Exercise Dependence and Muscle Dysmorphia in Novice
Exercise Dependence and Muscle Dysmorphia in Novice
Exercise Dependence and Muscle Dysmorphia in Novice
244–248 (2013)
DOI: 10.1556/JBA.2.2013.4.8
(Received: July 9, 2013; revised manuscript received: September 13, 2013; accepted: October 13, 2013)
Background and aims: Extensive research has shown that male bodybuilders are at high risk for exercise depend-
ence, but few studies have measured these variables in female bodybuilders. Prior research has postulated that mus-
cular dysmorphia was more prevalent in men than women, but several qualitative studies of female bodybuilders
have indicated that female bodybuilders show the same body image concerns. Only one study has compared female
bodybuilders with control recreational female lifters on eating behaviors, body image, shape pre-occupation, body
dissatisfaction, and steroid use. The purpose of this study was to compare exercise dependence and muscle dysmor-
phia measures between groups of female weight lifters. Methods: Seventy-four female lifters were classified into
three lifting types (26 expert bodybuilders, 10 or more competitions; 29 novice bodybuilders, 3 or less competitions;
and 19 fitness lifters, at least 6 months prior lifting) who each completed a demographic questionnaire, the Exercise
Dependence Scale (EDS), the Drive for Thinness scale (DFT) of the Eating Disorder Inventory-2, the Bodybuilding
Dependence Scale (BDS), and the Muscle Dysmorphia Inventory (MDI). Results: Female bodybuilders scored
higher than fitness lifters for EDS Total, BDS Training and Social Dependence, and on Supplement Use, Dietary Be-
havior, Exercise Dependence, and Size Symmetry scales of the MDI. Discussion and conclusions: Female
bodybuilders seem to be more at risk for exercise dependence and muscle dysmorphia symptoms than female recre-
ational weight lifters.
INTRODUCTION size and/or strength even though they may actually be quite
large and muscular (Tod & Lavallee, 2010). Components of
Although medical practitioners agree that the majority of the MD include: body image distortion/dissatisfaction, dietary
population in western societies would benefit from more constraints, pharmacological aids, dietary supplements, ex-
regular exercise as part of a healthier lifestyle, a small per- ercise dependence, physique concealment, and low self-es-
centage of individuals may develop an obsessive approach teem (Muscle Dysmorphia Inventory, MDI; Rhea, Lantz &
that can be damaging physiologically, psychologically and Cornelius, 2004).
socially. Researchers and clinicians have recently reviewed One of these components, exercise dependence (ED),
the decades of emerging literature on excessive exercise in has been defined as “a craving for leisure time physical ac-
weight lifters, and some have concluded that the behaviors tivity that results in uncontrollable excessive exercise be-
are part of an obsessive-compulsive disorder diagnosis (e.g., havior and that manifests in physiological symptoms (e.g.,
Pope, Phillips & Olivardia, 2000), while others suggest that tolerance, withdrawal) and/or psychological symptoms
the symptoms are part of a body dysmorphia/body image (e.g., anxiety, depression)” (Hausenblas & Symons Downs,
disorder diagnosis (e.g., Lantz, Rhea & Mayhew, 2001; 2002, p. 90). It has also been measured by the Exercise De-
McCreary & Sasse, 2000), and still others have sought to pendence Scale (EDS; Symons Downs, Hausenblas & Nigg,
differentiate it from a primary eating disorder (e.g., Hausen- 2004). In the bodybuilding realm, Smith, Hale and Collins
blas & Symons Downs, 2002). More recently, Berczik et al. (1998) also created and validated the Bodybuilding Depend-
(2012) have argued forcefully that excessive exercise is a ence Scale (BDS).
