Purchase of this document is not possible at this time because the store is temporarily disabled. Continue searching

Processes of emotion dysregulation in restrictive eating pathology: A mediation model

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Deborah A Burkowski
Abstract:
Symptoms of disordered eating pose a serious problem among females in Westernized countries (Alloy, Riskind & Mano, 2005). While the eating pathology literature is growing, there is much to learn about development and maintenance of these complex disorders (Stice, 2002). Emotion dysregulation and inhibition have recently been identified important in the development of binging and purging behavior (SchupakNeuberg & Nemroff, 1993; Telch, Agras & Linehan, 2001; Telch & Linehan, 2000); however, limited research has assessed the importance of emotion dysregulation in the development of restrictive eating behavior and attitudes. The present study used structural equation modeling to test the proposed model that Emotion Dysregulation mediates the relationship between Emotion Inhibition and Restrictive Eating Pathology for samples of binging/purging and nonbinging/purging female college students. Model fit was adequate and structural coefficients statistically significant. With nonsignificant differences between the partial and full mediation models, a full mediation best represented the data. However, due to alternative interpretations of the model results, it was concluded that more research is needed to better understand the theorized relationships. Possible modifications to future models of emotion regulation and restrictive eating disorder symptoms are discussed.

TABLE OF CONTENTS I. List of Figures vii II. List of Tables viii III. Acknowledgments ix IV. Introduction 1 a. Anorexia Nervosa 2 b. Bulimia Nervosa 4 c. Eating Disorder Not Otherwise Specified 5 d. Prevalence/Incidence Rates 6 i. Disordered eating in college samples 7 V. Associated Symptoms and Personality Features 8 a. Symptomatology 9 i. Summary 13 b. Personality Features 13 i. Perfectionism 14 ii. Features that differentiate ED diagnoses 17 iii. Summary 20 VI. Emotion Regulation and Eating Disorders 20 a. Emotion Regulation and Dysregulation 20 i. Emotions and eating 24 b. Processes of Emotion Dysregulation in Eating Pathology 25 i. Alexithymia 26 ii. Emotion dysregulation 30 iii. Emotion inhibition 33 iv. Summary 37 VII. Current study 37 a. Using SEM to Test Mediation Effects 40 b. Hypotheses 41 VIII. Method 43 a. Participants 43 b. Measures 46 i. General information sheet 46 ii. Eating disorder examination questionnaire 46 iii. Difficulties in emotion regulation scale 48 iv. Silencing the self scale 50 c. Latent and Measured Variables 53 i. Restrictive eating pathology 53 ii. Emotion dysregulation 54 v

iii. Emotion inhibition 56 d. Procedure 58 e. Statistical Analysis 59 i. Model estimation 59 ii. Samples evaluated 61 iii. Model Fit 62 IX. Results 62 a. Confirmatory Factor Analysis 62 b. Structural Model without BMI for Each Subsample 73 c. Structural Model with BMI for Each Subsample 73 X. Discussion 77 a. Model Comparison 78 b. BMI as a Covariate 79 c. Limitations 80 d. Future Research 81 i. Further considerations: Inhibition as dysregulation 84 e. Conclusions 85 XL References 86 XII. Appendix A, General Information Sheet 101 XIII. Appendix B, Eating Disorder Examination Questionnaire 102 XIV. Appendix C, Difficulties in Emotion Regulation Scale 105 XV. Appendix D, Silencing the Self Scale 107 vi ยป

LIST OF FIGURES Figure 1, page 42: The hypothesized model that Emotion Dysregulation mediates the relationship between Emotion Inhibition and Restrictive Eating Pathology Figure 2, page 65: The final model that Emotion Dysregulation mediates Emotion Inhibition and Restrictive Eating Pathology vn

LIST OF TABLES Table 1, page 51: Internal consistency of measured variables Table 2, page 52: Means and standard deviations of observed variables Table 3, page 53: Restrictive eating pathology Table 4, page 55: Emotion dysregulation Table 5, page 56: Emotion inhibition Table 6, page 60: Correlation matrices between restrictive eating pathology, emotion inhibition, emotion dysregulation and BMI Table 7, page 66: Estimated factor loadings for the whole sample (N= 702) Table 8, page 68: Model fit for the full sample and subsamples for the CFA models with and without BMI Table 9, page 69: Factor loadings and correlated residuals for all three subsamples tested without BMI included in the model Table 10, page 70: Factor loadings and correlated residuals for all three subsamples tested with BMI included in the model Table 11, page 71: Model fit statistics for all three subsamples tested without BMI included in the model Table 12, page 72. Model fit statistics for all subsamples tested with BMI included in the model Table 13, page 75: Structural models without BMI Table 14, page 76: Structural models with BMI included vni

