An exploratory investigation of purging disorder

Eating Behaviors 14 (2013) 26–34

Contents lists available at SciVerse ScienceDirect

Eating Behaviors

An exploratory investigation of purging disorder Kathryn E. Smith ⁎, Janis H. Crowther Kent State University, Kent, OH 44242, United States

a r t i c l e

i n f o

Article history: Received 14 March 2012 Received in revised form 23 July 2012 Accepted 3 October 2012 Available online 9 October 2012 Keywords: Purging Disorder Eating Disorder Not Otherwise Specified (EDNOS) Bulimia Nervosa Dietary restraint Body image

a b s t r a c t Objective: Purging Disorder (PD) is an understudied pattern of behaviors within the Eating Disorder Not Otherwise Specified (EDNOS) category. Such categorization may suggest that PD is not clinically significant as other eating disorders. However, evidence has suggested that PD is associated with significant impairments in psychosocial functioning and well-being. Despite the apparent clinical significance of PD, it remains to be determined if PD is distinct from other clinically significant eating disorders. The present study sought to assess the phenomenology, clinical significance, and distinctiveness of PD. Method: Group scores on measures of eating pathology, body image disturbance, and psychological correlates were compared using MANOVA among a female undergraduate sample (N = 94) meeting diagnostic criteria for PD (n = 20), Bulimia Nervosa (BN; n = 35), restrained eating (n = 18), and healthy controls (n = 21). Results: Overall, results indicated the PD group reported less severe symptoms than BN but more severe symptoms than controls. The PD and restraint groups were similar on most variables (including subjective binge behavior), with the exception of perfectionism and hunger. Discussion: Findings support the conceptualization of PD as existing along a spectrum of bulimic spectrum disorders rather than as a distinct diagnostic category. © 2012 Elsevier Ltd. All rights reserved.

1. Introduction Although there is a general consensus among psychologists that Anorexia Nervosa (AN) and Bulimia Nervosa (BN) are clinically significant and distinct categories of eating disorders, most individuals who have clinically significant symptoms of eating pathology do not meet the diagnostic criteria for either diagnosis (Keel, Haedt, and Edler, 2005; Machado, Machado, Goncalves, and Hoek, 2007; Wade, Bergin, Tiggemann, Bulik, and Fairburn, 2006). Thus, such individuals most often receive a diagnosis of an Eating Disorder Not Otherwise Specified (EDNOS), which is a category that describes “disorders of eating that do not meet the criteria for any specific eating disorder” (American Psychiatric Association, 2000, p. 594). However, many experts have recognized the problematic nature of this broad categorization, as this is a group of highly heterogeneous patterns of behaviors that are believed to have clinical significance (Crow, 2007; Fairburn and Cooper, 2007; Fairburn et al., 2007; Machado et al., 2007). 1.1. Defining Purging Disorder Purging Disorder (PD) is one EDNOS category that has been the focus of increasing study. Currently PD is classified as “EDNOS” or ⁎ Corresponding author at: 144 Kent Hall, Kent State University, Kent, OH 44242, United States. Tel.: + 1 330 672 2166; fax: + 1 330 672 3786. E-mail address: [email protected] (K.E. Smith). 1471-0153/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eatbeh.2012.10.006

“EDNOS-P” in the DSM-IV-TR, yet recent literature suggests that it may be a distinct and clinically significant disorder, with prevalence rates comparable to other eating disorders (Crowther, Armey, Luce, Dalton, and Leahey, 2008; Favaro, Ferrara, and Santonastaso, 2003; Haedt and Keel, 2010; Keel et al., 2005; Wade et al., 2006). In a review of the published literature on PD, Keel and Striegel-Moore (2009) proposed five diagnostic criteria for PD: (1) recurrent purging in order to influence weight or shape (e.g., self-induced vomiting, laxative abuse, enemas, diuretics), (2) purging occurs, on average, at least once a week for three months, (3) self-evaluation is unduly influenced by body shape or weight or there is an intense fear of gaining weight or becoming fat, (4) the purging is not associated with objectively large binge episodes, and (5) the purging does not occur exclusively during the course of anorexia nervosa or bulimia nervosa. Additionally, in an investigation of the syndrome validity of PD, Haedt and Keel (2010) varied the diagnostic criteria by type of compensatory behavior (purging vs. non-purging) and frequency (once vs. twice per week). Results indicated that distinguishing between purging and non-purging behavior (i.e., excluding non-purging behavior from PD diagnostic criteria) was associated with larger effect sizes when PD was compared to healthy controls on external validators (e.g., psychosocial functioning, perfectionism, substance use). However, reducing the minimum frequency of purging behavior from twice to once per week was associated with similar effect sizes when PD was compared to healthy controls on these variables. Thus, this research supports the diagnostic criteria proposed above (Keel and Striegel-Moore, 2009).

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

27

Recent research provides evidence for the clinical significance of PD. Compared to individuals without PD, individuals with PD show increased levels of impairment, including higher levels of general psychopathology, distress, eating pathology, and personality disorders (Keel, Wolfe, Gravener, and Jimerson, 2008; Keel et al., 2005). However, research has yielded mixed findings regarding how individuals with PD compare to individuals with other eating disorder diagnoses, particularly BN. For example, some studies have demonstrated that women with PD do not differ significantly from those with BN on measures of symptom severity, impairment, body dissatisfaction, or dietary restraint (Binford and le Grange, 2005; Keel, Mayer, and Harnden-Fischer, 2001; Keel et al., 2005). In contrast, other literature suggests that compared to individuals with PD, individuals with BN generally report greater shape/weight/eating concerns (Binford and le Grange, 2005), lower levels of self-esteem (Binford and le Grange, 2005), and greater levels of psychopathology, including current mood disorders (Keel et al., 2005, 2008), anxiety levels (Fink, Smith, Gordon, Holm-Denoma, and Joiner, 2009), and impulsive behaviors (Fink et al., 2009; Keel et al., 2001), although other studies have failed to replicate the findings regarding impulsivity (Keel et al., 2005, 2008). These discrepant findings may be due to variations in the criteria used to define PD and a lack of statistical power to find significant differences between PD and other forms of eating pathology. Given these mixed findings and the lack of theoretical conceptualizations of PD, the present study further explored the phenomenology of PD within the framework of empirically supported theories of eating pathology. This included an examination of constructs that may inform the conceptualization of PD, including dietary restraint, subjective and objective binge episodes, and variables which have not previously been compared between BN and PD (e.g., perfectionism, emotion regulation).

restrained eating increase the likelihood of subsequent binge eating (Herman and Polivy, 1975, 1980, 1983; Polivy and Herman, 1985). In an effort to lose weight, restrained eaters adhere to unrealistic, rigid dietary rules in order to restrict their food intake (Polivy and Herman, 1985). Rigid dietary restraint, when coupled with biological (e.g., starvation effects), cognitive (e.g., dichotomous thinking), and affective (e.g., mood fluctuations) factors related to dieting, may lead dieters to feel a loss of control after any lapse in their diet. This increases the likelihood that the person will temporarily abandon all dietary rules and engage in binge episodes (Wilson, 2002). In support of this theory, empirical evidence suggests that dietary restraint may potentiate binge eating and contribute to the development of eating disorders (Lowe et al., 1996; Stice, 2002; Wilson, 2002). Two findings are particularly relevant. First, using ecological momentary assessment, Engelberg et al. (2005) found that restraint was elevated before binge cravings, but not binge episodes. Such findings support restraint theory, in that dietary restraint potentiates, but does not directly cause, binge eating. Rather, other factors, such as emotional distress, may trigger binge episodes. Thus, in PD, it is possible that restraint is associated with binge cravings, but not necessarily objectively large binge episodes unless other factors are present. Second, Kerzhnerman and Lowe (2002) examined dieting intensity and the frequency of objective and subjective binge episodes; results indicated higher levels of dieting intensity were related to more frequent subjective, yet not objective, binge episodes. This finding suggests that the relationship between dietary restriction and subsequent perceived disinhibited eating (i.e., SBEs) is stronger and more significant than the relationship between dietary restriction and objective caloric consumption (i.e., OBEs).

