.
;)
Figure 1: Schematic representation of the relationship between muscle dysmorphia and eating disorders.
According to the delta between actual and ideal weight, a group of body construction pattern, a group of body destruction pattern, and a group of ED patients have been considered in this research. Eating disorder behavior and body image concerns (BICs) have been assessed by means of Eating Disorder Inventory-2 (EDI-2) and Body Uneasiness Test (BUT). Eating Disorder Inventory-2 (10) is a worldwide validated questionnaire that provides a multidimensional evaluation of the very remarkable psychological characteristics of anorexia and bulimia nervosa, through 11 subscales: drive for thinness (DT), bulimia (BU), interoceptive awareness (IA), asceticism, body dissatisfaction (BD), perfectionism (P), interpersonal distrust, impulse regulation, ineffectiveness (IN), maturity fears, and social insecurity. In the literature (9), high Cronbach's alphas (range: 0.80-0.91) were reported for the internal consistency of the Eating Disorder Inventory 2 (EDI2) scales and high test-retest reliability in patients with ED diagnoses (range: 0.81-0.89) (28). Because several cut-offs can been used to identify subjects with eating disorders (EDs) (4,26), in this study, EDI2 total score > 40 for men and >50 for women and DT > 14 have been used.
To evaluate the BD and some associated emotions (i.e., anxiety, alarm, trepidation, worry, mistrust, misgiving, doubt, suspicion, and embarrassment) of individuals regarding their body weight, the self-report questionnaire BUT has been validated in large nonclinical and clinical samples of individuals suffering from EDs (5). This 71-item instrument is divided into 2 parts with a score ranging from 0 (never) to 5 (always). Body Uneasiness Test (BUT-A) consists of 34 items, accounting for 5 dimensions: weight phobia (fear of being or becoming fat), BICs (worries related to physical appearance), avoidance (A, body image-related avoidance behavior), compulsive self monitoring (CSM, compulsive checking of physical appearance), and depersonalization (D, feelings of detachment and estrangement toward the body), which showed good psychometric properties, a satisfactory internal consistency (Cronbach's alpha: WP = 0.84; BIC = 0.90; A = 0.79; CSM = 0.82; and D = 0.85), and significant test-retest correlation coefficients (0.90) (6). The average of the scores of the BUT-A items represents the Global Severity Index (GSI), with a cut-off point >1.2 indicating a clinically significant body uneasiness. Because the remaining 37 items of BUT-B look at specific worries about particular body parts (i.e., mouth, moustaches, and skin) that are not modifiable with weight loss, for the aim of the present study only Part A was used.
In this study, MD behavior has been evaluated by means of the Muscle Dysmorphia Inventory (MDI) (25), which measures the risk of related behaviors having as principal target adults who consider their body image and physical exercise fundamental for their success. This inventory includes 27 items, which provide information regarding 6 dimensions: size/symmetry (SS), physique protection (PP), exercise dependence (EDp), supplement use (SU), dietary behavior (DB), and pharmacological use (PU), with satisfactory reliability estimates (Cronbach's alpha ranging from 0.72 to 0.94) (25). Although there is no cut-off point, this instrument provides a dimensional description, with high scores indicating increased risk for MD.
Thus, it was hypothesized that differences in gender and desired body weight might influence the tendency to body destructive (i.e., ED) or body constructive (i.e., MD) behaviors of individuals who overexercise.
Subjects
In a fitness center in Southern Italy, 193 clients (≥18 years) who spent more than 1.5 hours at least 3 d·wk−1, were invited to participate in this study. An informed consent was obtained from 134 subjects (86 men and 48 women). Because the Italian law does not allow ED patients to engage in structured exercise activities, 20 ED women (10 restricter anorexics and 10 purger bulimics) were enrolled from a clinical setting. The study did not include ED men and MD women because of a lack of a representative sample and the absence of women who declared to overexercise to gain weight, respectively. Thus, 4 groups were identified: (A) men who overexercise to gain weight (n = 52); (B) men who overexercise to lose weight (n = 34); (C) women who overexercise to lose weight (n = 48); and (D) women with EDs (n = 20).
Procedures
One hour before the physical activity, the questionnaires were completed individually, although an investigator was present to provide assistance if required. After ensuring the confidentiality of the responses, subjects were assured that there were no right or wrong answers. The participant's height (cm) and body mass (kg) were recorded (SECA, Hamburg, Germany) and body mass index (BMI) calculated (kg·m−2). Information regarding the subjects' desired BMI, motivation to do exercise, time spent weekly exercising, eating patterns, self-induce vomiting and other compensatory behaviors (use of diuretics or laxative drugs), and the use of food integrators (i.e., proteins, l-carnitine, vitamins, and sport drinks) or hormones (i.e., steroids) were obtained. Hence, differences between desired and actual BMI were calculated.