ACKNOWLEDGEMENTS For my dissertation committee: Robyn Ridley, Ph.D. (Chair) Hobart Davies, Ph.D. Jonathan Kanter, Ph.D. Gwynne Kohl, Ph.D. Rodney Swain, Ph.D. For assistance in data analysis and interpretation: Daniel Sass, Ph.D. David Osmon, Ph.D. IX

1 Processes of Emotional Dysregulation in Restrictive Eating Pathology: A Mediation Model Based on the Diagnostic and Statistical Manual for Mental Disorders - Forth Edition Revised, there are three diagnosable eating disorders (ED): Anorexia Nervosa (AN), Bulimia Nervosa (BN) and Eating Disorder Not Other-wise Specified (EDNOS). A number of factors are related to and predictive of ED symptomatology including low self- esteem (Halvorsen & Heyerdahl, 2006), anxiety, depression (Herzog, Keller, Sacks, Yeh, & Lavori, 1992; Hudson et al., 2006; Piran, Lerner, Garfinkel, Kennedy, & Brovillette, 1988), and perfectionism (Bastiani, Rao, Weltzin, & Kaye, 1995; Casper et al., 1992; Castro, Gila, Gual, Lahortiga, & Toro, 2004; Sutandar-Pinnock et al., 2003). In a meta- analytic review, Stice (2002) concluded that the predictive power of individual risk and maintenance factors is limited and thus it was suggested that researchers continue to search for other risk and maintenance factors and develop more comprehensive multivariate models. Both clinical and research evidence suggests that emotion regulation and emotion inhibition are important to understanding ED pathology (Forbush & Watson, 2006; Geller, Cockell, Hewitt, Golder, & Flett, 2000; Whiteside et al., 2007). While bulimic behavior has been conceptualized as a means to regulate emotions (Schupak-Neuberg & Nemroff, 1993; Telch, Agras & Linehan, 2001; Telch & Linehan, 2000; Whiteside et al, 2007), the relation of restrictive eating behavior/attitudes to emotion dysregulation and inhibition is unclear. A better understanding of emotion dysregulation at it relates to restricting ED pathology may be particularly important for the treatment of AN.

2 For the present study, a model of restrictive eating pathology was tested using structural equation modeling (SEM) in a non-clinical female college sample. A non clinical college sample was used in order to dimensionally assess the proposed model, thus allowing for a range of eating pathology scores. Also, Longitudinal studies suggest a progression from less to more severe disordered eating (Polivy & Herman, 2002). In order to control for the influence of hinging and purging behavior on model fit, participants were assessed for the presence or absence of binging and/or purging behaviors. The proposed model was then tested for three participant subgroups: 1) participants who reported an absence of binging or purging behavior, 2) participants who reported an absence of binging and/or purging plus participants who reported excessive/compulsive exercise, and 3) participants who reported binging or purging behavior. It was proposed that emotion dysregulation mediates the relationship between emotion inhibition and restrictive eating behavior and attitudes. The current paper presents discussion of diagnostic criteria for ED diagnoses, a brief review of associated symptoms and personality features, followed by a review of emotion regulation and a review of emotion dysregulation in the ED literature. Finally, the method, results and discussion of the current study are presented. Anorexia Nervosa Based on the Diagnostic and Statistical Manual for Mental Disorders (DSM-IV; APA, 2000) AN is an ED defined by the following criteria: 1) a refusal to maintain body weight at or above a minimally normal weight for age and height (85% or less of ideal body weight), 2) an intense fear of gaining weight or becoming fat, 3) disturbance in perception of shape or weight of one's body or denial of the seriousness of one's low