1.2. Conceptualizing PD

1.3. The present study

Though mounting evidence suggests that PD may be a clinically significant and unique pattern of behavior, it is unclear what factors precipitate the purging behavior. In the purging subtype of BN, there is a consensus that objective binge episodes (OBEs) usually precede purging behavior (Stice, 2001); thus, purging behaviors are generally studied in conjunction with binge episodes. However, consistent with PD diagnostic criteria (Keel and Striegel-Moore, 2009), individuals with PD do not engage in OBEs, which potentially challenges the existing theories that conceptualize purging behavior as a response to OBEs. While OBEs require both the consumption of an objectively large quantity of food and a sense of loss of control, subjective binge episodes (SBEs), which are not addressed in current BN diagnostic criteria, only require that the individual experiences a sense of loss of control over eating. Thus, it is possible that SBEs precede purging in PD, that is, individuals with PD may experience a loss of control when they consume an objectively moderate amount of food. On the basis of clinical reports, Fairburn and Garner (1986) have posited that the perception of excessive consumption and the perception of loss of control may be more important than the actual amount of food consumed. Moreover, there is some evidence that the distinction between SBEs and OBEs may not be clinically significant (Mond et al., 2006; Niego, Pratt, and Agras, 1997; Pratt, Niego, and Agras, 1998). Because studies have yielded inconclusive findings regarding the presence of SBEs in PD (Keel et al., 2001), research should investigate the frequency and phenomenology (i.e., individuals' subjective experiences) of such episodes in PD. The identification of possible antecedents to purging in PD may be elucidated by Restraint Theory, which provides a unifying, empirically supported conceptualization of eating pathology (Engelberg, Gauvin, and Steiger, 2005; Polivy and Herman, 1985; Wilson, 2002). Although studies have evidenced significant dietary restraint among individuals with PD (e.g., Keel, Holm-Denoma, and Crosby, 2011), thus far, no studies have conceptualized PD within restraint theory, which postulates that both the physiological and psychological aspects of

Taken together, it appears that the validity of PD as a distinct category remains unclear. In theory and consistent with previous research, the clinical significance of PD would be evidenced by differences between individuals with PD and healthy controls on various measures of eating pathology and psychosocial variables, while the distinctiveness of PD would be evidenced by relative differences between individuals with PD and other eating disorders. Given the mixed findings of previous studies, it appears that two possibilities exist regarding the significance and distinctiveness of PD from EDNOS. First, PD may be a clinically significant and distinct diagnostic category. That is, the clinical significance of PD would be evidenced by significant differences between PD and control groups on a variety of variables, with PD individuals reporting greater psychopathology than controls. The distinctiveness of PD would be supported by significant differences between PD and both BN and restrained eaters, including a lack of reported SBEs in PD. The latter finding may suggest a different etiology of PD, as this would not support Restraint Theory and current conceptualizations of bulimic spectrum eating psychopathology. Second, PD may not be a distinct disorder, but rather a disorder that falls on the spectrum of bulimic symptomatology. This conceptualization would be supported by the finding that individuals with PD report subjective binge episodes and differ from controls on variables that are also associated with BN and restrained eating, which would support the possibility that PD exists along a continuum of bulimic eating psychopathology. Thus, the present study sought to compare the core eating disorder symptomatology, body image disturbance, and other psychological correlates of individuals with PD to those of healthy controls, restrained eaters, and individuals with BN. A second purpose of this study was to evaluate the conceptualization of PD within restraint theory by including participants with a range of bulimic spectrum eating pathology and assessing levels of restraint, SBEs, and OBEs. Given the state of the current literature, no specific hypotheses are presented.

28

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

2. Methods 2.1. Participants Participants were 94 undergraduate female students at a large Midwestern university who were recruited following their participation in a large mass testing session. During this session, participants completed the Eating Disorder Diagnostic Scale (EDDS), a brief diagnostic measure of eating pathology (Stice, Telch, and Rizvi, 2000). The EDDS was used to identify potential participants who met criteria for Bulimia Nervosa (BN, both purging and non-purging) and Purging Disorder (PD). The Revised Restraint Scale (RRS) was used as a screener to identify individuals who exhibited dietary restraint (Herman and Polivy, 1975). Additionally, a random sample of individuals who did not report any symptoms of eating pathology on the EDDS was recruited from the same mass testing sample to serve as a healthy control group. Participants were classified into four groups (i.e., BN, PD, Restrained Eater, Control) based on the diagnosis generated by the EDDS algorithm (Stice, Fisher, and Martinez, 2004) and the RRS. Participants were assigned to the BN group if their responses on the EDDS met criteria for a diagnosis of BN according to the algorithm (at least two objectively large binge episodes per week, at least two instances of compensatory behavior per week over the past three months, and an undue influence of weight and/or shape on self-evaluation). Based on previous findings regarding the definition of PD (Keel and Striegel-Moore, 2009), participants were assigned to the PD group if they (1) did not meet criteria for AN or BN according to the EDDS algorithm, (2) reported at least one episode of self-induced vomiting and/or laxative abuse per week over the past three months, and (3) reported an undue influence of weight and/or shape on self-evaluation. Participants were assigned to the Restrained Eater group if they (1) did not meet the aforementioned criteria for inclusion in the BN or PD groups, or any other clinical or subclinical diagnostic groups defined by the EDDS algorithm, (2) had scores of at least 14 on the RRS (a score of 14 or above has typically been used to discriminate between restrained and unrestrained eaters) (Jarry, Polivy, Herman, Arrowood, and Pliner, 2006), and (3) had a BMI less than 24.9, so as to exclude overweight participants who may be engaging in dietary restraint for health purposes. Lastly, participants were assigned to the Control group if they (1) were classified as asymptomatic according to Stice et al.'s algorithm, and (2) scored below a 14 on the RRS. All of the eligible BN and PD individuals were invited to participate, while individuals who met criteria for the Restrained Eater group and control groups were recruited randomly from the mass testing pool. The resulting sample consisted of 35 BN participants, 20 PD participants, 18 restrained eaters, and 21 healthy controls. Participants' mean age was 19.77 years (SD= 4.22), and their mean Body Mass Index (BMI) was 23.70 (SD= 4.98). The majority (96.7%) of participants were in their first, second, third, or fourth year of their undergraduate degree. The sample consisted of participants who identified as Caucasian (90.4%), African American (5.3%), Asian/Asian American (2.1%), and Hispanic/Latina (2.1%). 2.2. Procedure After obtaining participants' informed consent, participants completed a packet of self-report questionnaires, which included the measures below. The Institutional Review Board at the university approved this research project, and participants received research credits toward their undergraduate psychology course for participation.