type of behavioral addiction. Unfortunately, almost all of the Leone, Sedory and Gray (2005) postulated that MD was
research that has been reviewed to date on addictive anaero- more prevalent in men than women, but several qualitative
bic exercise behavior (Hale & Smith, 2012; Tod & Lavallee, studies of female bodybuilders (Bolin, 1992; Guthrie &
2010) has involved male bodybuilding and weightlifting Castelnuovo, 1992; Klein, 1986, 1992) have stated that fe-
samples. male bodybuilders show the same body image concerns,
Whereas most western women seem to score high on the motivation for muscularity, and workout behaviors as
Drive for Thinness scale (DFT; Garner’s (1991) Eating Dis- males. While extensive reviews of quantitative studies have
order Inventory-2) and yearn to be thin and toned (Thomp- shown that male bodybuilders are at high risk for ED (Hale
son, Heinberg, Altabe & Tantleff-Dunn, 1999), men in the
last three decades are showing increasing scores in the drive
for muscularity (e.g., McCreary and Sasse’s (2000) Drive * Corresponding author: Bruce Hale, PhD, Dept. of Kinesiology,
for Muscularity Scale). According to researchers, some Penn State Berks, 114a BCC, Tulpehocken Rd., Reading, PA
weight lifters develop muscle dysmorphia (MD), view 19610, USA; Phone: +1-610-396-6156; Fax: +1-610-396-6155;
themselves as too thin, and may feel pressure to gain muscle E-mail: bdh1@psu.edu
& Smith, 2012; Smith & Hale, 2011) and may also suffer ican Psychiatric Association, 1994). The seven subscales
from MD (Tod & Lavallee, 2010), few quantitative designs (Tolerance, r = .78; Withdrawal Effects, r = .90; Continu-
(Smith & Hale, 2004; Goldfield, 2009) have measured these ance, r = .90; Lack of Control, r = .82; Reductions in Other
variables in female bodybuilders. Activities, r = .75; Time, r = .86; Intention, r = .89) have all
Goldfield (2009) was one of the few studies to compare shown acceptable scale score reliability; internal consis-
20 female bodybuilders with recreational female lifters on tency for total EDS score for this study was r = .92. Partici-
eating behaviors, body image, shape pre-occupation, body pants are categorized by a total score as “exercise depend-
dissatisfaction, and steroid use. He reported that bodybuild- ent”, “non-dependent symptomatic”, or “non-dependent
ers scored higher on the Bulimia subscale, Drive for Bulk asymptomatic”. Hausenblas and Symons Downs (2002) and
scale (Blouin & Goldfield, 1995), and Drive for Tone scale Hausenblas and Giacobbi (2004) presented evidence of con-
(Goldfield, 2009). More recently Hale, Roth, DeLong and current validity of the EDS.
Briggs (2010) found that male bodybuilders and power lift-
ers were significantly higher than fitness lifters on EDS To- Bodybuilding Dependence Scale
tal, seven EDS-R scales, and the three BDS scales. No study
to date has compared measures of MD and ED between dif- The Bodybuilding Dependence Scale (BDS; Smith et al.,
ferent groups of female weight lifters. 1998) is a 9-item, 7-choice Likert scale (“Strongly Disagree”
Although the estimates of bodybuilders suffering from to “Strongly Agree”) with three dimensions (Social Depend-
ED and MD may be small in western populations, the study ence, Training Dependence, and Mastery Dependence) to
of ED and MD in female weight lifters is warranted. This measure the degree to which ED is exhibited in weight lifters
study hypothesized that female bodybuilders would score based on Veale’s (1987) biomedical and psychosocial diag-
significantly higher in ED, MD (Hale & Smith, 2012; Smith nostic criteria. This measure has demonstrated adequate
& Hale, 2011; Tod & Lavallee, 2010), and lower in DFT psychometric internal reliability (Cronbach’s alpha of .83,
(Goldfield, 2009) than female fitness lifters. .70, and .89, respectively, in this study), construct and concur-
rent validity (Hurst et al., 2000; Smith & Hale, 2004), and ad-
equate test–retest reliability (Smith & Hale, 2005) for each
METHODS scale (r = .97, .96, and .94, respectively).
tomatic”, and 4.1% (n = 3) as “nondependent asymptom- differences between different lifting types. More partici-
atic”. Of the 10 “at risk” for ED, nine were bodybuilders and pants should have been measured in each lifting group,
one was a fitness lifter. In this sample of female lifters, the prevalence of
‘at-risk’ behavior for ED behaviors (13.5%) was found to be
Bodybuilding Dependence Scale higher than other college-age samples measured by the EDS
(3.4%, Hausenblas & Symons Downs, 2002; 3.6–5%,
A significant overall MANOVA group main effect (Wilks’ Symons Downs et al., 2004). Past findings (Allegre,
lambda = .66, F(6, 138) = 5.30, p < .05) was calculated. Uni- Souville, Therme & Griffiths, 2006; Hausenblas & Symons
variate F-tests showed that Training Dependence was signif- Downs, 2002; Terry, Szabo & Griffiths, 2004) of mixed
icant (F(2, 71) = 12.35, p < .001), and Tukey-tests indicated gender samples have been conservatively in the 3–13%
that expert (M = 14.46) and novice (M = 14.41) bodybuilders range. In this study nine out of 10 of the “at risk” scores
were significantly higher than fitness lifters (M = 9.79). A came from bodybuilder groups.