Statistical Analyses
Type I error was set at p ≤ 0.05. Age, BMI, and EDI2, BUT, and MDI subscales were compared using 2-tailed analysis of variance (ANOVA). Bonferroni post hoc tests were conducted when significant effects were found and Cohen's effect sizes (ESs) were calculated (3), considering ES ≤ 0.2 trivial, from 0.3 to 0.6 small, from 0.7 to 1.2 moderate and >1.2 large.
Stepwise multiple regression analysis was performed to assess associations between independent variables (delta BMI, purgative behaviors, food integrators intake, amount of physical activity, EDI-2 and BUT positivity and MDI dimensions) and exercise dependence as dependent variable, corrected by sex and age.
Results
Group D was significantly younger (22.1 ± 5.6 years; p = 0.02; ES: 0.37-0.50) than groups A (27.2 ± 6.8 years), B (28.4 ± 7.0 years), and C (28.6 ± 5.8 years). Differences emerged for time spent at the gymnasium (p < 0.0001, ES: 0.23-0.80) with groups D showing lower values (1.0 ± 2.6 h·wk−1) and group A showing higher values (9.0 ± 3.3 h·wk−1) than group B (7.5 ± 2.9 h·wk−1), and C (6.3 ± 2.4 h·wk−1). All the women (groups C and D) declared to currently follow or to have followed in the previous 12 months a restricted diet to lose weight. Of the participants included in group A, 90.4% affirmed to actually follow a hyperproteic diet to increase their muscle mass, whereas 41.2% of men in group B were currently following or had followed a hypocaloric diet during the last year.
Table 1 reports the BMI data and the occurrence of purgative behaviors. For actual BMI differences emerged between groups (p < 0.0001; ES: 0.45-0.83). According to BMI norms for adults over 20 years old (29), groups A and C showed BMI values within the normal range (19-25 kg·m−2), group D within the mild thinness category (17.0-18.5 kg·m−2) and group B within the overweight (≥25 kg·m−2) category. Group A showed a desired BMI higher than actual, whereas the opposite trend emerged for the other groups (p < 0.0001, ES: 0.15-0.82). Differences were found for delta values between desired and actual BMI (p < 0.0001, ES: 0.10-0.75), with positive values for men who desired to gain weight (group A) and negative for the other groups. This discrepancy was higher for men who exercised to lose weight (group B) (−2.24 ± 1.3 kg·m−2) and lower for women with EDs (group D) (−0.82 ± 2.3 kg·m−2). Regarding purgative behaviors, half of eating disordered patients (group D) admitted self-induced vomiting, whereas the occurrence of this behavior was low (2-4%) in the other groups (p < 0.0001, ES: 0.31-0.99). Although women declared a more frequent use (15-21%) of diuretics or laxatives than their male counterparts (10-12%), no significant difference was found between groups (p = 0.49). Differences between groups emerged also for use of food integrators (p < 0.0001, ES: 0.84-0.99), with higher occurrence (21-54%) for men and no occurrence for group D.
Table 1: Mean ± SDs of BMI data and frequency of occurrence (%) of purgative behaviors and use of food integrators in groups A-D.*
A main effect (p < 0.0001, ES: 0.44-0.63) for group emerged for total EDI2 scores (Figure 2), with significantly higher values for group D (96.3 ± 47.3 pt) compared with the other groups (group A: 37.6 ± 19.5 pt: group B: 51.5 ± 33.1 pt; group C: 55.6 ± 33.7 pt). According to a cut-off point on DT > 14 nearly two-thirds of patients from group D resulted positive, whereas only 1 man of group A did. When EDI-2 cut-offs for probable ED (i.e., men > 40, women > 50) were considered, 46% of men in group A and two-thirds of subjects in groups B and C resulted in being at high risk for EDs.
Figure 2: Mean and SDs of the Eating Disorder Inventory-2 (EDI2) dimensions (i.e., drive for thinness, DT; bulimia, BU; interoceptive awareness, IA; asceticism, ASC; body dissatisfaction, BD; perfectionism, P; interpersonal distrust, ID; impulse regulation, IR; ineffectiveness, IN; maturity fears, MF; and social insecurity, SI) for men who overexercise to gain weight (group A), men who overexercise to lose weight (group B), women who overexercise to lose weight (group C), and women with eating disorders (group D).