the last three to six months. The symptom composite score can be used as an overall assessment of eating pathology, with higher scores indicating more severe pathology. Stice et al. (2004) found that the EDDS demonstrated adequate criterion validity with interviewbased diagnoses, as assessed by the Eating Disorder Examination (EDE). Stice et al. reported a sensitivity of .88 (i.e., proportion of individuals with a positive EDE interview diagnosis who were correctly identified by the EDDS) and a specificity of .98 (i.e., proportion of individuals with a negative interview diagnosis who were correctly identified by the EDDS); additionally, the EDDS demonstrated concurrent validity of .74 (i.e., proportion of individuals classified with a positive diagnosis by the EDDS who met criteria for the diagnosis on the EDE). The latter finding suggests that the EDDS is likely to over-diagnose as compared to the EDE. Regarding reliability, Stice et al. (2004) reported a mean Cronbach's alpha of .89 for the EDDS symptom composite score among four studies. In the present sample, the Cronbach's alpha for the symptom composite score was .88. 2.3.2. Revised Restraint Scale (RRS; Herman and Polivy, 1975) The RRS contains ten multiple-choice items that assess weight fluctuations, dieting behaviors, and thoughts about eating, with higher scores indicating higher levels of dietary restraint. Previous studies have reported adequate internal consistency of the RRS (Allison, Kalinsk, and Gorman, 1992). For the present sample, the internal consistency was .86. 2.4. Study measures These measures were used to assess conceptually related variables, which were divided into three categories: core eating pathology symptomatology (as measured by the EDDS symptom composite, eating concerns, restraint, disinhibition, and perceived hunger), body image disturbance (as measured by shape and weight concerns, body dissatisfaction, and internalization of sociocultural ideals) and psychological variables (as measured by perfectionism, difficulties in emotion regulation, impulsivity, obsessive compulsiveness, and self-esteem). 2.4.1. Core eating pathology and body image disturbance The Symptom Composite scale of the EDDS (described above) was used as one measure of core eating pathology. 2.4.1.1. Eating Disorder Examination Questionnaire (EDE-Q4; Fairburn and Beglin, 1994). The EDE-Q4 is a 32-item self-report measure that assesses various dimensions of eating pathology. The EDE-Q4 consists of four subscales: Weight Concern, Shape Concern, Eating Concern, and Restraint (with higher values indicating higher levels of the symptom), as well as items that assess both the frequency of binge episodes and compensatory behaviors in the preceding 28 days. Additionally, the present study used Item 15 (“Over the past 28 days, on how many days have episodes of overeating [i.e., you have eaten an usually large amount of food and have had a sense of loss of control at the time]”?) to assess objective binge frequency, and Items 9 and 20 (“Have you had a definite fear of losing control over eating?” and “On what proportion of the times that you have eaten have you felt guilty [felt that you've done wrong] because of its effect on your shape or weight?”) as proxies for subjective binge frequency. Items 9 and 20 were assessed separately to measure SBEs. This research also used the Weight Concern and Shape Concern subscales as measures of body image disturbance, and the Eating Concern subscale as a measure of core eating pathology; for this sample, these scales had internal consistencies of .90, .95, and .89, respectively.

2.3. Screening measures 2.3.1. Eating Disorder Diagnostic Scale (EDDS; Stice et al., 2000) This 22-item self-report measure is both a diagnostic measure of eating disorders and a continuous measure of eating pathology over

2.4.1.2. Three-Factor Eating Questionnaire (TFEQ; Stunkard and Messick, 1985). The TFEQ is a 54-item self-report measure that assesses three aspects of eating pathology: cognitive (dietary) restraint, perceived hunger, and disinhibition, with higher scores indicating greater levels

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

of symptoms. Stunkard and Messick (1985) reported coefficient alpha reliabilities for the TFEQ subscales that ranged from .85 to .92, and research has supported the validity of the three subscales (French, Jeffery, and Wing, 1994; Laessle, Tuschl, Kotthaus, and Prike, 1989; Stunkard and Messick, 1985). The internal consistencies of the TFEQ subscales for this sample were .86 (restraint), .81 (disinhibition), and .80 (hunger). 2.4.1.3. Body Shape Questionnaire (BSQ; Cooper, Taylor, Cooper, and Fairburn, 1987). The BSQ is a 34-item self-report instrument designed to measure concerns about body shape, with higher scores indicating greater concern about one's body image. Cooper et al. (1987) demonstrated good concurrent and discriminate validity of the BSQ, and Evans and Dolan (1993) reported a previous internal consistency of .97. The internal consistency for this sample was .98.

29

expression, with higher scores indicating greater levels of emotional expression and/or impulsivity. Gross and John (1995) reported satisfactory internal consistency of the BEQ subscales, with Cronbach's alpha values ranging from .71 to 76. The present study used the Impulse Strength subscale, which had an internal consistency of .81 for this sample. 2.4.2.6. Difficulties in Emotion Regulation Scale (DERS; Gratz and Roemer, 2004). The DERS is a 36-item self-report measure designed to assess various aspects of emotion dysregulation, with higher scores indicating greater emotional dysregulation. Gratz and Roemer (2004) reported high internal consistency of the DERS composite score (α =.93). The internal consistency of the DERS composite score in this sample was .95. 3. Results

2.4.1.4. Sociocultural Attitudes Towards Appearance Scale-3 (SATAQ-3; Thompson, van, Roehrig, Guarda, and Heinberg, 2004). The 30-item SATAQ-3 is comprised of four subscales: Internalization-General, Information, Pressures, and Internalization-Athlete. Cronbach's alpha values ranged from .92 to .96 among the subscales. The present study used the Internalization-General subscale, which had an internal consistency of .97 for this sample. 2.4.2. Psychological variables 2.4.2.1. Frost Multidimensional Perfectionism Scale (MPS; Frost, Marten, Lahart, and Rosenblate, 1990). The 35-item MPS assesses six dimensions of perfectionism, with higher scores indicating greater levels of perfectionism. Grange et al. (2006) reported an alpha coefficient of .93 for the total MPS score, with subscale alpha coefficients ranging from .79 to .93. The present study used the Concern over Mistakes, Doubts about Action, and Personal Standards subscales, as the former two subscales have been conceptualized as a maladaptive form of perfectionism, and the latter subscale an adaptive form of perfectionism (Bardone-Cone, Weishuhn, and Boyd, 2009; Chang, Watkins, and Banks, 2004). The Concern over Mistakes, Personal Standards, and Doubts about Action subscales had internal consistencies of .91, .85, and .80, respectively. 2.4.2.2. Vancouver Obsessional Compulsive Inventory (VOCI; Thordarson et al., 2004). The 55-item VOCI is based on the Maudsley Obsessional Compulsive Inventory (Rachman, Thordarson, and Randomsky, 1995), and assesses a wide range of obsessions, compulsions, and other behaviors relevant to Obsessive–Compulsive Disorder (OCD). Higher scores indicate greater levels of obsessive compulsive symptomatology. Thordarson et al. (2004) reported excellent internal consistency for the VOCI and its subscales (.79 to .98). The present study used the full scale, which had an alpha coefficient of .96 for this sample. 2.4.2.3. Barratt Impulsivity Scale (BIS-11; Patton, Stanford, and Barratt, 1995). The BIS is a 30-item self-report questionnaire that assesses impulsive behaviors, with higher scores indicating greater levels of impulsiveness. The internal consistency of the composite BIS score for this sample was .83. 2.4.2.4. Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965). The 10-item RSES is a self-report assessment of global self-worth. Higher scores indicate greater levels of self-esteem. Although originally intended for use among adolescents, the RSES has demonstrated good psychometric properties in a variety of populations (Sinclair et al., 2010). The internal consistency of the RSES in this sample was .88. 2.4.2.5. Berkley Expressivity Questionnaire (BEQ; Gross and John, 1995). The BEQ is a 16-item self-report questionnaire that assesses the strength of emotional response tendencies and the degree of emotional