significant Social Dependence scale group main effect This at-risk finding is further supported by results from
(F(2, 71) = 16.68, p < .001) indicated that expert (M = 20.92) the BDS analysis. On two of the three scales, female
and novice (M = 21.59) bodybuilders were significantly bodybuilders scored higher than fitness lifters. These find-
higher than fitness lifters (M = 12.47) (see Table 1). No sig- ings are supported by Hale et al. (2010) and Smith and Hale
nificant group main effect was found for Mastery Depend- (2004) with male bodybuilders and fitness lifters.
ence, F(2, 71) = 2.56, p = .08. The results for the MD assessment also provided partial
support for the hypothesis that female novice and expert
Muscle Dysmorphia Inventory bodybuilders would score higher. The significant finding on
the Exercise Dependence scale suggests that women body-
A significant overall MANOVA group main effect (Wilks’ builders, like male bodybuilders, are at extremely high risk
lambda = .44, F(12, 132) = 5.59, p < .05) occurred. Univa- for MD symptoms (Smith & Hale, 2004; Tod & Lavallee,
riate F-tests indicated significant differences in Supplement 2010). These findings further support the qualitative find-
Use (F(2, 71) = 23.43, p < .001), Dietary Behavior (F(2, 71) ings of Bolin (1992), Guthrie and Castelnuovo (1992), and
= 21.80, p < .001), Exercise Dependence (F(2, 71) = 27.24, Klein (1986, 1992) and quantifiable results of Goldfield
p < .001), and Size Symmetry (F(2, 71) = 11.09, p < .01). (2009) with women bodybuilders. With no differences re-
Follow up Tukey post hoc tests showed that expert and nov- ported here between novice and experienced bodybuilders,
ice bodybuilders scored significantly higher than fitness lift- it suggests that women who are attracted to serious body-
ers on these four scales (see Table 1). building programs may arrive with symptoms of MD intact
already or may develop these symptoms soon after commit-
Drive for Thinness Scale ting to regimented training.
Finally, the non-significant findings between lifting
No significant differences occurred in DFT score between groups on the Drive for Thinness scale rejected our research
the lifting groups, F(2, 71) = .12, p = .89. hypothesis. Since Hausenblas and Symons Downs’ (2002)
criteria for secondary exercise dependence is a score of 14 or
better, all three groups seem to be highly at risk for an eating
DISCUSSION disorder. The high scores also suggest that in addition to a
potentially high drive for muscularity (McCreary & Sasse,
In general, bodybuilders spent more years, time in the gym, 2000), all female weight lifters may also want to remain ex-
and worked out harder than fitness lifters. This finding is tremely lean.
similar to differences in workout frequency reported for Recently Berczik et al. (2012) have suggested that ED
male bodybuilders and fitness lifters (Hale et al., 2010), but should be more appropriately labeled as exercise addiction
as Hausenblas and Symons Downs (2002) reported, exercise with six common symptoms (salience, mood modification,
behavior and history alone are not adequate predictors of tolerance, withdrawal symptoms, personal conflict, and re-
ED. The finding that bodybuilders’ typical workout was lapse). Furthermore, they assert that exercise addiction is a
moderate-high in intensity compared to fitness lifters’ form of behavioral addiction because of its preoccupation
light-moderate intensity was similar to recent findings of with the behavior when it is prevented or delayed. It is clear
Cook, Hausenblas and Rossi (2013), who reported that par- that measurement of ED (addiction?) needs improved diag-
ticipants who wanted to gain weight (e.g., bodybuilders) had nostic processes. A recent study by Heaney, Ginty, Carroll
significantly higher amounts of strenuous exercise than and Phillips (2011) showed that ED participants produced a
women who wanted to lose weight (e.g., fitness lifters). blunted cardiovascular and cortisol reaction to stress, similar
Other antecedent variables must be examined to try to un- to those seen in alcohol and smoking dependence; this find-
derstand the etiology of the disorder. ing may offer a more accurate, future objective measure-
The hypothesis that bodybuilders would show higher ment.