weight, and 4) in postmenarcheal females, amenorrhea. The Body Mass Index (BMI) is frequently used to determine whether a patient's weight is at a healthy range. A BMI of 18 or lower is considered to be in the anorexic range. Two subtypes of AN have been identified: restricting type and binge/purge type. Restricting type AN is specified when the person diagnosed with AN has not engaged in binge-eating or purging behavior (self- induced vomiting, misuse of laxatives, diuretics, or enemas). Binge-eating/purging type is considered present when the person diagnosed with AN has regularly engaged in binge- eating or purging behavior. One identified problem with the DSM-IV diagnostic criteria for AN is criterion of amenorrhea. This has been considered a problem due to difficulty obtaining accurate information of patients' menstrual cycle (Halmi, 2005). In addition, although distorted body image and/or fear of gaining weight are generally considered key features of AN, some individuals with AN report not experiencing these symptoms. Individuals who lack this drive for thinness are referred to as having atypical AN (APA, 2002; Vervaet, van Heeringen & Audenaert, 2004). Vervaet and colleagues found that 27% of their sample had a low score on the drive for thinness scale of the Eating Disorder Inventory (EDI). These clients included both inpatients and outpatients. Some significant differences between the atypical and typical patients diagnosed with AN have been noted. Those who had atypical AN developed the disorder at a later age, had a lower lifetime body mass index and were more likely to be diagnosed with AN restricting type rather than binge/purge type. Overall, these authors concluded that the primary driving force for food restriction in persons with AN may ultimately be a means to reduce high levels of anxiety, with drive for thinness contributing further for those with typical AN.

4 Some research defines AN based on the criteria of the DSM-III (APA, 1980) rather than the current version of the DSM; therefore, the criteria for the DSM-III is specified. According to the DSM-III, AN is based on the following criteria: 1) intense fear of becoming obese that does not diminish as weight loss progresses, 2) disturbance of body image, 3) loss of at least 25% of original body weight, 3) refusal to maintain body weight over a minimal normal weight for age and height, and 4) no known physical illness that could account for weight loss. Bulimia Nervosa BN is an ED defined by the presence of recurrent episodes of binge eating followed by compensatory behavior. Based on the DSM-IV, criteria for this disorder are as follows: 1) Recurrent episodes of binge eating characterized by eating an amount of food that is unquestionably larger than most people would eat in a relatively short period of time, with perceived lack of control over eating. 2) Recurrent inappropriate behaviors to prevent weight gain, such as self-induced vomiting; misuse of laxatives, diuretics, enemas, or other medications; fasting or excessive exercise. 3) The binge eating and inappropriate purging behaviors both occur, on average, at least twice a week for three months. 4) Self-evaluation is influenced by body weight and shape. There are two subtypes of BN, purging type and non purging type. BN purging type is defined by the presence of self-induced vomiting after binging episodes, whereas BN nonpurging type is defined by the absence of vomiting behavior, and instead the individual engages in another form of compensatory behavior such as excessive exercise or fasting. BN is distinguished from AN as individuals with this disorder maintain an average to above average weight for their height (APA, 2000).

5 According to the DSM-III, criteria for BN includes: 1) recurrent episodes of rapid consumption of large amounts of food in a discrete period of time. 2) At least three of the following: a. consumption of high caloric easily ingested food during a binge, b. inconspicuous eating during a binge, c. termination of eating episodes by abdominal pain, sleep, social interruption, or self-induced vomiting, d. repeated attempts to lose weight by severely restrictive diets, self-induced vomiting or use of cathartics or diuretics, e. frequent weight fluctuations greater than 10 pounds due to alternately binges and fasts. 3) Awareness that the eating pattern is abnormal and fear of not being able to stop eating voluntarily. 4) Depressed mood and self-depreciating thoughts of not being able to stop, and 5) bulimic episodes are not due to AN or any other known physical disorder. Despite difference in diagnostic criteria for AN and BN, there is significant overlap with both disorders having the primary symptoms of excessive concern and self- evaluation of their weight and shape (APA, 2000). In fact, many individuals with a diagnosis of AN develop symptoms of BN over time due to the inability to abstain from eating food (Gowers, 2006). EDNOS Individuals who do not meet criteria for AN or BN but do engage in clinically significant ED behavior are diagnosed with EDNOS (APA, 2000). The presentation of EDNOS includes the following possibilities: 1) most diagnostic criteria for AN are met, and despite significant weight loss, the individual's weight has remained within the normal range, 2) the individual purges food after eating only small portions, or repeatedly chews and spits food out, 3) all of the criteria for BN have been met except binging/purging episodes occur less than two times a week or for a shorter duration than