3.1. Sample characteristics Of the participants screened using the EDDS and RRS 4.2% met criteria for BN (43 of 1,016), and 3.6% met criteria for PD (61 of 1,709). Table 1 displays descriptive statistics on demographic variables by group. A one-way analysis of variance (ANOVA) indicated that participants' BMI differed significantly by group, F(3, 90)= 3.27, p b .05; however, a post-hoc Student-Neuman–Keuls analysis revealed no significant between-group differences. (The Student Newman– Keuls test was chosen b/c it is designed to maximize power while providing some control over the family-wise error rate; Howell, 2002). Upon examination of BMI means and effect sizes for group comparisons it appears that the BN and PD participants had higher BMIs than the Restrained Eater and Control groups. There also were no significant group differences for age, F(3, 90)=.56, p>.05. Because of the very low frequency of African American, Asian/American, and Hispanic/Latina participants, ethnicity was recoded into two groups: Caucasian and non-Caucasian. A Chi-square analysis comparing the diagnostic groups by ethnicity was non-significant, χ2(3, N=94)=5.94, p>.05. 3.2. Binge eating and compensatory behavior Table 1 also presents descriptive data and statistical findings for binge episodes and compensatory strategies. In regards to binge episode frequency, an independent t-test compared the frequency of Objective Binge Episodes (OBEs; as assessed by EDE-Q4 Item 15) between the BN and Restrained Eater group. These two groups were compared because neither the PD nor control group reported OBEs on the EDDS. However, given the EDDS scoring algorithm, it was possible that restrained eaters reported a subthreshold frequency (i.e., less than twice per week) of binge episodes. Results indicated that the BN group reported significantly more OBEs than the Restrained Eater group over the last 28 days, t(51) = 2.55, p b .05. Two one-way ANOVAs compared the four groups on fear of loss of control (EDE-Q4 Item 9) and guilt while eating (EDE-Q4 Item 20), which served as proxies for Subjective Binge Episodes (SBEs). Results were significant: Item 9: F(3, 90) = 14.63, p b .001; Item 20: F(3, 90) = 20.82, p b .001. The patterns of post-hoc Student Newman–Keuls analyses for both items were similar, indicating that the BN group reported significantly more SBEs than all other groups, and the PD and Restrained Eater groups also reported significantly more SBEs than the control group. Because both the BN and PD groups engage in purging behaviors, the frequencies of self-induced vomiting and laxative use were assessed by summing participants' responses to Items 16 and 17 on the EDE-Q4 to create the dependent variable of frequency of purging behavior. An independent t-test comparing the BN and PD groups on the frequency of purging behavior over the last 28 days was nonsignificant, t(53) = 1.06, p = .30. Two one-way ANOVAs comparing the

30

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

Table 1 Demographic variables, binge eating, and compensatory strategies: Descriptive data and statistical findings.

Age BMI Exercise Fasting Fear of losing control Guilt over eating

OBEs Purging frequency

Bulimia Nervosa1

Purging Disorder2

Restrained eater3

Control4

n = 35

n = 20

n = 18

n = 21

M (SD)

M (SD)

M (SD)

M (SD)

F

p

ηp2

20.53(4.78) 25.48(6.33) 7.10(7.87) 2.49(2.44) 3.34(1.48) 2.80(2.04)

19.47(3.34) 23.88(4.11) 6.40(9.06) 2.43(2.59) 2.00(1.95) 1.40(1.60)

19.33(5.19) 22.26(1.90) 4.56(4.25) 2.52(4.05) 1.67(1.68) 1.00(1.33)

19.19(2.98) 21.78(4.12) .46(1.47) 0.00(0.00) .14(.46) .48(.22)

.58 3.27 4.70 4.94 20.82 14.63

.63 b.05 b.05 b.05 b.001 b.001

.02 .10 .14 .14 .41 .33

t (df)

p

6.60(7.22) 1.92(4.19)

– 1.05(1.88)

1.94(3.81) –

– –

2.55(53) 1.06(53)

b.05 .30

Between subjects effects

SNK post-hoc results n.s. n.s. 1,2,3,>4 1,2,3,>4 1 > 2,3 > 4 1 > 2,3 > 4

Cohen's d effect size

1vs2

1vs3

1vs4

2vs3

2vs4

3vs4

.26 .30 .08 .02 .77 .76

.24 .69 .40 .01 1.05 1.04

.34 .69 1.17 1.44 2.92 1.60

.03 .51 .26 −.03 .18 .27

.09 .51 .92 1.33 1.31 .81

.03 .15 1.29 .88 1.24 .55

– .27

.81 –

– –

– –

– –

– –

Note. SNK = Student Newman Keuls; BMI = Body Mass Index; SBE = Subjective Binge Episode; OBEs = Objective Binge Episode frequency.

four groups on the frequency of fasting and excessive exercise as compensatory strategies (as assessed by Item 17 and 18 on the EDE-Q4) were significant, F(3, 90) = 4.94, p b .05 and F(3, 90) = 4.70, p b .05, respectively. Post-hoc Student-Neuman–Keuls tests indicated that the BN, PD, and Restrained Eater groups reported significantly greater use of fasting and excessive exercise than the Control group; no other group differences were significant. 3.3. Eating pathology, body image disturbance, and psychosocial variables Three separate multivariate analyses of variance (MANOVA) compared the four groups on conceptually related dimensions (i.e., eating psychopathology, body image disturbance, and related psychosocial variables). The first MANOVA compared the groups on four measures of body image disturbance: EDE-Q4 shape concern, EDE-Q4 weight concerns, BSQ body dissatisfaction, and SATAQ-3 general internalization of sociocultural ideals about appearance (See Table 2). There was a significant multivariate effect, F(3, 90) = 9.76, p b .001. Univariate ANOVAs yielded significant group differences on each of these measures (EDE-Q4 shape concern: F(3,90) = 43.77, p b .001; EDE-Q4 weight concern: F(3,90) = 41.03, p b .001; BSQ body dissatisfaction: F(3,90) = 39.05, p b .001; SATAQ general internalization: F(3,90) = 22.63, p b .001). Post-hoc tests indicated that the BN group reported significantly greater levels of shape concern, weight concern, and body dissatisfaction than all other groups, and both the PD and Restrained Eaters groups reported significantly greater levels of these variables compared to the Control group. For general internalization, the BN, PD, and Restrained Eaters group all reported significantly greater levels of internalization than the Control group. The second MANOVA compared the groups on eight psychosocial variables: BES impulse strength, BIS general impulsivity, VOCI obsessive compulsive symptoms, three aspects of perfection (MPQ concern over

1,2,3,4

SNK post-hoc results.

mistakes, MPQ personal standards, and MPQ doubts about actions subscales), RSE self-esteem, and DERS emotion regulation (See Table 3). There was a significant multivariate effect, F(3, 90) = 2.09, p b .01. Univariate ANOVAs yielded significant group differences on five psychosocial variables (VOCI obsessive compulsive symptoms: F(3,90) = 4.66, p b .01; MPQ concern over mistakes: F(3,90) = 6.68, p b .001; MPQ doubts about actions: F(3,90)= 5.13, p b .05; RSE self-esteem: F(3,90) = 11.45, p b .001; and DERS emotion regulation: F(3,90)= 7.56, p b .001). Post-hoc tests indicated that both the BN and PD groups reported significantly higher levels of concern over mistakes than the Control group. The BN group reported significantly higher levels of obsessive compulsive symptoms and doubts about actions compared to the Control group. The BN group also reported significantly more emotion regulation difficulties and lower self-esteem than all other groups. The third MANOVA compared the groups on five measures of eating pathology: EDE-Q4 eating concerns, EDDS symptom severity, TFEQ dietary restraint, TFEQ disinhibition, and TFEQ perceived hunger (See Table 4). There was a significant multivariate effect, F(3, 90) = 12.12, p b .001. Univariate ANOVAs yielded significant group differences on all five measures (EDE-Q4 eating concerns: F(3,90) = 23.41, p b .001; EDDS symptom severity: F(3,90) = 55.20, p b .001; TFEQ dietary restraint: F(3,90) = 7.54, p b .001; TFEQ disinhibition: F(3,90) = 32.97, p b .001; and TFEQ hunger: F(3,90) = 11.88, p b .001). Post-hoc tests indicated that the BN group reported significantly greater levels of eating concerns, disinhibition, and overall symptom severity than each of the other groups. Additionally, both the PD and Restrained Eaters group reported significantly higher levels of eating concerns, disinhibition, and symptom severity than the Control group. Both the BN and Restrained Eaters groups reported significantly higher levels of perceived hunger than the PD and Control groups. Finally, the BN, PD, and Restrained Eaters groups reported significantly higher levels of dietary restraint than the Control group.

