Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders

Document Sample
Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders Powered By Docstoc
					Original Research

Body Image, Binge Eating, and Bulimia Nervosa in Male

Gary S Goldfield, PhD1,2, Arthur G Blouin, PhD2, D Blake Woodside, MD3

                  Objective: Male bodybuilders (MBB) exhibit more severe body dissatisfaction, bulimic eating
                  behaviour, and negative psychological characteristics, compared with male athletic and nonathletic
                  control subjects, but few studies have directly compared MBB and men with eating disorders. This
                  study compared men with bulimia nervosa (MBN), competitive male bodybuilders (CMBB), and
                  recreational male bodybuilders (RMBB) on a broad range of eating attitudes and behaviours and
                  psychological characteristics to more accurately determine similarities and differences among these
                  Method: Anonymous questionnaires, designed to assess eating attitudes, body image, weight and
                  shape preoccupation, prevalence of binge eating, weight loss practices, lifetime rates of eating
                  disorders, anabolic androgenic steroid (AAS) use, and general psychological factors, were
                  completed by 22 MBN, 27 CMBB, and 25 RMBB.
                  Results: High rates of weight and shape preoccupation, extreme body modification practices, binge
                  eating, and bulimia nervosa (BN) were reported among MBB, especially among those who
                  competed. CMBB reported higher rates of binge eating, BN, and AAS use compared with RMBB,
                  but exhibited less eating-related and general psychopathology compared with MBN. Few
                  psychological differences were found between CMBB and RMBB.
                  Conclusions: MBB, especially competitors, and MBN appear to share many eating-related features
                  but few general psychological ones. Longitudinal research is needed to determine whether men
                  with a history of disordered eating or BN disproportionately gravitate to competitive bodybuilding
                  and (or) whether competitive bodybuilding fosters disordered eating, BN, and AAS use.
                  (Can J Psychiatry 2006;51:160–168)
                  Information on funding and support and author affiliations appears at the end of the article.

                    Clinical Implications

                    · Bodybuilding among men is associated with an increased risk of body dissatisfaction, weight and shape
                      preoccupation, and pathological eating behaviours.
                    · Compared with RMBB, CMBB are more likely to engage in binge eating and meet criteria for BN
                      during their lifetime
                    · CMBB are also at higher risk of using AAS, compared with RMBB, but the high prevalence of use in
                      RMBB indicates the use of steroids for cosmetic reasons.


                    · A clinical sample of eating disorder patients and a community-based sample of MBB comprised the
                      overall sample. The differences in recruitment may have biased the results.
                    · Diagnostic criteria for BN was established by self-report without a confirmatory clinical interview;
                      thus, the prevalence and severity of binge eating and abnormal weight control practices may be
                      overestimated or underestimated.
                    · The sample of MBB was relatively small; therefore, results may not represent the studied populations.

160                                                                              W Can J Psychiatry, Vol 51, No 3, March 2006
                                                                  Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders

Key Words: bodybuilding, body image, binge eating,                This may have important implications for psychological
bulimia nervosa, anabolic androgenic steroids                     health, given that activities or sports that require
                                                                  overinvestment in body shape and physical appearance have
     arge-scale surveys reveal that male body dissatisfaction
L    has increased dramatically during the last 3 decades, from
15% to 43%, making current rates almost comparable to those
                                                                  been noted as a risk factor for developing disordered eating or
                                                                  eating disorders (11).

found in women (1). There is converging evidence from             Uncontrolled research indicates that MBB report high rates of
cross-sectional and experimental research that exposure to the    body dissatisfaction (12), weight and shape preoccupa-
exceptionally thin beauty standards for women as advertised       tion (13), and unhealthy methods of body modification prac-
in the media, as well as exposure to the lean and muscular male   tices, such as strict dieting and dehydration methods (13) and
ideal, increases body dissatisfaction and negative affect in      AAS use (14). Controlled studies show inconsistent findings,
both women and men (2–6). Female body dissatisfaction typi-       with data showing greater body dissatisfaction or disordered
cally manifests in feeling too heavy or fat with a concomitant    eating in MBB in comparison with male athletic and
desire to be thinner (7), while most young men seek to be         nonathletic groups (15), while other data show that MBB
leaner, yet larger and more muscular (4). These expressions of    exhibit a more positive body image, compared with active
body dissatisfaction are consistent with standards of attrac-     exercisers (16) and nonactive control subjects (17). Regard-
tiveness for each sex. The high prevalence of body dissatisfac-   ing psychological characteristics, evidence suggests that,
tion is concerning, given that body image issues are often the    compared with control subjects, MBB report more narcis-
driving force underlying disordered eating, compensatory          sism (18); hypermasculine beliefs and homophobia (19); inef-
bulimic behaviours, full-blown eating disorders (8), and use      fectiveness (14); and elevated scores on obsessionality,
of AAS (9).                                                       perfectionism, and anhedonia (18). However, the differences
                                                                  between CMBB and RMBB in terms of eating behaviour or
In response to this hypermesomorphic somatype portrayed as
                                                                  general psychological characteristics have not been well stud-
the masculine ideal, many adolescent and young adult men are
                                                                  ied; thus, the psychological effects of pursuing the
engaging in serious weight training or bodybuilding (10).
                                                                  hypermesomorphic and masculine ideal imposed on CMBB
                                                                  remains unknown.