Figure 2 shows the means and SDs for the EDI2 subscales. Only maturity fears subscale showed no difference between groups. Post hoc analysis showed significantly higher values only for group D with respect to the other groups for DT, BU, IA, P, and IN. No difference for DT emerged between groups B and C. Group A showed significantly lower BD scores than the other groups.
A similar trend was found for BUT-A scores with increasing values from groups A to D (Figure 3). Differences emerged for D (p < 0.0001, ES: 0.13-0.56), CSM (p < 0.0001, ES: 0.05-0.50), WP (p < 0.0001, ES: 0.24-0.71), A (p= 0.002, ES: 0.09-0.37), and BIC (p < 0.0001, ES: 0.15-0.51). Post hoc analysis maintained differences only between group D and the other groups. No difference was found between groups B and C for BIC and between groups A, B, and C for D, CSM and A. Body uneasiness was lower than 1.2 cut-off for groups A and B only. Groups C showed GSI values slightly higher than 1.2 (1.29 ± 0.9), whereas group D showed the highest scores (1.99 ± 1.0).
Figure 3: Means and SDs of Body Uneasiness Test-A (BUT-A) dimensions (i.e., depersonalization, D; compulsive self-monitoring, CSM; weight phobia, WP; avoidance, A; and body image concerns, BIC) for men who overexercise to gain weight (group A), men who overexercise to lose weight (group B), women who overexercise to lose weight (group C), and women with eating disorders (group D).
For MDI subscales (Figure 4), no difference between groups was found for PP and PU (Figure 4). Conversely, differences emerged for DB (p = 0.001, ES: 0.09-0.39), SU (p < 0.0001, ES: 0.20-0.48), EDp (p < 0.0001, ES: 0.20-0.60), and SS (p < 0.001, ES: 0.33-0.42). Higher values were found for groups A and B with respect to groups C and D. In particular, group A always showed significantly higher values than groups C and D, whereas group B showed significantly higher scores than group D only for ED.
Figure 4: Means and SDs of Muscle Dysmorphia Inventory (MDI) dimensions (i.e., dietary behavior, DB; supplement use, SU; physique protection, PP; exercise dependence, ED; size/symmetry, SS; and pharmacological use, PU) for men who overexercise to gain weight (group A), men who overexercise to lose weight (group B), women who overexercise to lose weight (group C), and women with eating disorders (group D).
Multiple regression analyses (R2 = 0.625; p < 0.0001) revealed that EDp could be predicted by the presence of purgative behaviors (p = 0.028), the time spent weekly at the gymnasium (p = 0.004), BUT positivity (p = 0.01), and DB (p < 0.0001), SU (p = 0.01), and SS (p < 0.0001). The regression ANOVA and the scatter plot of the standardized residuals against the standardized predicted values showed no obvious pattern, confirming that the assumptions of linearity and homogeneity of variance had been met.
Discussion
The aim of our study was to elucidate whether people who overexercise resemble more eating disordered patients or muscle dysmorphic individuals. Actually, the growing concerns for body image and medical complications related to obesity have caused an increasing interest for diets and physical activity. Similarly to the findings reported by Drewnowski and Yee (8), all the women (i.e., groups C and D) desired to reduce their weight, whereas 55% of men wanted to increase their weight and 45% wanted to become thinner. In fact, men in group B were slightly overweight (i.e., BMI around 27), whereas those in group A showed a normal-weight (BMI around 24). According to the literature (11,22), this gender difference can be because of cultural influences, with women aiming to achieve a thin figure and men being more concerned of an idealized brawny masculine shape. In fact, to control their weight, men and women adopted different strategies: Women mainly followed restrictive diets (someone also inducing self-vomiting) and were involved in exercise programs as a complement to dieting; men who wanted to lose weight preferred physical activity to diet (8). As expected, people who overexercise (groups A-C) spent more time at the gymnasium with respect to group D because patients with EDs are not allowed to engage in structured exercise programs, but they exercise in loneliness to “burn calories.”
The interviewees were equivocal to the basic norms of a balanced diet, to the recommended supplement dosages and their related health risks. This disinformation was frequent for men who desired to gain weight (54%) and some of them also admitted to use anabolic steroids, proteins, beta 2 agonist, l-carnitine, creatine, and vitamins to enlarge their muscle volume, to reduce their anxiety level, and the perception of fatigue rather than to improve their sport performance and health (23). In general, these substances were assumed without prescriptions, overlooking the potential severe complications such as hypertension and atherosclerosis, alteration of blood lipid levels and coagulation factors, jaundice, hepatotoxicity, carcinoma, tendon damage, influence on insulin production, feedback inhibition of endogenous creatine synthesis, long-term renal damages and psychiatric and behavioral alterations (2,19,27). Conversely, a smaller proportion of people in groups B and C admitted to assume food integrators (mineral salts) mainly with the aim of integrating their electrolytes after excessive physical activity.