Table 2 Body image variables: Descriptive data and statistical findings. Dependent variable

Shape concern Weight concern Body dissatisfaction Internalization

Bulimia Nervosa1

Purging Disorder2

Restrained eater3

Control4

n = 35

n = 20

n= 18

n = 21

M (SD)

M (SD)

M (SD)

M (SD)

F

p

ηp2

4.49(1.09) 4.17(1.08) 137.60(28.57) 36.49(6.63)

3.19(1.63) 2.86(1.58) 105.80(41.12) 33.15(9.67)

3.26(1.35) 2.73(1.21) 101.45(26.44) 32.44(7.92)

.72(.55) .62(.66) 51.68(11.92) 18.48(8.94)

43.77 41.03 39.05 2.63

b.001 b.001 b.001 b.001

.59 .58 .57 .43

Between subjects effects

SNK post-hoc results 1 > 2,3 >4 1 > 2,3 >4 1 > 2,3 >4 1,2,3 > 4

Cohen's d effect size

1vs2

1vs3

1vs4

2vs3

2vs4

3vs4

.94 .97 .90 .40

1.00 1.26 1.31 .55

4.37 3.97 3.93 2.29

.05 .09 .13 .08

2.03 1.85 1.16 1.58

2.46 2.16 2.43 1.65

Note. SNK = Student Newman Keuls; Shape concern = EDE-Q4 Shape Concerns subscale score; Weight concern = EDE-Q4 Weight Concerns subscale score; Body dissatisfaction = BSQ score; Internalization = SATAQ-3 General Internalization subscale score.

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

31

Table 3 Psychosocial variables: Descriptive data and statistical findings. Dependent variable

Impulse strength Impulsivity Obsessive compulsiveness Concern over mistakes Personal standards Doubts about actions Self-esteem Emotion regulation difficulties

Bulimia Nervosa1

Purging Disorder2

Restrained eater3

Control4

n = 35

n= 20

n= 18

n = 21

M (SD)

M (SD)

M (SD)

M (SD)

2.39(.82) 65.96(10.50) 56.00(39.84)

2.93(1.57) 64.70(10.01) 37.35(25.44)

2.60(1.46) 66.27(10.41) 35.77(19.91)

2.36(1.09) 61.25(12.24) 26.33(24.09)

28.11(7.87) 25.14(6.14) 12.66(3.34) 27.03(4.97) 100.73(25.92)

26.60(6.62) 25.30(3.92) 11.20(3.31) 31.40(4.37) 80.65(21.22)

23.11(5.96) 25.00(5.20) 10.83(2.83) 31.50(4.96) 84.24(27.11)

19.67(7.68) 6.68 b.001 22.83(5.16) 1.04 .38 9.03(3.89) 5.13 b.01 34.82(5.48) 11.45 b.001 70.90(19.51) 7.56 b.001

Cohen's d effect size SNK post-hoc results

Between subjects effects

F

p 2.31 .08 1.00 .40 4.66 b.01

ηp2

1vs2

.07 n.s. .03 n.s. .13 1 > 4

−.43 −.18 .12 −.03 .56 .64

.18 .03 .15 .28 .20

.21 .72 1.09 .55 .97 .50 −.03 .02 .41 .07 .54 .42 .44 .59 1.00 .12 .60 .53 −.93 −.90 −1.49 −.02 −.69 −.64 .85 .62 1.30 −.15 .48 .56

1,2 > 4 n.s. 1>4 1 b2,3,4 1 > 2,3,4

1vs3

1vs4

2vs3

2vs4

3vs4

.03 .22 .41 −.15 .90 .07

.42 .31 .44

.19 .44 .43

Note. SNK = Student Newman Keuls; Impulse strength = BEQ Impulse Strength subscale score; Impulsivity = BIS score; Obsessive Compulsiveness = VOCI score; Concern over mistakes = MPQ Concern over Mistakes subscale score; Personal standards = MPQ Personal Standards subscale score; Doubts about actions = MPQ Doubts about Actions subscale score; Self-esteem = RSE score; Emotion regulation = DERS score. 1,2,3,4 SNK post-hoc results.

4. Discussion and conclusion These findings have implications for the conceptualization of PD, which is informed by Restraint Theory. First, results indicate that PD represents a clinically significant syndrome. Consistent with previous research (Keel and Striegel-Moore, 2009; Keel et al., 2005, 2008), individuals with PD consistently reported higher levels of body image disturbance, and eating pathology compared to controls; additionally, many of the effect sizes for comparisons between PD and control groups on psychosocial variables were medium to large. Second, although the statistical analyses employed do not allow conclusions regarding the latent structure of bulimic syndromes, the findings with respect to the eating-related variables suggest that PD may represent a midpoint on a continuum of bulimic symptomatology, with dietary restraint representing a less severe syndrome and BN representing the most severe syndrome. This continuum was illustrated by the fact that the restrained eaters reported significantly more dietary restraint, greater fear of losing control, and greater guilt over eating than controls; the PD group reported significantly more dietary restraint, greater fear of losing control, greater guilt over eating, and more frequent compensatory behaviors than controls; and the BN group reported significantly more dietary restraint, greater fear of losing control, greater guilt over eating, and more frequent compensatory behaviors and OBEs than controls. Consistent with previous research documenting a constellation of symptomatology associated with BN (e.g., Fairburn, Cooper, and Shafran, 2003), the BN group also reported the most severe levels of psychopathology in several other domains, including body dissatisfaction, low self-esteem, emotion regulation difficulties, doubts over actions (a maladaptive dimension of perfectionism), and obsessive compulsive symptoms. These findings replicate those of Lowe et al. (1996), who found that restrained eaters reported higher levels of psychopathology and weight concerns compared to unrestrained eaters, yet individuals

with BN reported the highest levels of psychopathology, weight concerns, and binge eating. Interestingly, with the exception of the moderate to large effect sizes for concerns over mistakes, eating concerns, and perceived hunger, the small effect sizes between individuals with PD and restrained eaters suggest more similarities than differences. To our knowledge, no study has yet compared PD to a group of restrained eaters. It appears that PD is both similar to and different from the BN and Restrained Eater groups in the topography of behavior and associated symptomatology. Though the presence of purging behavior distinguished individuals with PD from restrained eaters, the frequency of purging behavior reported by those with PD did not differ significantly from those with BN. Interestingly, the PD and BN groups also reported significantly higher levels of perfectionism related to concern over mistakes, which has been conceptualized as a maladaptive form of perfectionism (Bardone-Cone et al., 2009; Chang et al., 2004). The maladaptive perfectionism reported by the BN and PD groups may influence these individuals to engage in more extreme methods (i.e., purging behavior) of attaining the thin-ideal. In sum, maladaptive perfectionism may be an important psychosocial factor that accounts for why PD and BN represent more severe syndromes along a continuum of eating pathology. Interestingly, although individuals with BN and PD as well as restrained eaters reported subjective losses of control and guilt over their eating behavior when compared to controls, individuals with PD do not experience objective binge episodes, unlike individuals with BN and restrained eaters. These findings are intriguing in light of the fact that despite the significant levels of dietary restraint reported by all three groups, the PD group reported lower levels of perceived hunger than the BN and Restrained Eater groups. One possibility is that individuals with PD are more effective at suppressing their level of hunger than restrained eaters or those with BN. Thus, although they appear to have concerns about their eating and report experiencing some loss of control after eating even moderate amounts