                                                                  When combined, there is evidence that MBB exhibit a psy-
                                                                  chological profile resembling people with eating disorders.
 Abbreviations used in this article                               Valid conclusions regarding the degree of concordance of
 AAS           anabolic androgenic steroids                       eating-related and psychological profiles between MBB and
 AN            anorexia nervosa                                   individuals with eating disorders requires that these groups be
 ANCOVA        analysis of covariance                             directly compared, but only 2 such studies exist. Davis and
 ANOVA         analysis of variance
                                                                  Robertson found no significant differences between female
                                                                  patients with AN and CMBB in terms of obsessionality,
 ASQ           Anabolic Steroid Questionnaire
                                                                  perfectionisms, anhedonia, and narcissism, but both of these
 BDI           Beck Depression Inventory
                                                                  groups scored significantly higher than control subjects (18).
 BMI           body mass index
                                                                  However, Davis and Robertson found that CMBB reported
 BN            bulimia nervosa                                    more positive perceptions of self-worth and body image, com-
 C-DIS         Computerized Diagnostic Interview Schedule         pared with women with anorexia (18). In the second study,
 CMBB          competitive male bodybuilders                      Mangweth and others found that MBB displayed higher rates
 df            degrees of freedom                                 of body dissatisfaction and weight preoccupation than
 EDI           Eating Disorder Inventory                          nonathletic male control subjects, but lower rates than men
 MANCOVA       multivariate analysis of covariance                with a history of eating disorders (20). Men with eating disor-
                                                                  ders reported higher rates of psychiatric disorders than both
 MANOVA        multivariate analysis of variance
                                                                  MBB and control subjects, while MBB reported more mood
 MBB           male bodybuilders
                                                                  disorders than control subjects (20). The proposed study is
 MBN           men with bulimia nervosa
                                                                  designed to build on past research by comparing and contrast-
 NIMH-DIS      National Institute of Mental Health Diagnostic     ing CMBB, RMBB, and MBN on a broad scope of eating atti-
               Interview Schedule
                                                                  tudes and behaviour and general psychological characteristics
 RMBB          recreational male bodybuilders
                                                                  in order to further identify similarities and differences among
 SD            standard deviation                                 these groups. The current study differs from the Davis and
                                                                  Robertson and the Mangweth and others (20) studies in

Can J Psychiatry, Vol 51, No 3, March 2006 W                                                                                    161
The Canadian Journal of Psychiatry—Original Research

several important ways. First, the current study focuses less on   overconcerned with body shape or weight, and were using at
general psychological characteristics or psychiatric disorders     least one compensatory method of weight control (that is,
and assesses a broader measure of eating-related behaviours,       purging) within the 2 weeks prior to assessment. Owing to
including binge eating, dieting, and compensatory behaviours       insufficient numbers of men seeking treatment of eating dis-
designed to prevent weight gain common in eating disorders         orders during the study period, men who had previously pre-
(that is, excessive exercise, vomiting, and diuretic and laxa-     sented for treatment at the clinics were contacted and invited
tive use). Second, as opposed to other studies that used either    to participate, but all were symptomatic according to the
women with AN (18) or men with several eating dis-                 aforementioned criteria.
orders (20), the current study used an eating disorder group       After complete description of the study, participants gave
solely comprising men with BN. We believe the most relevant        their written informed consent. This study was approved by
comparison group is men with BN because our previous               the research ethics boards at Carleton University, the Ottawa
research showed elevations on the Bulimia subscale of the          Hospital, and the Toronto Hospital.
EDI and because CMBB eat a very restricted diet that may
predispose them to binge eating (21) and compensatory              Procedure
behaviours designed to prevent weight gain in fat from binges.     MBB were recruited by posting advertisements in local gym-
Also, one cannot meet criteria for AN and still be a               nasia. They were given a cover letter describing the purpose of
bodybuilder, but a bodybuilder may still meet the criteria for     the study as to examine “eating attitudes and behavior, weight
BN, which suggests that BN may be a more appropriate com-          loss practices, general psychological characteristics, as well
parison group. Third, the current study compares MBB and           prevalence of current and past use of anabolic androgenic ste-
MBN chosen from the same culture and data collected during         roid use, and attitudes toward steroid use in recreational and
the same time period, whereas previous research compared           competitive bodybuilders.” To protect their anonymity,
Austrian MBB to American men with eating disorders. Thus           bodybuilders were instructed not to include their name or con-
the current study better controls for effects of current           tact information on any of the surveys, and they were guaran-
sociocultural norms on eating-related behaviour. Fourth, we        teed that data obtained would only be available to study staff.
decided to include RMBB as a comparison group instead of           Any publication of data would involve aggregated data so that
nonathletic control subjects used in previous research (18,19)     no individual could be identified. Inclusion criteria for the
for 2 main reasons: research has already shown that MBB            CMBB group required a person to be either actively training
exhibit more disordered eating than other athletes and             for a competition or to have competed within the past 12
nonathletic control subjects, and the current design controls      months. RMBB were defined as those people who engaged in
for the attitudinal and environmental effects of the gym on        traditional forms of weight training (for example, free weights
eating-related parameters to better isolate the effects of com-    or nautilus) at least twice weekly for 7 months or more
petition on selected dependent measures.                           (although all did significantly more) and who had never com-
We predicted that CMBB would exhibit an eating and psycho-         peted in a bodybuilding competition and had no plans to do so
logical profile that was similar to, though less severe than,      in the next 12 months.
men with BN and more pathological than that of RMBB.               Patients presenting for treatment at the eating disorder clinics
                                                                   in Ottawa and Toronto were randomly approached to partici-
Methods                                                            pate in the study. During the patients’ initial assessments, a
Subjects                                                           clinical interview was conducted in order to verify the BN
The volunteer sample of 74 participants comprised 22 MBN,          diagnosis. Previous patients with BN who were not in treat-
27 CMBB, and 25 RMBB. MBB were recruited from flyers               ment were contacted and invited to participate in a telephone
posted in gymnasia in Ottawa, Ontario. The clinical sample         screening interview. Those who were symptomatic according
consisted of volunteers who sought treatment from eating dis-      to the above criteria, and also interested, were mailed ques-
order clinics located in large hospitals in Ottawa and Toronto.    tionnaires and instructed to complete and return them by mail
MBN recruited from Ottawa and Toronto did not significantly        in prestamped envelopes. These procedures also applied to
differ in demographic variables, eating-related pathology, or      MBB. Participants were not financially remunerated for their
general psychological characteristics and were therefore           participation.
combined to form one group.                                        Measurement Instruments
All 22 men (100%) met the DSM-III-R lifetime criteria for          Subjects completed an assessment package containing a brief
BN as assessed by clinician interview, were symptomatic at         demographics questionnaire, the BDI (22), and the EDI (23),
the time of testing, defined as currently engaging in binge eat-   to assess eating-related attitudes and psychological factors
ing at least once weekly in the past 3 months, were persistently   associated with eating disorders. A paper and pencil version