scores in exercise dependence was partially supported. Al- This study contains several design limitations. It was a
though a predicted higher score was calculated for expert correlational, cross-sectional design that could not provide
bodybuilders over fitness lifters in total EDS, the MANOVA cause-and-effect findings which might help predict the etiol-
for scale differences just failed to reach significance. Re-ex- ogy of ED and MD. In addition, the sample was voluntary
amination showed that the statistical power was .87, which and limited to a small group of bodybuilders from one major
is barely adequate for a multivariate analysis involving competition and several health clubs with a non-random fit-
seven dependent variables and 74 participants. The finding ness group used as a comparator, which reduced statistical
for higher total EDS scores is similar to other previous stud- power for internal validity and decreased the potential for
ies (Hale et al., 2010; Hurst et al., 2000) that have examined external validity. The questionnaires selected are only indic-
ative of ‘at risk’ symptoms inherent in ED and MD; only Hale, B. D., Roth, A., DeLong, R. & Briggs, M. (2010). Exercise de-
clinical diagnostic procedures combined with question- pendence and the drive for muscularity in male bodybuilders,
naires and possible biochemical analyzes can lead to clear power lifters, and fitness lifters. Body Image, 7, 234–239.
diagnosis. Future research needs more diverse samples of fe- Hale, B. D. & Smith, D. (2012). Bodybuilding. In T. Cash (Ed.),
male weight lifters and ED and MD self-report measures Encyclopedia of body image and human performance (Chap-
given in random order that include measures that control for ter 46). Oxford, UK: Elsevier.
social desirability. Hausenblas, H. & Giacobbi, P. (2004). Relationship between exer-
In conclusion, the overall findings of this study support cise dependence symptoms and personality. Personality and
Individual Differences, 36, 1265–1273.
the hypothesis that female bodybuilders, whether new or ex-
Hausenblas, H. & Symons Downs, D. (2002). Exercise depend-
perienced competitors, show the same high risks for ED and
ence: A systematic review. Psychology of Sport and Exercise,
MD that male bodybuilders have shown (Hale & Smith,
3, 89–123.
2012; Tod & Lavallee, 2010). This study is one of the first Heaney, J., Ginty, A., Carroll, D. & Phillips, A. (2011). Preliminary
quantifiable cross-sectional designs to compare differences evidence that exercise dependence is associated with blunted
in potentially pathological exercise behaviors and eating dis- cardiac and cortisol reactions to acute psychological stress. In-
orders in different female weight lifting groups. The find- ternational Journal of Psychophysiology, 79(2), 323–329.
ings suggest that either gender of bodybuilders may be at Hurst, R., Hale, B. D & Smith, D. (2000). Exercise dependence in
high risk for potentially addictive behavioral disorders (ex- bodybuilders and weight lifters. British Journal of Sports Med-
ercise dependence, muscle dysmorphia). icine, 11, 319–325.
Klein, A. M. (1986). Pumping iron: Crisis and contradiction in
bodybuilding subculture. Society of Sport Journal, 3, 112–133.
Klein, A. M. (1992). Man makes himself: Alienation and self-ob-
Funding sources: No financial support was received for this jectification in bodybuilding. Play and Culture, 5, 326–337.
study. Lantz, C. D., Rhea, D. J. & Mayhew, J. L. (2001). The drive for
size: A psycho-behavioral model of muscle dysmorphia. Inter-
Authors’ contribution: BDH (study concept and design, national Sport Journal, 5, 71–85.
analysis and interpretation of data, study supervision); DD Leone, J. E., Sedory, E. J. & Gray, K. A. (2005). Recognition and
(data collection); KW (data collection); MB (analysis and treatment of muscle dysmorphia and related body image disor-
interpretation of data; statistical analysis). ders. Journal of Athletic Training, 40, 352–359.
Little, R. J. A. (1992). Regression with missing x’s: A review. Journal
Conflict of interest: The authors declare no conflict of interest. of the American Statistical Association, 87(420), 1227–1237.