Table 4 Eating pathology variables: Descriptive data and statistical findings. Dependent variable

Eating concerns EDDS composite Restraint Disinhibition Hunger

Bulimia Nervosa1

Purging Disorder2

Restrained eater3

Control4

n = 35

n= 20

n = 18

n = 21

M (SD)

M (SD)

M (SD)

M (SD)

F

p

ηp2

2.51(1.37) 39.33(9.46) 12.69(4.79) 11.29(2.68) 9.15(3.10)

1.54(1.36) 24.96(10.45) 13.17(4.87) 6.20(3.20) 4.99(3.18)

.92(.76) 23.02(11.92) 11.70(5.48) 6.99(2.76) 7.50(3.05)

.06(.15) 6.24(9.46) 7.25(3.04) 4.39(2.24) 5.22(2.40)

23.41 55.20 7.54 32.97 11.88

b.001 b.001 b.001 b.001 b.001

.44 .65 .20 .52 .28

Between subjects effects

SNK post-hoc results 1 > 2,3 >4 1 > 2,3,>4 1,2,3 > 4 1 > 2,3 >4 1,3 >2,4

Cohen's d effect size

1vs2

1vs3

1vs4

2vs3

2vs4

3vs4

.71 1.44 −.10 1.7 1.32

1.44 1.52 .19 1.58 .54

2.51 3.50 1.36 2.79 1.42

.56 .17 .28 −.26 −.81

1.53 1.88 1.46 .66 −.08

1.57 1.56 1.00 1.03 .83

Note. SNK = Student Newman Keuls; Eating concerns = EDE-Q4 Eating Concerns subscale score; Restraint = TFEQ Restraint subscale score; Disinhibition TFEQ Disinhibition subscale score; Hunger = TFEQ Perceived Hunger subscale score.

32

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

of food, the restraint that they report does not lead to objective binge episodes. While most of the present findings were consistent with previous research, there were no group differences on levels of impulsivity or impulse strength, which contradicts evidence that suggests that impulse control problems are associated with eating pathology (Fischer, Smith, and Cyders, 2008; Hawkins and Clement, 1984). However, the present findings are consistent with the meta-analytic review by Stice (2002), which found that while impulsivity did not directly predict increases in bulimic symptoms across studies. It would be useful for future studies to further assess levels of impulsivity among individuals with eating psychopathology, perhaps using more ecologically valid methodologies (e.g., Ecological Momentary Assessments). Additionally, it is important to note that the mean frequency of purging for BN participants was less than two episodes per week, which is inconsistent with the EDDS diagnostic criteria used to select such participants. This may be due in part to the time frame of the assessments used in the study. That is, the EDDS was used to diagnose participants at the time of their screening during mass testing sessions. Some time may have passed between the screening and the completion of study measures, including the EDE-Q4 items that assessed binge and purge frequency. Therefore, participants' symptoms may have fluctuated in the interim time period between screening and study participation, resulting in the observed inconsistency regarding purging frequency. 4.1. Limitations Though the aforementioned findings are compelling, the present study is not without limitations. First, the relatively small sample size limits the statistical power and generalizability of the results. As previously mentioned, latent structure analyses were not feasible to examine the continuous nature of bulimic syndromes. However, many of the present findings are consistent with previous research (e.g., Bardone-Cone et al., 2009; Keel and Striegel-Moore, 2009; Keel et al., 2005; Legenbauer, Vocks, and Ruddel, 2008; Polivy and Herman, 1985; Stice, 2002). Nevertheless, it is notable that the response rate for BN and PD participants was low (e.g., less than half of the invited PD individuals participated). This selection bias may have influenced the results, as it is not clear whether the BN and PD individuals who did respond were similar in symptomatology to those who participated. Second, these results are based solely on self-report measures, which pose several potential limitations. For instance, there is a possibility that some participants intentionally minimized symptoms due to social desirability factors. Furthermore, the possibility of reporting biases was magnified due to the fact that study utilized a non-clinical sample. That is, participants were not seeking treatment for eating disturbances and thus may have been motivated to underreport symptoms. If this was the case, the present study was a conservative test of group differences. Conversely, there is also evidence that the EDDS may over-diagnose eating psychopathology (Stice et al., 2004), which was suggested by the higher prevalence of EDDS diagnoses in the present study compared to existing prevalence estimates of BN and PD (e.g., American Psychiatric Association, 2000). Thus, some participants may have been inaccurately assigned to symptomatic groups, which may have resulted in minimizing group differences on later assessments. However, despite the relatively higher prevalence rates of BN and PD in the present study, these groups' mean scores on measures of eating psychopathology were comparable to scores of the BN group reported by Stice et al. (2000) and fell within 1 SD of the scores reported by Keel et al. (2005). Third, there is also a limitation inherent within the existing diagnostic criteria for the BN and PD groups regarding the frequency of eating disorder behavior (e.g., twice weekly compensatory behavior in BN compared to once weekly purging in PD). Therefore, the greater level of severity in multiple domains evidenced by the BN group may be, in part, due to the criteria used to define these groups. Fourth, the items from the EDE-Q that were used to assess the frequencies of SBEs

are not clear-cut proxies for SBEs. However, we believe that these items address subjective feelings of loss of control and guilt over eating; moreover, the pattern of responses for both items were similar, suggesting that the items measure the same construct. Fifth, control participants were not matched to the ED groups on demographic variables (e.g., age, ethnicity), which may have limited the ability to detect larger group differences. Finally, these findings are crosssectional in nature. Thus, no conclusions can be drawn as to whether one form of eating pathology increases risk for a more (or less) severe form of eating pathology. Although previous research has found little diagnostic crossover in PD (Keel et al., 2005), prospective research is needed to examine the natural course of this disorder and whether diagnostic migrations may occur. 4.2. Conclusions Despite such limitations, these findings suggest an interesting conceptualization of PD, which has its foundation in the literature that has examined the role of purging behavior in BN. Studies have found that individuals with BN not only refrain from binge eating if the opportunity to purge is prevented, but they also eat minimal amounts of anxiety-provoking foods (e.g., one bite of one candy bar) and significantly less compared to healthy controls during laboratory test meals (Rosen and Leitenberg, 1982; Rosen, Leitenberg, Gross, and Willmuth, 1985). Rosen and Leitenberg (1982, 1988) note that BN may be analogous to obsessive–compulsive disorder in that individuals have a fear of eating (especially foods perceived as “fattening”) and may consider even normal quantities of many foods to be “unsafe, repulsive, and fattening”; thus vomiting becomes a “magic ritual” that protects against anxiety and feared consequences (i.e., weight gain) in BN. Although individuals may experience temporary pleasure during binge episodes, the act of vomiting may be more reinforcing via the reduction of overall negative affect. Previous studies have evidenced reductions in not only anxiety but also a range of other negative emotions following selfinduced vomiting (e.g., anger, inadequacy, lack of control) (Johnson and Larson, 1982). Furthermore, vomiting “removes inhibitions against binge-eating episodes in the future, thereby maintaining a cycle of bulimic symptomatology” (Rosen and Leitenberg, 1982, p. 166). Although these conclusions are speculative in nature, it seems reasonable that individuals with PD restrict their dietary intake and purge in response to the consumption of normal amounts of food, both of which are due to a fear of eating, gaining weight, and/or appearing “fat.” As stated by Rosen and Leitenberg (1982, p. 167), “in the usual progression of (BN), once self-induced vomiting is learned, binge-eating typically becomes more severe and more frequent.” It seems possible that individuals with PD, who likely already experience SBEs, may eventually engage in OBEs and cross the diagnostic threshold to BN. Additional prospective studies are needed to examine the possible progression of restrained eating behavior to PD and from PD to BN. Role of funding source Funding for this study was provided by Kent State University. Kent State University had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. Contributors Kathryn Smith and Janis Crowther designed the study. Kathryn Smith conducted literature searches, collected data, conducted the statistical analysis, and drafted the manuscript. Janis Crowther assisted with statistical analyses and manuscript revisions. Both authors contributed to and have approved the final manuscript. Conflict of interest Both authors declare that they have no conflicts of interest.