As expected, no ED patient (group D) made use of food integrators and widely (50%) practiced self-induced vomiting to control their weight. These results are in line with the fact that half of them were diagnosed as purger bulimics. Interesting to note, all groups showed a similar use of diuretics and laxatives (range 10-21%). This phenomenon is alarming (12), especially considering the potentially dangerous consequences (i.e., gastritis, constipation, hypokalemia, hyponatremia, metabolic alkalosis, and dehydration) linked to the abuse of purgative methods. Although EDs were clinically diagnosed in ED patients, EDI2 data showed a high risk for these alterations for the other groups too, especially relevant for groups B and C. These findings point to the need of further research making use of a clinical interview to find symptoms in the Eating Disorder Not Otherwise Specified (EDNOS) frame (1).
In this study, body uneasiness emerged in most ED patients and in men and women who wanted to lose weight. Conversely, men who wanted to gain weight showed the greatest body satisfaction and the lowest occurrence of body concerns, probably because of a bias of the BUT-A that measures uneasiness mainly in people who want to lose weight. In fact, this group showed high body size/symmetry and exercise dependence scores of the MDI questionnaire, indicating exaggerate physique concerns and altered body image perceptions that compelled the individual to use food integrators and steroids. Despite MDI having no clear cut-off to discriminate exercise-dependent individuals, group A reported significantly higher scores in all the dimension with respect to the other groups. These results, strengthened by the significant linear regression for dietary behavior, supplement use and size/symmetry MDI dimensions could underline the presence of MD symptoms in group A.
This research highlighted the presence of minor body image disorders in individuals who engage in physical activity programs to control their body weight, slightly resembling EDNOS individuals. Conversely, body image was remarkably altered in patients with EDs and in men who overexercise to increase their muscle mass. These findings substantiate a transdiagnostic view of eating disorders, where ED and MD could represent the extremes of a continuum with low self-esteem levels inducing behavioral disorders and increasing obsessive-compulsive mechanisms for self-control through extreme physical exercise in men oriented toward a pathological “body construction” and extreme diet in women oriented toward a pathological “body destruction” (21). In both situations, the achievement of such a rigid behavioral control would activate reward mechanisms to increase self-esteem.
Practical Applications
In Italy, a medical evaluation is required before being admitted into structured exercise programs. However, the results of this study emphasize the need of an early recognition of possible risk factors for ED (i.e., rigid dieting patterns) or MD (i.e., exercise dependence) in gym clients who overexercise. Although physical trainers cannot diagnose nor treat pathological disorders, they could easily notice exaggerated or abnormal behaviors of clients who spend many hours at the gymnasium and recommend further psychological counseling. In particular, physical trainers should focus particular attention on the main characteristics of MD such as exaggerate worries for body shape, rigid eating patterns, exercise dependence, perfectionism, and use of food integrators. More information should be provided to gym clients regarding the benefits of a healthy balanced diet and the health risks associated with purgative behaviors, use of steroids, and abuse of food supplements.
The research has been conducted without any grant.
References
1. American Psychiatric Association.
Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association; 2000.
2. Benzi, G and Ceci, A. Creatine as nutritional supplementation and medicinal product.
J Sports Med Phys Fitness 41: 1-10, 2001.
3. Cohen, J.
Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Erlbaum, 1988.
4. Cotrufo, P, Barretta, V, Monteleone, P, and Maj, M. Full-syndrome, partial-syndrome and subclinical eating disorders: An epidemiological study of female students in Southern Italy.
Acta Psychiatr Scand 98: 112-115, 1998.
5. Cuzzolaro, M, Vetrone, G, Marano, G, and Battacchi, M. Body uneasiness test. In:
Repertorio delle Scale di Valutazione in Psichiatria (Vol. 3). Conti, L, ed. Firenze, Italy: Societa' Editrice Universo; 2000. pp. 1759-1761.
6. Cuzzolaro, M, Vetrone, G, Marano, G, and Garfinkel, PE. The Body Uneasiness Test (BUT): Development and validation of a new body image assessment scale.
Eat Weight Disord 11: 1-13, 2006.
7. Dalle Grave, R, Calugi, S, and Marchesini, G. Compulsive exercise to control shape or weight in eating disorders: Prevalence, associated features, and treatment outcome.