6 3 months, or 4) recurrent episodes of binge eating without inappropriate compensatory behavior. This final example is commonly referred to as Binge ED. Prevalence/Incidence ofEDs Development of EDs generally begins with dieting during the adolescent or college years (APA, 2000; Goldstein, 2005; Hoek & van Hoeken, 2003) with females making up approximately 90-95% of cases (American Psychiatric Association, 2002). The prevalence of EDs range from 3-10% of females aged 15-29 with diagnoses of BN outweighing AN 2 to 1 (Polivy & Herman, 2002). Approximately .5-3.7% of females meet criteria for AN and 1-3% of women meet criteria for BN (APA, 2000; Hoek & van Hoeken, 2003). A recent study from the National Comorbidity Survey Replication, a representative survey of the U.S. household population aged 18 and older, found that .9% of females and .3% of males met lifetime criteria for AN, 1.5% of women and .5% of males met lifetime criteria for BN, and 3.5% of women and 2% of men met lifetime criteria for Binge ED (Hudson, Hiripi, Pope & Kessler, 2006). In a review, Hoek and van Hoeken reported that most studies found much higher prevalence rates for partial syndrome AN and BN. Also, it is difficult to know the exact prevalence and incidence rates, as individuals who would meet criteria are often reluctant to admit their symptoms. While individuals with BN are more likely to eventually seek treatment, those with AN often deny the disorder. Hoek and van Hoeken reported that incidence rates have been based on cases presenting to health care facilities. Based on their review, the incidence of AN is approximately 8/100,000 cases per year and the incidence of BN is approximately 12/100,000 cases per year.

7 Most individuals who seek help for an ED do not meet the full diagnostic criteria for AN or BN and therefore fall into the category of EDNOS (Fairburn et al., 2007; Machado, Machado, Goncalves, and Hoek, 2007; Polivy & Herman, 2002). In outpatient settings, EDNOS cases account for approximately 60% of ED cases (Fairburn & Bohn, 2005; Fairburn et al., 2007). Turner and Bryant-Waugh (2004) assessed a sample of ED clients from an adult community ED service using a structured interview. Based on this assessment 67% met criteria for EDNOS. Of those with an EDNOS diagnosis, 47% missed a diagnosis of AN on the basis of the weight criteria, 35% missed a diagnosis of AN based on the absence of amenorrhea, 50% missed a diagnosis of BN due to not meeting the frequency of binging, and 37% based on the frequency of compensatory behaviors. Disordered eating in college samples. A number of studies have found high rates of ED symptomatology in college samples. A sample of 682 undergraduate females were assessed for symptoms of ED as well as associated variables (Mintz & Betz, 1988). A high prevalence of dieting and binging behaviors was found among the sample. Eight participants were in the anorexic weight range and 31 had a weight in the obese range. Of the remainder of the sample 3% were classified as bulimic, and 61% were found to have an intermediate eating behavior problem (chronic dieting, binging or purging behavior, or subclinical bulimia). Only 33% of the participants were classified as having normal eating habits. Mintz and O'Halloran (2000) assessed a sample of 136 females attending a university. Based on structured clinical interviews, 33 individuals met a DSM-IV diagnosis for a full syndrome ED. One individual met a diagnosis of AN, 9 met a