References Allison, D. B., Kalinsk, L. B., & Gorman, B. S. (1992). A comparison of the psychometric properties of three measures of dietary restraint. Psychological Assessment, 4(3), 391.

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34 American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders, 4th ed. (Text Revision (DSM-IV-TR) ed.). Washington, DC: Author. Bardone-Cone, A., Weishuhn, A. S., & Boyd, C. A. (2009). Perfectionism and bulimic symptoms in African American college women: Dimensions of perfectionism and their interactions with perceived weight status. Journal of Counseling Psychology, 56(2), 266–275. http://dx.doi.org/10.1037/a0015003. Binford, R. B., & le Grange, D. (2005). Adolescents with bulimia nervosa and eating disorder not otherwise specified-purging only. International Journal of Eating Disorders, 38(2), 157–161. http://dx.doi.org/10.1002/eat.20167. Chang, E. C., Watkins, A., & Banks, K. H. (2004). How adaptive and maladaptive perfectionism relate to positive and negative psychological functioning: Testing a stress-mediation model in black and white female college students. Journal of Counseling Psychology, 51(1), 93–102. http://dx.doi.org/10.1037/0022-0167.51.1.93. Cooper, P. J., Taylor, M. J., Cooper, Z., & Fairburn, C. G. (1987). The development and validation of the body shape questionnaire. International Journal of Eating Disorders, 6(4), 485–494. http://dx.doi.org/10.1002/1098-108X(198707)6:4b485:: AID-EAT2260060405>3.0.CO;2-O. Crow, S. (2007). Eating disorder not otherwise specified: Next steps. International Journal of Eating Disorders, 40, S104–S106. http://dx.doi.org/10.1002/eat.20449. Crowther, J. H., Armey, M., Luce, K. H., Dalton, G. R., & Leahey, T. (2008). The point prevalence of bulimic disorders from 1990 to 2004. International Journal of Eating Disorders, 41(6), 491–497. http://dx.doi.org/10.1002/eat.20537. Engelberg, M. J., Gauvin, L., & Steiger, H. (2005). A naturalistic evaluation of the relation between dietary restraint, the urge to binge, and actual binge eating: A clarification. International Journal of Eating Disorders, 38(4), 355–360. http://dx.doi.org/10.1002/ eat.20186. Evans, C., & Dolan, B. (1993). Body shape questionnaire: Derivation of shortened “alternative forms”. International Journal of Eating Disorders, 13(3), 315–321. http://dx.doi.org/10.1002/1098-108X(199304)13:3b315::AID-EAT2260130310> 3.0.CO;2–3. Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16(4), 363–370. Fairburn, C. G., & Cooper, Z. (2007). Thinking afresh about the classification of eating disorders. International Journal of Eating Disorders, 40, S107–S110. http://dx.doi.org/ 10.1002/eat.20460. Fairburn, C. G., Cooper, Z., Bohn, K., O'Connor, M. E., Doll, H. A., & Palmer, R. L. (2007). The severity and status of eating disorder NOS: Implications for DSM-V. Behaviour Research and Therapy, 45(8), 1705–1715. http://dx.doi.org/10.1016/j.brat.2007.01.010. Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment. Behaviour Research and Therapy, 41(5), 509–528. http://dx.doi.org/10.1016/S0005-7967(02)00088-8. Fairburn, C. G., & Garner, D. M. (1986). The diagnosis of bulimia nervosa. International Journal of Eating Disorders, 5(3), 403–419. http://dx.doi.org/10.1002/1098-108X (198603)5:3b403::AID-EAT2260050302>3.0.CO;2-Z. Favaro, A., Ferrara, S., & Santonastaso, P. (2003). The spectrum of eating disorders in young women: A prevalence study in a general population sample. Psychosomatic Medicine, 65(4), 701–708. http://dx.doi.org/10.1097/01.PSY.0000073871.67679.D8. Fink, E. L., Smith, A. R., Gordon, K. H., Holm-Denoma, J., & Joiner, T. E. (2009). Psychological correlates of purging disorder as compared with other eating disorders: An exploratory investigation. International Journal of Eating Disorders, 42(1), 31–39. http://dx.doi.org/10.1002/eat.20556. Fischer, S., Smith, G. T., & Cyders, M. A. (2008). Another look at impulsivity: A meta-analytic review comparing specific dispositions to rash action in their relationship to bulimic symptoms. Clinical Psychology Review, 28(8), 1413–1425. http://dx.doi.org/10.1016/j.cpr.2008.09.001. French, S. A., Jeffery, R. W., & Wing, R. R. (1994). Food intake and physical activity: A comparison of three measures of dieting. Addictive Behaviors, 19(4), 401–409. http://dx.doi.org/10.1016/0306-4603(94)90063-9. Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14(5), 449–468. http://dx.doi.org/10.1007/BF01172967. Grange, D. l, Binford, R. B., Peterson, C. B., Crow, S. J., Crosby, R. D., Klein, M. H., et al. (2006). DSM-IV threshold versus subthreshold bulimia nervosa. International Journal of Eating Disorders, 39(6), 462–467. http://dx.doi.org/10.1002/eat.20304. Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 41–54. http://dx.doi.org/10.1023/B:JOBA.0000007455.08539.94. Gross, J. J., & John, O. P. (1995). Facets of emotional expressivity: Three self-report factors and their correlates. Personality and Individual Differences, 19(4), 555–568. http://dx.doi.org/10.1016/0191-8869(95)00055-B. Haedt, A. A., & Keel, P. K. (2010). Comparing definitions of purging disorder on point prevalence and associations with external validators. International Journal of Eating Disorders, 43(5), 433–439. Hawkins, R., & Clement, P. F. (1984). Binge eating: Measurement problems and a conceptual model. In R. Hawkins, W. J. Fremouw, & P. F. Clement (Eds.), The binge purge syndrome: Diagnosis, treatment, and research. New York: Springer. Herman, C. P., & Polivy, J. (1975). Anxiety, restraint, and eating behavior. Journal of Abnormal Psychology, 84(6), 666–672. http://dx.doi.org/10.1037/0021-843X.84.6.666. Herman, C., & Polivy, J. (1980). Restrained eating. In A. Stunkard (Ed.), Obesity (pp. 208). WB Saunders: Philadelphia. Herman, C. P., & Polivy, J. (1983). A boundary model for the regulation of eating. Psychiatric Annals, 13(12), 918–927. Howell, D. C. (2002). Statistical methods for psychology (5th Ed.). Pacific Grove, CA: Duxbury. Jarry, J. L., Polivy, J., Herman, C. P., Arrowood, A. J., & Pliner, P. (2006). Restrained and unrestrained eaters' attributions of success and failure to body weight and