Compr Psychiatry 49: 346-352, 2008.
8. Drewnowski, A and Yee, DK. Men and body image: Are males satisfied with their body weight?
Psychosom Med 49: 626-634, 1987.
9. Eberenz, KP and Gleaves, DH. An examination of the internal consistency and factor structure of the Eating Disorder Inventory-2 in a clinical sample.
Int J Eat Disord 16: 371-379, 1994.
10. Garner, DM.
Eating Disorder Inventory 2. Professional Manual. Odessa, FL: Psychological Assessment Resources, 1993.
11. Geist, R, Heinmaa, M, Katzman, D, and Stephens, D. A comparison of male and female adolescents referred to an eating disorder program.
Can J Psychiat 44: 3748, 1999.
12. Hay, PJ, Mond, J, Buttner, P, and Darby, A. Eating disorder behaviors are increasing: Findings from two sequential community surveys in South Australia.
PLoS One 3: e1541, 2008.
13. Istituto Nazionale di Statistica. La pratica sportiva in Italia. Anno 2006. Available at:
http://www.istat.it/salastampa/comunicati/non_calendario/20070620_00/testointegrale.pdf. Accessed September, 2008.
14. Lantz, CD, Rhea, DJ and Cornelius, AE. Muscle dysmorphia in elite-level power lifters and bodybuilders: A test of differences within a conceptual model.
J Strength Cond Res 16: 649-655, 2002.
15. Lejoyeux, M, Avril, M, Richoux, C, Embouazza, H, and Nivoli, F. Prevalence of exercise dependence and other behavioral addictions among clients of a Parisian fitness room.
Compr Psychiatr 49: 353-358, 2008.
16. Leone, JE. Muscle dysmorphia symptomatology and extreme drive for muscularity in a 23-year-old woman: A case study.
J Strength Cond Res 23: 988-995, 2009.
17. Leone, JE, Sedory, EJ, and Gray, KA. Recognition and treatment of muscle dysmorphia and related body image disorders.
J Athl Train 40: 352-359, 2005.
18. Mangweth, B, Pope, HG Jr, Kemmler, G, Ebenbichler, C, Hausmann, A, De Col, C, Kreutner, B, Kinzl, J, and Biebl, W. Body image and psychopathology in male bodybuilders.
Psychother Psychosom 70: 38-43, 2001.
19. Maravelias, C, Dona, A, Stefanidou, M, and Spiliopoulou, C. Adverse effects of anabolic steroids in athletes. A constant threat.
Toxicol Lett 158: 167-175, 2005.
20. McFarland, MB and Kaminski, PL. Men, muscles, and mood: The relationship between self-concept, dysphoria, and body image disturbances.
Eat Behav 10: 68-70, 2009.
21. Mosley, PE. Bigorexia: Bodybuilding and muscle dysmorphia.
Eur Eat Disord Rev 17: 191-198, 2009.
22. Olivardia, R, Pope, HG, and Hudson, J. Muscle dysmorphia in male weightlifters: A case-control study.
Am J Psychiatry 157: 1291-1296, 2000.
23. Oliver, AJ, León, MT, and Hernández, EG. Statistical analysis of the consumption of nutritional and dietary supplements in gyms.
Arch Latinoam Nutr 58: 221-227, 2008.
24. Pope, HG Jr, Gruber, AJ, Choi, P, Olivadic, R, and Philips, KA. Muscle dysmorphia. An under-recognised form of body dysmorphic disorder.
Psychosomatics 38: 548-557, 1997.
25. Rhea, DJ, Lantz, CD, and Cornelius, AE. Development of the Muscle Dysmorphia Inventory (MDI).
J Sports Med Phys Fitness 44: 428-435, 2004.
26. Sepulveda, AR, Carrobles, JA, and Gandarillas, AM. Gender, school and academic year differences among Spanish university students at high-risk for developing an eating disorder: An epidemiologic study.
BMC Public Health 8: 102, 2008.
27. Shahidi, NT. A review of the chemistry, biological action, and clinical applications of anabolic-androgenic steroids.
Clin Ther 23: 1355-1390, 2001.
28. Thiel, A and Paul, T. Test-retest reliability of the Eating Disorder Inventory 2.
J Psychosom Res 61: 567-569, 2006
29. World Health Organization. BMI classification. Available at:
http://www.who.int. Accessed October, 2007.
30. World Health Organization. Global Strategy on Diet, Physical Activity and Health. Available at:
http://www.who.int/dietphysicalactivity/pa/en/index.html. Accessed January 2009.