McCreary, D. R. & Sasse, D. K. (2000). An exploration of the drive
for muscularity in adolescent boys and girls. Journal of Ameri-
can College Health, 48, 297–304.
REFERENCES Pope, H. G., Phillips, K. A. & Olivardia, R. (2000). The Adonis
complex: The secret crisis of male body obsession. New York:
Free Press.
Allegre, B., Souville, M., Therme, P. & Griffiths, M. (2006). Defi-
Rhea, D. J., Lantz, C. D. & Cornelius, A. E. (2004). Development
nitions and measures of exercise dependence. Addiction Re-
of the Muscle Dysmorphia Inventory (MDI). The Journal of
search and Theory, 14, 631–646.
Sports Medicine and Physical Fitness, 44, 428–435.
American Psychiatric Association (1994). Diagnostic and statisti-
Smith, D. & Hale, B. D. (2004). Validity and factor structure of the
cal manual of mental disorders (4th Edition). Washington, DC:
Bodybuilding Dependence Scale. British Journal of Sports
Author.
Medicine, 38, 177–181.
Berczik, K., Szabo, A., Griffiths, M. D., Kurimay, T., Kun, B., Ur-
Smith, D. & Hale, B. D. (2005). Exercise dependence in body-
ban, R. & Demetovics, Z. (2012). Exercise addiction: Symp-
building: Antecedents and reliability of measurement. Journal
toms, diagnosis, epidemiology, and etiology. Substance Use &
of Sports Medicine and Physical Fitness, 45, 401– 408.
Misuse, 47, 403–417.
Smith, D. & Hale, B. D. (2011). Exercise dependence. In D.
Blouin, A. G. & Goldfield, G. S. (1995). Body image and steroid
Lavallee & D. Tod (Eds.), The psychology of strength training
use in male bodybuilders. International Journal of Eating Dis-
and conditioning (Chapter 8). London, UK: Routledge.
orders, 18, 159–165.
Smith, D. K., Hale, B. D. & Collins, D. J. (1998). Measurement of
Bolin, A. (1992). Flex appeal, food, and fat: Competitive body-
exercise dependence in bodybuilders. Journal of Sport Medi-
building, gender, and diet. Play and Culture, 5, 378–400.
cine and Physical Fitness, 38, 66–74.
Cook, B., Hausenblas, H. & Rossi, J. (2013). The moderating effect
Symons Downs, D., Hausenblas, H. & Nigg, C. R. (2004). Facto-
of gender on ideal-weight goals and exercise dependence
rial validity and psychometric examination of the Exercise De-
symptoms. Journal of Behavioral Addictions, 2(1), 51–55.
pendence Scale-Revised. Measurement in Physical Education
doi:10.1556/JBA.1.2012.010 and Exercise Science, 8, 183–201.
Donders, A. R. T., Van der Heijden, G. J. M. G., Stijnen, T. & Terry, A., Szabo, A. & Griffiths, M. (2004). The Exercise Addic-
Moons, K. G. M. (2006). Review: A gentle introduction to im- tion Inventory: A new brief screening tool. Addiction Research
putation of missing values. Journal of Clinical Epidemiology, and Theory, 12, 489–499.
59, 1087–1091. Thompson, J. K., Heinberg, L. J., Altabe, M. & Tantleff-Dunn, S.
Garner, D. M. (1991). Eating Disorder Inventory-2 manual. (1999). Exacting beauty: Theory, assessment, and treatment of
Odessa, FL: Psychological Assessment Resources. body image disturbance. Washington, D.C.: American Psycho-
Goldfield, G. S. (2009). Body image, disordered eating, and anabo- logical Association.
lic steroid use in female bodybuilders. Eating Disorders, 17, Tod, D. & Lavallee, D. (2010). Towards a conceptual understanding
200–210. doi: 10.1080/10640260902848485 of muscle dysmorphia development and sustainment. Interna-
Guthrie, S. R. & Castelnuovo, S. (1992). Elite women body- tional Review of Sport and Exercise Psychology, 3(2), 111–131.
builders: Models of resistance or compliance? Play and Cul- Veale, D. (1987). Exercise dependence. British Journal of Addic-
ture, 5, 401–408. tion, 82, 735–740.
248 | Journal of Behavioral Addictions 2(4), pp. 244–248 (2013)