diagnosis of BN and 23 met a diagnosis of EDNOS. In addition, Aronson, Fredman and Gabriel (1990) assessed 122 female undergraduates aged 18-27 and 40, 30-56 year old women. It was found that 10% of the young group met the cut-off score of the Eating Attitudes Test (EAT) while none of the older group met the cut-off criteria. Binge ED, included under the diagnosis of EDNOS, has been found prevalent among college samples. Specifically, Whiteside and colleagues (2007) found that of 695 college participants 15% (n = 106) met criteria for Binge ED. Twenty percent of female participants reported an average of 1 or more binge-eating episodes per week in the past 3 months. Of participants who binged but did not meet criteria for Binge ED, 75% did not meet the criteria due to the presence of compensatory behaviors, and 44% did not meet criteria because they binged less than the specified number of days. Associated Symptoms and Personality Features Bruch (1962) described individuals with AN syndrome as having a disturbance in perceptual and cognitive interpretation abilities, rigid restriction of food intake as well as disorganized eating patterns and habits. He further observed that individuals with the AN syndrome experience themselves as ineffective in that they do not do anything to please themselves but only act in response to the demands coming from other people. Today we remain perplexed by ED symptomatology and continue to conduct research to more fully understand associated features. Symptomatology All ED diagnoses, including EDNOS are severe in symptomatology (Fairburn et al., 2007). AN has the highest mortality rate of all psychiatric diagnoses and is one of the most difficult to treat (Goldstein, 2005; Key & Lacey, 2002; Fichter, Quadflieg, &

9 Hedlund, 2006). Approximately 10% will die due to starvation, complications of the disorder (acute diseases, and chronic diseases such as renal failure) or suicide (APA, 2000; Finfgeld, 2002). AN and BN can lead to numerous physical problems and complications including osteoporosis, heart disease, and severe dental problems (Goldstein, 2005). In a literature review of the long-term outcome of AN, it was concluded that complications of the disorder include abnormal bone mineral density, reproductive complications, medical emergencies and physical complaints (Finfgeld, 2002). A review of the research, Mehler (2003) indicated that osteoporosis disease is a serious health risk for those with AN, with high prevalence rates of osteoporosis among AN patients. Low bone mineral density makes this population at increased risk for fracturing bones as well as associated morbidity. The severity of low bone mineral density is associated with the duration of AN and once bone density has been lost, it is not normalized. In a review of the cardiovascular effects of AN, Winston and Stafford (2000) reported that AN can damage the heart, structurally and functionally, in multiple ways including the development of bradycardia, hypotension, electrocardiogram abnormalities, ventricular arrhythmias and heart failure. Due to these problems, the authors advise that a cardiovascular examination and an electrocardiogram be performed on patients with AN. In a review of research assessing the prevalence of suicide attempts and self- injurious behaviors among individuals with EDs, it was concluded that 39% of BN inpatients, 23% of BN outpatients, and 16% of AN outpatients reported past suicide attempts. In addition, 25% of BN inpatients, 25% of BN outpatients and 23% of AN outpatients had engaged in self-injurious behavior (Sansone & Levitt, 2002). In a review,

10 Franko and Keel (2006) reported that 3-20% of patients with AN and 25-35% of patients with BN attempt suicide. While successful suicides were not elevated for BN, studies have reported elevated rates of suicide in patients with AN. They report that purging behaviors, depression, substance abuse, and a history of childhood abuse are correlated with suicidal ideation in patients with EDs. The high rates of suicidal ideation are frightening, especially since only approximately one-third of individuals who meet diagnostic criteria for AN receive mental health care (Sansone & Levitt). Individuals with AN and BN experience symptoms of depression, anxiety, poor concentration, as well as feelings of resentment, alienation (Piran, Lerner, Garfinkel, Kennedy, & Brovillette, 1988), low self-esteem and poor life satisfaction (Halvorsen & Heyerdahl, 2006; Polivy & Herman, 2002). Bachar and colleagues (2002) found that individuals with AN and BN, as compared to a normal comparison group, had the tendency to ignore their own needs and serve the needs of others. Individuals with AN and BN were also less attracted to life and more repulsed by life than normal control participants. The authors concluded that individuals with EDs experience such high levels of guilt and that they reject life by caring less for themselves and utilizing their energy to serve the needs of others. Individuals with AN tend to experience high levels of obsessionality and compulsivity (Halmi, 2005; Halmi et al., 2005). Halmi discussed the function of obsessionality among individuals with AN, with this population having the tendency to obsess over concern with body size and shape. For individuals with AN, weight loss is a skill that they have competency in and for which they receive strong positive