33

perception of social consensus: The special case of romantic success. Journal of Social and Clinical Psychology, 25(8), 885–905. http://dx.doi.org/10.1521/jscp. 2006.25.8.885. Johnson, C., & Larson, R. (1982). Bulimia: An analysis of moods and behavior. Psychosomatic Medicine, 44(4), 341–351. Keel, P. K., Haedt, A., & Edler, C. (2005). Purging disorder: An ominous variant of bulimia nervosa? International Journal of Eating Disorders, 38(3), 191–199. http://dx.doi.org/ 10.1002/eat.20179. Keel, P. K., Holm-Denoma, J. M., & Crosby, R. D. (2011). Clinical significance and distinctiveness of Purging Disorder and Binge Eating Disorder. International Journal of Eating Disorders, 44(4), 311–316. Keel, P. K., Mayer, S. A., & Harnden-Fischer, J. (2001). Importance of size in defining binge eating episodes in bulimia nervosa. International Journal of Eating Disorders, 29(3), 294–301. http://dx.doi.org/10.1002/eat.1021. Keel, P. K., & Striegel-Moore, R. (2009). The validity and clinical utility of purging disorder. International Journal of Eating Disorders, 42(8), 706–719. http://dx.doi.org/10.1002/ eat.20718. Keel, P. K., Wolfe, B. E., Gravener, J. A., & Jimerson, D. C. (2008). Co-morbidity and disorder-related distress and impairment in purging disorder. Psychological Medicine: A Journal of Research in Psychiatry and the Allied Sciences, 38(10), 1435–1442. http://dx.doi.org/10.1017/S0033291707001390. Kerzhnerman, I., & Lowe, M. R. (2002). Correlates of subjective and objective binge eating in binge-purge syndromes. International Journal of Eating Disorders, 31(2), 220–228. http://dx.doi.org/10.1002/eat.10026. Laessle, R. G., Tuschl, R. J., Kotthaus, B. C., & Prike, K. M. (1989). A comparison of the validity of three scales for the assessment of dietary restraint. Journal of Abnormal Psychology, 98(4), 504–507. http://dx.doi.org/10.1037/0021-843X.98.4.504. Legenbauer, T., Vocks, S., & Ruddel, H. (2008). Emotion recognition, emotional awareness and cognitive bias in individuals with bulimia nervosa. Journal of Clinical Psychology, 64(6), 687–702. http://dx.doi.org/10.1002/jclp. 20483. Lowe, M. R., Gleaves, D. H., DiSimone-Weiss, R., Furgueson, C., Gayda, C. A., Kolsky, P. A., et al. (1996). Restraint, dieting, and the continuum model of bulimia nervosa. Journal of Abnormal Psychology, 105(4), 508–517. http://dx.doi.org/10.1037/0021-843X.105.4.508. Machado, P. P. P., Machado, B. C., Goncalves, S., & Hoek, H. W. (2007). The prevalence of eating disorders not otherwise specified. International Journal of Eating Disorders, 40(3), 212–217. http://dx.doi.org/10.1002/eat.20358. Mond, J., Hay, P., Rodgers, B., Owen, C., Crosby, R., & Mitchell, J. (2006). Use of extreme weight control behaviors with and without binge eating in a community sample: Implications for the classification of bulimic-type eating disorders. International Journal of Eating Disorders, 39(4), 294–302. http://dx.doi.org/10.1002/eat.20265. Niego, S. H., Pratt, E. M., & Agras, W. S. (1997). Subjective or objective binge: Is the distinction valid? International Journal of Eating Disorders, 22(3), 291–298. http://dx.doi.org/ 10.1002/(SICI)1098-108X(199711)22:3b291::AID-EAT8>3.0.CO;2-I. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt impulsiveness scale. Journal of Clinical Psychology, 51(6), 768–774. http://dx.doi.org/ 10.1002/1097-4679(199511)51:6b768::AID-JCLP2270510607>3.0.CO;2-1. Polivy, J., & Herman, C. P. (1985). Dieting and binging: A causal analysis. American Psychologist, 40(2), 193–201. http://dx.doi.org/10.1037/0003-066X.40.2.193. Pratt, E. M., Niego, S. H., & Agras, W. S. (1998). Does the size of a binge matter? International Journal of Eating Disorders, 24(3), 307–312. http://dx.doi.org/10.1002/(SICI) 1098-108X(199811)24:3b307::AID-EAT8>3.0.CO;2-Q. Rachman, S., Thordarson, D. S., & Randomsky, A. S. (1995). A revision of the Maudsley Obsessional Compulsive Inventory. World Congress of Cognitive and Behavioral Therapies. Rosen, J. C., & Leitenberg, H. (1982). Bulimia nervosa: Treatment with exposure and response prevention. Behavior Therapy, 13(1), 117–124. http://dx.doi.org/10.1016/ S0005-7894(82)80055-5. Rosen, J. C., & Leitenberg, H. (1988). Eating behavior in bulimia nervosa. In B. Walsh (Ed.), Eating behavior in eating disorders (pp. 161). Washington, DC: American Psychiatric Association. Rosen, J. C., Leitenberg, H., Gross, J., & Willmuth, M. (1985). Standardized test meals in the assessment of bulimia nervosa. Advances in Behaviour Research and Therapy, 7(3), 181–197. http://dx.doi.org/10.1016/0146-6402(85)90008-6. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Sinclair, S. J., Blais, M. A., Gansler, D. A., Sandberg, E., Bistis, K., & LoCicero, A. (2010). Psychometric properties of the Rosenberg self-esteem scale: Overall and across demographic groups living within the United States. Evaluation & the Health Professions, 33(1), 56–80. http://dx.doi.org/10.1177/0163278709356187. Stice, E. (2001). A prospective test of the dual-pathway model of bulimic pathology: Mediating effects of dieting and negative affect. Journal of Abnormal Psychology, 110(1), 124–135. http://dx.doi.org/10.1037/0021-843X.110.1.124. Stice, E. (2002). Risk and maintenance factors for eating pathology: A meta-analytic review. Psychological Bulletin, 128(5), 825–848. http://dx.doi.org/10.1037/00332909.128.5.825. Stice, E., Fisher, M., & Martinez, E. (2004). Eating disorder diagnostic scale: Additional evidence of reliability and validity. Psychological Assessment, 16(1), 60–71. http://dx.doi.org/10.1037/1040-3590.16.1.60. Stice, E., Telch, C. F., & Rizvi, S. L. (2000). Development and validation of the eating disorder diagnostic scale: A brief self-report measure of anorexia, bulimia, and binge-eating disorder. Psychological Assessment, 12(2), 123–131. http://dx.doi.org/10.1037/10403590.12.2.123. Stunkard, A. J., & Messick, S. (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. Journal of Psychosomatic Research, 29(1), 71–83. http://dx.doi.org/10.1016/0022-3999(85)90010-8. Thompson, J. K., van, d. B., Roehrig, M., Guarda, A. S., & Heinberg, L. J. (2004). The sociocultural attitudes towards appearance scale-3 (SATAQ-3): Development and validation.

34

K.E. Smith, J.H. Crowther / Eating Behaviors 14 (2013) 26–34

International Journal of Eating Disorders, 35(3), 293–304. http://dx.doi.org/10.1002/ eat.10257. Thordarson, D. S., Radomsky, A. S., Rachman, S., Shafran, R., Sawchuk, C. N., & Hakstian, A. R. (2004). The Vancouver obsessional compulsive inventory (VOCI). Behaviour Research and Therapy, 42(11), 1289–1314. http://dx.doi.org/10.1016/j.brat.2003.08.007. Wade, T. D., Bergin, J. L., Tiggemann, M., Bulik, C. M., & Fairburn, C. G. (2006). Prevalence and long-term course of lifetime eating disorders in an adult Australian

twin cohort. The Australian and New Zealand Journal of Psychiatry, 40(2), 121–128. http://dx.doi.org/10.1111/j.1440-1614.2006.01758.x. Wilson, G. (2002). The controversy over dieting. In C. G. Fairburn, & K. D. Brownell (Eds.), Eating Disorders and Obesity: A Comprehensive Handbook (2nd Ed.). New York: Guilford Press.