162                                                                                    W Can J Psychiatry, Vol 51, No 3, March 2006
                                                                        Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders

          Table 1 Demographic and weight training characteristics of the sample

                                                     MBN                           CMBB                             RMBB
                                                     n = 22                        n = 27                           n = 25

          Variable                          Mean              SD           Mean             SD            Mean               SD

          Agea,b                             33.6             8.9           26.7            5.0            24.9              5.0

          Educationa,b                         4.0            1.5            4.9            1.4             5.1              1.1

          Height (cm)a,c                    178.6             5.7          173.3            7.4           176.4              6.1

          Weight (kg)                        82.2             30.4          93.6            16.1           82.8              18.4

          BMI                                25.8             9.6           31.1            4.5            26.6              5.3

          MBN compared with CMBB, P < 0.05; bMBN compared with RMBB, P < 0.05; cCMBB comapred with RMBB, P < 0.05

of the Eating Disorder section of the validated C-DIS (24),             tests using Tukey’s tests. Owing to differences in
which was derived from the NIMH-DIS (25,26), was admin-                 demographics among groups and high intercorrelations
istered to examine the prevalence of bulimic eating practices,          among demographic characteristics and outcome measures,
weight and shape preoccupation, and lifetime rates of BN                analyses (ANCOVA and MANCOVA) were conducted with
based on DSM-III-R criteria (24). A bodybuilding question-              age, BMI, and education as covariates with a pattern of results
naire designed to classify MBB as competitive or recreational           emerging that was identical to those that did not use
was used. Previous research has shown this instrument to have           covariates; thus, results are presented without covariates to
good discriminative validity (14). The ASQ (27) was                     maximize statistical power. The prevalence of AAS use and
expanded to assess a wider range of motivations for using               bulimic symptoms across groups was examined by chi-square
AAS, as well as capturing possible effects of AAS, data for             analyses.
which are reported elsewhere (28). In addition, the 6-item
Drive for Bulk scale was developed as a modification of the             Results
Body Dissatisfaction scale of the EDI. The direction of items           Demographic and Weight Training Characteristics
was reversed (for example, “too big” changed to “too small”)            Table 1 shows the demographic characteristics of each group.
and references to body parts were adapted to the masculine              MBN were significantly older than MBB. CMBB had signifi-
(mesomorphic) ideal. This scale was sensitive in previous               cantly higher BMI (assessed by self-report) than MBN and
research, significantly discriminating MBB from both run-               RMBB, with no differences between MBN and RMBB. Both
ners and martial artists (14). Finally, a 5-item Drive for Tone         bodybuilding groups reported a significantly higher level of
subscale was developed and assessed participants’ desire for            education (some college courses to college graduate) com-
leaner and more toned body parts with the same 5-point likert           pared with MBN (high school graduate to some college
rating scale as the EDI.                                                courses). CMBB lifted weights more frequently (weekly)
                                                                        than RMBB and MBN (mean 4.7, SD 0.71, compared with
Analytic Plan                                                           mean 3.9, SD 0.78), but no differences among bodybuilding
MANOVA was used to test the effects of Group (MBN,                      groups were found for number of years in which participants
CMBB, RMBB) on 2 categories of dependent measures: eat-                 had lifted weights (mean 8.6, SD 3.9, compared with mean
ing-related attitudes and behaviour, which included the Drive           6.8, SD 4.7) or duration (minutes) of their workout (mean
for Thinness, Bulimia, and Body Dissatisfaction subscales of            86.5, SD 41.7, compared with mean 78.0, SD 21.4).
the EDI, as well as the Drive for Bulk and Drive for Muscle
Tone scales developed for this study; and general psychologi-           Body Image and Eating Attitudes
cal factors commonly observed in, but not specific to, eating           Table 2 displays the results of eating-related and general psy-
disorders, such as depression, ineffectiveness, perfectionism,          chological factors analyzed across conditions. MANOVA
interpersonal distrust, interoceptive awareness, and maturity           (Wilk’s Lambda) revealed a significant main effect of group
fears. Significant multivariate findings (P < 0.05, 2-tailed)           (F10,134 = 5.97; P < 0.001). ANOVA and Tukey’s post hoc
were followed by ANOVA and multiple comparison post hoc                 tests indicated that, compared with either bodybuilding