11 reinforcement. Therefore, having a thin body is thought to act as a security that compensates for insecurities. According to Halmi (2005), in order to capture the true psychopathology of individuals with AN, it is necessary to measure and define the lack of self-confidence that these individuals face in dealing with problems of everyday living. Halmi suggested that these symptoms represent a complex disturbance that requires further study. Congruent with this, in a qualitative study, patients diagnosed with AN attributed their severe restrictive behavior to providing feelings of security, avoidance of negative emotions, mental strength, self-confidence, sense of identity, care from others, and to the desire for death (Nordbo, Espeset, Gulliksen, Skarderud & Holte, 2006). Comorbid diagnoses are common among individuals with EDs, with 56% of AN respondents and 94.5% of BN respondents meeting criteria for at least one additional DSM-IV Axis I disorder (Hudson et al., 2006). Among a large sample of treatment seeking individuals with AN, high rates of Axis I comorbidity were found. Seventy-three percent of participants with restricting type AN, and 82% of those who met criteria for both AN and BN met criteria for at least one additional Axis I disorder (Herzog, Keller, Sacks, Yeh & Lavori, 1992). Affective disorders were most common, particularly Major Depressive Disorder. In a longitudinal study of a large representative female sample, Lewinsohn, Striegel-Moore, and Seeley (2000) found that more than 70% of adolescent participants who met a full or partial syndrome ED met criteria for an Axis I disorder in young adulthood. In adulthood, full or partial syndrome EDs were associated with comorbidity, treatment seeking, and risk for psychopathology.

12 The severity of anxiety, depression and obsessionality have been shown to decrease after recovery (Pollice, Kaye, Green & Weltzin, 1997). Specifically, Pollice and colleagues found that these symptoms decreased after short-term weight restoration from AN, and among clients who had been recovered for a long period of time (6 months to 10 years), symptoms were even less severe. However, compared to a healthy control group, individuals who were recovered long-term had significantly elevated symptoms of obsessionality. Stice (2002) conducted a meta-analytic review to determine risk and maintenance factors of ED pathology using only prospective and experimentally designed studies. Based on his review, only studies of bulimic and general eating pathology were found. The results of this review showed that perceived pressure to be thin is a causal risk factor for body dissatisfaction, dieting, negative affect and bulimic pathology. In addition the internalization of a thin ideal was found to be a causal risk factor for body dissatisfaction, dieting, negative affect, and bulimic pathology as well as a maintenance factor for bulimic pathology. Body dissatisfaction was found to be a risk factor for dieting, negative affect, and eating pathology and maintenance factor for bulimic pathology. Body dissatisfaction emerged as one of the most consistent and robust risk and maintenance factors for eating pathology. Contradictory to this, in a longitudinal study Leon, Fulkerson, Perry and Farly- Zald (1995) found that over the course of their 3-year study, body dissatisfaction was not a significant predictor of ED development. The strongest predictor of an ED during the 3r year of the study was risk scores for years 1 and 2. Leon and colleagues concluded

13 that the presence of disordered eating behavior overrides individual and group differences in attitudinal factors and personality. Summary. EDs present severe pathology that can be detrimental to one's mental and physical health. While researchers have begun to develop a better understanding ED pathology, there is still much to learn and understand about these complex disorders. The next section of this paper discusses personality features that are commonly found to be associated with ED pathology. These features are thought to be important in understanding the development and maintenance of ED pathology as well as in the treatment of ED symptoms (Key & Lacey, 2002; Signorini, Bellini, Pasanisi, Contaldo & Filippo, 2003). Personality Features The interest in personality of those diagnosed with EDs has spurred much study over the past decade (e.g. Claes, Vandereycken, and Vertommen, 2005; Fassino et al. 2002; Massoubre et al., 2005). EDs are associated with a number of personality features. Features common to all ED subtypes include harm avoidance (Blok set al., 2004; Farvo et al, 2005; Wagner et al., 2006), narcissism (Bloks et al, 2004; Holliday et al., 2006; Steiger et al., 1997), identity problems (Holliday et al., 2006; Livesley, Jang & Thordarson, 2005; Schupak-Neuberg & Nemeroff, 1993) and perfectionism (Bastiani et al., 1995; Sutander et al., 2003). There are also features primarily specific to diagnoses of AN restricting type and BN. Features that are frequently associated with restricting AN include high levels of rigidity (Cassin & van Ranson, 2005; Fassino, Abbate-Daga et al., 2002), persistence (Bloks et al., 2004; Cassin & van Ranson, 2005; Farvo et al., 2005), compulsivity (Holliday et al., 2005; Holliday et al., 2006; Steiger et al., 1997; Westen et