Can J Psychiatry, Vol 51, No 3, March 2006 W                                                                                          163
The Canadian Journal of Psychiatry—Original Research

  Table 2 Mean (SD) scores of eating-related and psychological factors across groups

                                                              MBN                              CMBB                             RMBB
                                                              n = 22                           n = 27                           n = 25

  Variable                                            Mean              SD             Mean              SD             Mean              SD

  Eating-related measures

       Drive for thinnessa,b                            9.1             6.5              3.5             3.8              3.2             3.9
       Bulimia                                          6.1             5.6              1.2             1.7              1.2             2.6

       Body dissatisfactiona,b                         10.5             6.7              4.6             3.6              4.9             3.3
       Drive for bulk                                  22.2             9.6             28.9             6.1             29.0             5.8

       Drive for tone                                  20.6             6.8             23.8             6.5             24.5             6.2

  Psychological measures

       Ineffectivenessa,b                              10.6             8.5              2.2             3.8              1.8             2.9

       Pefectionism                                     5.3             4.3              6.8             4.8              7.1             4.4

       Interpersonal distrusta,b                        6.8             4.9              3.4             3.9              3.2             2.6
       Interoceptive awareness                          8.4             6.8              2.3             3.6              2.8             4.7

       Maturity fearsa.b                                7.1             6.8              2.8             3.1              3.4             3.6
       Depression                                      21.2             11.9             6.2             6.4              4.2             3.4
   MBN compared with CMBB, P < 0.05; bMBN compared with RMBB, P < 0.05; All measures are from EDI, except that depression was measured by the BDI,
  P < 0.05

condition, MBN reported significantly higher scores on all                     were common among CMBB and RMBB. Table 3 shows the
subscales of the EDI, with no significant differences between                  lifetime prevalence and symptoms of BN across groups.
bodybuilding groups on these measures. As expected, both
CMBB and RMBB reported significantly more Drive for                            Steroid Use
Bulk and than MBN (P < 0.001), but CMBB did not report
                                                                               The prevalence of admitted AAS use among MBB was 40%
higher scores than RMBB. Interestingly, there were no signif-
                                                                               (21 of 52), a rate significantly higher than the 4% (1 of 22)
icant differences among groups on Drive for Muscle Tone.
                                                                               reported for MBN. As expected, steroid use was more preva-
                                                                               lent in CMBB compared with RMBB (59% compared with
Psychological Factors                                                          24%; c2 = 6.1, df 1; P < 0.05).
As shown in Table 2, MANOVA revealed a significant main
effect of Group (F 12,130 = 4.7; P < 0.001). ANOVA and post
hoc comparisons using Tukey’s tests revealed MBN reported
significantly higher scores on all psychological factors                       Owing to the exceptionally lean and muscular standards that
(except perfectionism), compared with either bodybuilder                       currently exist for CMBB, we hypothesized that CMBB
group. CMBB and RMBB did not differ significantly on any                       would exhibit more body dissatisfaction and extreme eating
of the general psychological variables.                                        and weight-control practices, compared with RMBB, but less
                                                                               than those with BN. This hypothesis was partially confirmed,
                                                                               but fewer differences than expected were obtained between
Prevalence of Eating Pathology in MBBs                                         MBN and CMBB, as well as between bodybuilding condi-
Although many eating disorder symptoms exhibited by MBB                        tions. Nevertheless, several important differences among
were significantly less prevalent in comparison with MBN,                      bodybuilding groups, as well as similarities between MBN
eating disturbances and preoccupation with weight and shape                    and MBB, were found.

164                                                                                                 W Can J Psychiatry, Vol 51, No 3, March 2006
                                                                             Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders

  Table 3 Prevalence of bulimia nervosa and bulimic symptoms

                                                                    MBN                          CMBB                         RMBB
                                                                    n = 22                       n = 27                       n = 25

  Characteristic                                               n               %             n             %             n              %

  Have ever met criteria for BNa,b,c                           22             100            8            29.6           2             8.0

  Have ever met criteria for EDNOS                             0               0             1            3.5            3             12.0
  Have ever binged                                             22             100           13            48.0           5             20.0

       Depressed after bingea,b                                17             77.3           5            38.5           1             20.0
       Angry after binge                                       21             95.5           8            61.5           2             40.0

  One binge weekly in the last 3 monthsa,b,c                   22             100            9            33.3           4             16.0
  Two binges weekly in the last 3 months                       9              40.9           5            18.5           3             12.0

  Worried eating too much and (or) being too fat               22             100           17            63.0           18            72.0

  Overconcerned with shape and (or) weight                     22             100           22            84.6           20            80.0

  Body shape affects self-esteema,b                            22             100           16            61.5           16            64.0

  Body shape as important as friends and work                  19             86.4          20            76.9           16            64.0

  Regular use of weight-loss methods in past

       Vigorous exercise                                       19             86.4          15            55.5           15            60.0

       Strict dieting                                          17             77.3          14            52.5           13            52.0

       Diuretics                                               3              13.6           5            18.5           2             8.0

       Laxativesa,b                                            6              28.6           1            4.0            2             8.0
       Vomiting                                                21             95.5           0            0.0            0             0.0
   MBN compared with CMBB, P < 0.05; bMBN compared with RMBB, P < 0.05; cCMBBs compared with RMBB, P < 0.05. EDNOS = Eating disorder not oth-
  erwise specified

Perhaps the most striking and unique finding is that almost                  in their sample of MBB (20), Pope and others found that 3%
30% of CMBB met criteria for BN at some point in their life-                 of MBB reported a history of AN (31), a rate markedly (150
time, a rate that is significantly higher than the 8% reported by            times) higher than the 0.02% rate typically found in North
RMBB, and also markedly higher than rates reported for ath-                  American men (30).
letic male groups (29) and men in the population, according to
self-reported diagnoses (30). The high prevalence of BN in                   Lifetime rates of binge eating and prevalence of binge eating
the RMBB group indicates that body dissatisfaction and                       (once weekly) in the 3 months preceding testing were higher,
engaging in unhealthy eating and weight-control practices in                 as predicted, in CMBB compared with RMBB. Restraint the-
pursuit of the lean and muscular male ideal cannot be attrib-                ory predicts that strict dieting, either by avoiding forbidden
uted solely to competition but may also be motivated more by                 foods or by caloric restriction—both common in competitive
personal reasons relating to overvaluing weight and shape.                   bodybuilding to reduce body fat during the season—would
Diagnostic criteria for BN were based on the DSM-III-R, but                  predispose CMBB to binge eating by increasing deprivation
careful examination of data reveals that all CMBB would have                 of desired high-calorie or high-carbohydrate foods or
also met the DSM-IV diagnostic criteria for BN (6 with purg-                 decreasing sensitivity to internal cues such as hunger and sati-
ing subtype and 2 with nonpurging type). Interestingly, while                ation (21). Although strict dieting was highly prevalent in this
Mangweth and others found no incidence of eating disorders                   group, rates were just as high for RMBB; thus, the elevated

Can J Psychiatry, Vol 51, No 3, March 2006 W                                                                                                    165
The Canadian Journal of Psychiatry—Original Research