14 al., 2006), and social isolation/avoidance (Casper et al., 1992; Fassino et al., 2002; Holliday et al., 2005). Common features of BN include novelty seeking and impulsivity (e.g. Cassin & von Ranson, 2005; Diaz-Marsa, Carrasco & Saiz, 2000; Fassino et al., 2002). Individuals with AN binge/purge type tend to have characteristics that are common to BN and AN restricting type (Cassin & van Ranson, 2005; Claes et al., 2005; Fassino, Abbate-Daga, et al., 2002; Holliday et al., 2006; Steiger et al., 1997). While impulsivity has generally found to be higher among binge/purge anorexics (e.g. Claes et al., 2005; Steiger et al., 1997), some individuals with restricting AN also rate high levels of impulsivity (Farvo et al., 2005). Because the feature of perfectionism has been considered a specific risk factor for AN and BN (Forbush, Heatherton & Keel, 2007; Stice, 2002), perfectionism will first be reviewed. Second, features that distinguish individuals who engage in binging and/or purging behavior from those who do not will be reviewed. As stated previously these features include impulsivity. Perfectionism. Based on a meta-analytic review, Stice found that perfectionism is a risk factor for bulimic pathology (AN was not included in the review). Forbush and colleagues assessed a large sample (N = 2,482) of college students using the Perfectionism subscale of the EDI as a measure of combined socially prescribed and self- oriented perfectionism. For female students it was found that perfectionism was associated with disordered eating behaviors; however, the correlations were strongest for restricting and purging behavior. Food restriction mediated the relationship between binge eating and scores of perfectionism. For male students perfectionism was only associated with food restriction.

15 Castro, Gila, Gual, Lahortiga and Toro (2004) administered measures of perfectionism to female adolescents in various levels of treatment for AN as well as to a control group. The authors found that compared with a control group, the AN group presented with higher levels of self-oriented perfectionism. In addition, self-oriented perfectionism was highly correlated with scores on attitudes toward eating. These results suggest that individuals with AN tend to subject themselves to high, unrealistic standards. However, the lack of difference between the AN and control group on socially prescribed perfectionism suggests that individuals with AN do not feel that others expect them to live up to these high standards. Bastiani, Rao, Weltzin, and Kaye (1995) also studied characteristics of perfectionism in female patients with AN: Some were assessed when underweight and others were assessed after they gained weight. The authors found that both of the AN groups had elevated scores of perfectionism as compared with a healthy control group. The underweight group had elevated scores on all subscales of perfectionism except for parental expectations, and the weight restored group had elevated scores on all subscales except parental expectations, personal standards and doubts about actions. Studies by Castro et al. and Bastiani et al. showed that individuals with AN do not feel that others subject them to the same high standards as they have for themselves. Casper and colleagues (1992) conducted a study with hospitalized patients diagnosed with AN restricting type or binge/purge type, BN, as well as a group of healthy individuals. Patients with restricting AN reported significantly higher self-control, inhibition of emotionality and conscientiousness compared to patients with BN or the comparison group, whereas patients with BN reported higher levels of impulsivity. Based

Full document contains 129 pages
Abstract: Symptoms of disordered eating pose a serious problem among females in Westernized countries (Alloy, Riskind & Mano, 2005). While the eating pathology literature is growing, there is much to learn about development and maintenance of these complex disorders (Stice, 2002). Emotion dysregulation and inhibition have recently been identified important in the development of binging and purging behavior (SchupakNeuberg & Nemroff, 1993; Telch, Agras & Linehan, 2001; Telch & Linehan, 2000); however, limited research has assessed the importance of emotion dysregulation in the development of restrictive eating behavior and attitudes. The present study used structural equation modeling to test the proposed model that Emotion Dysregulation mediates the relationship between Emotion Inhibition and Restrictive Eating Pathology for samples of binging/purging and nonbinging/purging female college students. Model fit was adequate and structural coefficients statistically significant. With nonsignificant differences between the partial and full mediation models, a full mediation best represented the data. However, due to alternative interpretations of the model results, it was concluded that more research is needed to better understand the theorized relationships. Possible modifications to future models of emotion regulation and restrictive eating disorder symptoms are discussed.