rates of binge eating seen in CMBB are not likely to be attrib-     fat, highlighting the severity of eating pathology in MBB. On
uted to strict dieting alone. Depression and negative affect are    a related note, despite higher AAS use in CMBB, as expected,
additional psychological factors implicated in the precipita-       almost 25% of RMBB used AAS even though they had no
tion and maintenance of binge eating (32). While CMBB did           desire to enter competitions. This provides more evidence that
not score in the clinical range for depression, they were almost    the RMBB motivation for these unhealthy eating and body
twice as likely to feel depressed after binge eating. Thus it is    modification practices was to attain a more attractive body for
possible that feeling depressed after binge eating, perhaps         cosmetic, rather than competitive, purposes. The unexpect-
owing to a loss of control over eating and (or) anticipation of     edly high acuity of eating-related symptoms exhibited by the
weight gain in fat, predisposed CMBB to subsequent binges,          RMBB likely accounted for the lack of differences between
but this hypothesis needs to be explored in future research.        CMBB and RMBBs observed in the present study.
Important similarities and differences in body image and            MBN exhibited far more general psychopathology than
weight and shape preoccupation across groups were found. A          CMBB and RMBB, but no differences were found among
unique and interesting finding was that no differences among        groups in terms of perfectionism, which was elevated in all
MBN and bodybuilding groups were found in the prevalence            groups. These findings are partially consistent with previous
of being overconcerned with weight and shape or ratings that        research (18,20). For example, Mangweth and others, found
body shape was as important as friends or work. Moreover, no        higher rates of mood disorders in men with eating disorders,
group differences were found for the desire to obtain a more        compared with MBB, but no differences among these groups
toned and lean body, indicating that MBB, who were already          in rates of alcohol or drug abuse, anxiety disorders, or sexual
large with toned musculature, were as dissatisfied with their       dysfunction (20). Davis and Robertson also found no differ-
muscle tone as MBN. This marked acuity of body dissatisfac-         ences between women with eating disorders and MBB on nar-
tion and preoccupation with weight and shape observed in            cissism, obssessionality, or anhedonia (18), suggesting that
MBB was associated with AAS use, reflecting a constellation         some MBB not only share unhealthy eating and weight con-
of symptoms and behaviour that is consistent with the muscle        trol practices as those with BN, but also exhibit some overlap
dysmorphia syndrome described by Pope and others (12).              in psychological characteristics.
Pope and others (31) hypothesized that MBB may be at
greater risk for body dysmorphic symptoms as a whole and            This study has several limitations. Ideally, MBN would have
that sociocultural factors at a particular time may determine       been recruited from the community to match recruitment
whether they move in the anorexic or “reverse anorexic”             methods for MBB, but the low prevalence of eating disorders
direction.                                                          in men would have made it difficult to achieve the necessary
                                                                    sample size required for comparison with MBB. Thus we
As expected, differences were found in the way body dissatis-       relied on using a clinical sample of MBN, and these differ-
faction was manifested between MBN and MBB. MBN                     ences in recruitment may have biased the results. Typically,
scored higher on the Body Dissatisfaction subscale of the           treatment seekers are more symptomatic and have more psy-
EDI, which emphasizes feeling too fat and a desire to be slim-      chiatric comorbidity than those meeting eating disorder diag-
mer. CMBB and RMBB scored higher on the Drive for Bulk              nostic criteria who do not seek treatment (35,36), which may
scale, indicating a perception of being smaller than ideal and a    explain, at least in part, the differences among the MBN group
strong desire to enhance the size of various upper body parts       and MBB groups. However, a lack of differences emerged on
that reflect masculinity (for example, shoulders, biceps, and       some eating-related measures between MBB and MBN,
chest). Similar results were obtained in previous                   underscoring the high degree of eating-related psycho-
research (18,20). The discrepancy in Drive for Bulk but simi-       pathology exhibited by MBB. Homosexual orientation has
larity in Drive for Muscle Tone may explain why many MBB            been identified as a risk factor in men with eating dis-
simultaneously engage in bulimic weight-control practices           orders (37) but was not measured in this study; thus, the
and AAS use, while MBN engage in bulimic weight-control             effects of sexual orientation on eating pathology in MBB
practices without the concomitant use of steroids. In this con-     remain unknown. Diagnostic criteria for BN in MBB was
text, MBB who endorse both traditional bulimic weight-              established by self-report without a confirmatory clinical
control practices in conjunction with AAS use may be at             interview; thus, the prevalence and severity of binge eating
greater risk to suffer adverse medical and psychiatric conse-       and abnormal weight-control practices may be overestimated
quences of steroid use (33,34), compared with MBN alone.            (38, 39) or underestimated (40,41) when compared with struc-
Although MBN were more likely to purge via laxatives and            tured clinical interviews. Nevertheless, the prevalence of BN
vomiting, it was surprising that no differences were found          and related symptoms was much higher in CMBB than rates
between MBN and MBB in the lifetime prevalence of using             found in other athletic and nonathletic male populations that
vigorous exercise, strict dieting, or diuretics to lose weight or   used self-report, indicating that this growing subgroup of men

166                                                                                    W Can J Psychiatry, Vol 51, No 3, March 2006
                                                                                    Body Image, Binge Eating, and Bulimia Nervosa in Male Bodybuilders

                                                                                    10. Pope HG, Jr, Gruber AJ, Choi P, Olivardia R, Phillips KA. Muscle dysmorphia.
exhibits substantial psychiatric morbidity, which warrants                              An underrecognized form of body dysmorphic disorder. Psychosomatics
more careful study. Finally, the sample of MBB was relatively                           1997;38:548–57.
                                                                                    11. Brownell KD, Rodin J, Wilmore J. Eating, body weight and performance in
small; thus, results may not be representative of the                                   athletes: disorders of modern society. Philadelphia (PA): Lea and Febiger; 1992.
bodybuilding population, though findings are generally                              12. Klein AM. Fear and self-loathing in Southern California: narcissism and fascism
                                                                                        in bodybuilding subculture. J Psychoanal Anthropol 1987;10:117–37.
consistent with previous research.                                                  13. Andersen RE, Barlett SJ, Morgan GD, Brownell KD. Weight loss,
                                                                                        psychological, and nutritional patterns in competitive male body builders. Int J
In summary, body dissatisfaction, weight and shape preoccu-                             Eating Disord 1995;18:49–57.
pation, strict dieting, AAS use, binge eating, and a history of                     14. Blouin AG, Goldfield GS. Body image and steroid use in male bodybuilders.
                                                                                        Int J Eating Disord 1995;18:159–65.
full-blown BN were highly prevalent in CMBB and, to a                               15. Loosemore, DJ, Moriarty, D. Body dissatisfaction and body image distortion in
lesser extent, in RMBB. Moreover, MBB exhibited levels of                               selected group of males. CAHPER 1990; 11:11–5.
                                                                                    16. Pasman L, Thompson JK. Body image and eating disturbance in obligatory
weight and shape preoccupation, body dissatisfaction, dis-                              runners, obligatory weightlifters, and sedentary individuals. Int J Eating Disord
ordered eating, and perfectionism similar to MBN, indicating                        17. Anderson SL, Zager K, Hetzler RK, Nahikian-Nelms M, Syler G. Comparison of
that the current group of MBB share many eating-related fea-                            Eating Disorder Inventory (EDI-2) scores of male bodybuilders to the male
                                                                                        college student subgroup. Int J Sport Nutr 1996;6:255–62.
tures but fewer general psychological characteristics with                          18. Davis C, Scott-Robertson L. A psychological comparison of females with
those having BN. Taken together, our findings suggest that                              anorexia nervosa and competitive male bodybuilders: body shape ideals in the
                                                                                        extreme. Eating Behaviours 2000;1:33–46.
there is a subgroup of MBB, especially those who have com-                          19. Rubinstein G. Macho man: Narcissism, homophobia, agency, communion, and
peted, who may be at increased risk of developing unhealthy                             authoritarianism—a comparative study among Israeli bodybuilders and a control
                                                                                        group. Psychol Men and Mascul 2003;4:100–10.
eating and weight-control practices, including binge eating or                      20. Mangweth B, Pope HG, Jr, Kemmler G, Hausmann A, De Col C, Kreutner B,
full-blown BN. Longitudinal research is needed to determine                             and others. Body image and psychopathology in male bodybuilders. Psychother
                                                                                        Psychosom 2001;70:38–43.
whether men with a history of BN or subclinical manifesta-                          21. Polivy J, Herman CP. Dieting and bingeing: a causal analysis. Am Psychol
tions of BN disproportionately gravitate toward bodybuilding                            1985;40:193–201.
                                                                                    22. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for
and AAS use as a vehicle to meet personal or societal stan-                             measuring depression. Arch Gen Psychiatry 1961;4:561–71.
dards of attractiveness and (or) whether the pursuit of the                         23. Garner DM, Olmstead MP, Polivy J. Development and validation of a
                                                                                        multidimensional eating disorder inventory for anorexia nervosa and bulimia. Int
exceptionally lean and hypermesomorphic bodybuilding                                    J Eating Disord 1983;2:15–34.
                                                                                    24. Blouin AG, Perez EL, Blouin JH. Computerized administration of the Diagnostic
ideal fosters bulimic attitudes, eating behaviours, and AAS                             Interview Schedule. Psychiatr Res 1988;23:335–44.
use.                                                                                25. Robins LN, Helzer JE, Croughan J, Ratcliff KS. National Institute of Mental
                                                                                        Health Diagnostic Interview Schedule. Its history, characteristics, and validity.
                                                                                        Arch Gen Psychiatry 1981;38:381–9.
                     Funding and Support                                            26. Erdman HP, Klein MH, Greist JH, Skare SS, Hustel JJ, Robins C, and others.
                                                                                        A comparison of two computer-administered versions of the NIMH Diagnostic
This study was funded, in part, by an Applied Sport Research                            Interview Schedule. J Psychiatr Res 1992;26:85–95.
grant from Sport Canada awarded to Dr Blouin and Dr Goldfield,                      27. Chng M, Moore A. A study of steroid use among athletes: knowledge, attitude,
as well as a New Investigator award from the Canadian Institutes                        and use. Health Educ 1990; 21(6):12–7.
of Health Research awarded to Dr Goldfield.                                         28. Goldfield GS. Risk factors in anabolic steroid use inmale bodybuilders: A case
                                                                                        control study. Forthcoming.
                                                                                    29. Brownell KD, Rodin J. Prevalence of eating disorders in athletes. In: Brownell
                      Acknowledgements                                                  KD, Rodin J, Wilmore JH, editors. Eating, body weight, and performance in
                                                                                        athletes. Philadelphia (PA): Lea and Febiger; 1992. p 128–45.
The authors thank Paula Cloutier, Stephanie Leclaire, Dr Hollie                     30. Hoek HW, van Hoeken D. Review of the prevalence and incidence of eating
Raynor, and Dr Paddi O’Hara for their assistance in preparation of                      disorders. Int J Eating Disord 2003;34:383–96.
this manuscript.                                                                    31. Pope HG, Jr, Katz DL, Hudson JI. Anorexia nervosa and “reverse anorexia”
                                                                                        among 108 male bodybuilders. Compr Psychiatry 1993;34:406–9.
                                                                                    32. Polivy J, Herman CP. Etiology of binge eating: psychological mechanisms.
                              References                                                In: Fairburn CG, Wilson GT, editors. Binge eating: nature, assessment, and
                                                                                        treatment. New York (NY): The Guilford Press; 1993. p 173–205.
                                                                                    33. Pope HG, Jr, Katz DL. Affective and psychotic symptoms associated with
 1. Garner, D. M. The Body Image Survey. Psychology Today                               anabolic steroid use. Am J Psychiatry 1988;145:487–90.
    1997;(January/February):32– 84.                                                 34. Pope HG, Katz DL. Psychiatric and medical effects of anabolic-androgenic
 2. Ogden J, Mundray K. The effect of the media on body satisfaction: The role of       steroid use: a controlled study of 160 athletes. Arch Gen Psychiatry
    gender and size. Euro Eat Disord Rev 1996;4:171– 82.                                1994;51:375–82.
 3. Andersen AE, DiDomenico L. Diet vs shape content of popular male and female     35. Fairburn CG, Welch SL, Norman PA, O’Connor ME, Doll HA. Bias and bulimia
    magazines: A dose response relationship to the incidence of eating disorders?       nervosa: how typical are clinic cases? Am J Psychiatry 1996;153:386–91.
    Int J Eating Disord 1992;11:283–7.                                              36. Perkins PS, Klump KL, Iacono WG, McGue M. Personality traits in women with
 4. Cohane GH, Pope HG, Jr. Body image in boys: a review of the literature. Int J       anorexia nervosa: evidence for a treatment-seeking bias? Int J Eating Disord
    Eating Disord 2001;29:373–9.                                                        2005;37:32–7.
 5. Leit RA, Gray JJ, Pope HG, Jr. The media’s representation of the ideal male     37. Carlat DJ, Camargo CA, Jr. Review of bulimia nervosa in males. Am J
    body: a cause for muscle dysmorphia? Int J Eating Disord. 2002;31:334 –8.           Psychiatry 1991;148:831–43.
 6. Pope HG, Jr. Evolving ideals of male body image as seen through action toys.    38. Black CM, Wilson GT. Assessment of eating disorders: interview versus
    Int J Eating Disord 1999;26:65–72.                                                  questionnaire. Int J Eating Disord 1996;20:43–50.
 7. Rodin J, Silberstein L, Striegel-Moore R. Women and weight: A normative         39. Fairburn CG, Beglin SJ. Assessment of eating disorders: interview or self-report
    discontent. Nebr Symp Motiv; 1984;32:267–307.                                       questionnaire? Int J Eating Disord 1994;16:363–70.
 8. American Psychiatric Association. Diagnostic and statistical manual of mental   40. Carter JC, Aime AA, Mills JS. Assessment of bulimia nervosa: a comparison of
    disorders, 4th ed. Washington (DC): American Psychiatric Association; 1994.         interview and self-report questionnaire methods. Int J Eating Disord
 9. Kanayama G, Pope HG, Jr, Hudson JI. “Body image” drugs: a growing                   2001;30:187–92.
    psychosomatic problem. Psychother Psychosom 2001;70:61–5.                       41. Wilfley DE, Schwartz MB, Spurrell EB, Fairburn CG. Assessing the specific
                                                                                        psychopathology of binge eating disorder patients: interview or self-report?
                                                                                        Behav Res Therapy 1997;35:1151–9.

Can J Psychiatry, Vol 51, No 3, March 2006 W                                                                                                                        167
The Canadian Journal of Psychiatry—Original Research

                                                                              Associate Professor, Department of Psychiatry, University of Toronto,
Manuscript received April 2005, revised, and accepted August 2005.           Toronto, Ontario.
  Investigator, Children=s Hospital of Eastern Ontario Research Institute,   Address for Correspondence: Dr GS Goldfield, CHEO Research Institute,
Mental Health Research, Ottawa, Ontario; Adjunct Research Professor,         Mental Health Research, 401 Smyth Road, Ottawa, ON, K1H 8L1,
Department of Human Kinetics, University of Ottawa, School of Human
Kinetics, Ottawa, Ontario.
   Adjunct Research Professor, Department of Psychology, Carleton
University, Ottawa, Ontario.

                 Résumé : L’image corporelle, l’alimentation excessive et la boulimie chez les
                 culturistes masculins
                 Objectif : Les culturistes masculins (CM) présentent une insatisfaction corporelle, un
                 comportement alimentaire boulimique et des caractéristiques psychologiques négatives plus graves,
                 comparés à des sujets témoins masculins athlétiques et non athlétiques, mais peu d’études ont
                 comparé directement les CM avec des hommes souffrant de troubles alimentaires. Cette étude a
                 comparé des hommes souffrant de boulimie (HB), des hommes culturistes de compétition (HCC) et
                 des hommes culturistes récréatifs (HCR) relativement à une vaste gamme d’habitudes et de
                 comportements alimentaires, et de caractéristiques psychologiques pour déterminer plus
                 précisément les similitudes et les différences entre ces groupes.
                 Méthode : Des questionnaires anonymes, destinés à évaluer les habitudes alimentaires, l’image
                 corporelle, les préoccupations de poids et de forme, la prévalence des épisodes d’alimentation
                 excessive, les méthodes de perte de poids, les taux à vie de troubles alimentaires, l’utilisation de
                 stéroïdes anabolisants androgéniques (SAA), et les facteurs psychologiques généraux ont été
                 remplis par 22 HB, 27 HCC, et 25 HCR.
                 Résultats : Des taux élevés de préoccupations de poids et de forme, des méthodes extrêmes de
                 modification du corps, des épisodes d’alimentation excessive et la boulimie ont été déclarés par les
                 HC, en particulier par ceux qui étaient en compétition. Les HCC ont déclaré des taux plus élevés
                 d’épisodes d’alimentation excessive, de boulimie et d’utilisation de SAA, comparativement aux
                 HCR, mais présentaient moins de psychopathologie liée à l’alimentation et générale,
                 comparativement aux HB. Peu de différences psychologiques ont été observées entre les HCC et les
                 Conclusions : Les HC, surtout les compétiteurs, et les HB semblent partager de nombreux traits liés
                 à l’alimentation, mais peu de caractéristiques psychologiques. Une étude longitudinale est
                 nécessaire pour déterminer si les hommes ayant des antécédents d’alimentation déréglée ou de
                 boulimie s’adonnent à la compétition culturiste de façon disproportionnée, ou si le culturisme de
                 compétition favorise l’alimentation déréglée, la boulimie et l’utilisation de SAA.

168                                                                                                W Can J Psychiatry, Vol 51, No 3, March 2006

Description: symptoms-of-bulimia pdf