A SUMMATIVE EVALUATION OF TREATMENT FOR BULIMIA NERVOSA AT THE PRINCE GEORGE EATING DISORDER CLINIC By Rebecca Lindsay Associate of Arts, Central Oregon Community College, 1993 B.Sc. Psy., Portland State University, 1995 A PROJECT SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS IN EDUCATIONAL COUNSELLNG. © Rebecca Lindsay, 1999 UNIVERSTIY OF NORTHERN British Columbia Aprill999 All rights reserved . This work may not be reproduced in whole or in part, by photocopy or other means, without permission of the author. LIBRARY ·- --- - - - _ j ! l1l Abstract This project evaluated the effectiveness of the Prince George Eating Disorder Clinic and the multidisciplinary components ofthe clinic (individual counseling, nutritional counseling, medical visits, group counseling and family counseling). Seventeen participants who either met the DSM-IV-R criteria for bulimia nervosa or were diagnosed with an eating disorder not otherwise specified (EDNOS) with binge/purge behavior completed a client questionnaire and the Eating Disorder Inventory-2 . Clients completed the questionnaires after finishing treatment or a majority of their treatment. A clinical staff member who had the most contact with the participant completed a questionnaire about the participant's treatment. Depending on the frequency ofbulimic symptoms after completing treatment, participants' were classified as having good, intermediate, or poor outcomes. Results showed that 35.3% had good outcome, 23 .5% intermediate outcome, and 41.2% had poor outcome after completing mean of 14.9 months of treatment. The poor outcome group had a mean rating of 4.5 and the good outcome group had a mean rating of 4.3. Both the poor and good outcome groups rated the clinic highly on effectiveness, based on a scale of 1 (unsuccessful) and 6 (very successful) . The intermediate outcome group had a mean rating of3 .3. Participants may not judge the success oftheir treatment according to decreases in binge/purge behaviors. The participants rated components of their treatment (individual counseling, nutritional counseling, physical examinations, group counseling, and family counseling) favorably. Recommendations for changes to the clinic include: ensuring follow-up sessions, making resources more available, and enhancing aspects of the program that clients already find helpful. jy Table of Contents Page Abstract l11 List of Tables VI List ofFigures Vll Acknowledgements VIII Chapter I Introduction Chapter II Purpose and Rationale 3 Research Questions 4 Methodology 5 Participants 5 Research Instruments 5 File Review 5 Client Questionnaire 6 Eating Disorder Inventory-2 (EDI-2) 6 Clinic Questionnaire 7 Process Chapter III 7 Analysis 9 General Evaluation 10 Evaluation According to Symptoms 13 Evaluation ofthe Therapist 19 Evaluation of the Nutritionist 20 Evaluation of the Physician 21 Chapter IV Evaluation of Group Counseling 22 Summary 24 Limitations to the Study 26 Concluding Comments 26 References 28 Appendixes Appendix A: Letter from the Clinic 30 Appendix B : File Review 32 Appendix C: Client Questionnaire 34 Appendix D : Eating Disorder Inventory-2 (EDI-2) 45 Appendix E : Clinic Questionnaire 51 Appendix F : Informed Consent 54 Appendix G: Data from the Client Questionnaire 56 vi List ofTables Page 1. DSM-IV Diagnostic Criteria for Bulimia Nervosa 2 2. A Comparison of the Population and the Sample 10 3. A Comparison of Scores on the EDI-2 Sub scales 19 List ofFigures Page 1. A comparison of client and clinic staff evaluation of treatment success. 11 2. A comparison of client and clinic mean score for (question 19A to 19AA of the client questionnaire and question 8A to 8AA ofthe clinic questionnaire). 14 3. Method of purging before and during the past three months. 15 4. The rate of purging before treatment and in the past three months. 15 5. The rate ofbingeing before treatment and in the pas~ three months. 16 \"Ill Acknowledgements I wish to express a sincere thanks to my supervisor Peter MacMillan for his guidance and support throughout the duration of this project. I also wish to express my appreciation to Annette Browne for her support and to Ron Lehr who guided me in the early stages of my project. I would like express thanks to the Prince George Eating Disorder Clinic for providing participants and to the clinical staff there who shared with me their expertise in the area. I greatly appreciated Intersect Youth and Family Services for providing a central location for me to work from. I would also like to thank my friend Anna Djuric for editing my paper no matter how painful it was for her. Finally, I would like to thank my other editor and husband Grant Lindsay for his loving support. CHAPTER ONE INTRODUCTION The Ameri<;:an Psychiatric Association first classified bulimia nervosa in the third edition ofDiagnostic and Statistical Manual ofMental Disorders (DSM III, 1980). The more recent edition DSM-IV (1994) defines two types ofbulimia: (a) "the purging type who engage in self-induced vomiting, misuse of laxatives, diuretics, and enemas," and (b) ' "the nonpurging type who is regularly involved in over-exercising and fasting without regular use of self-induced vomiting, laxatives, diuretics, and enemas" (p. 445). DSM-IV describes five criteria for the classification of Bulimia Nervosa (see Table 1). The criteria include bingeing followed by undoing behavior and self esteem evaluated according to body size. Each year, bulimia nervosa affects millions of people in western society. Garfinkel, Lin, Goering, Spegg, Goldbloom, Kennedy, Kaplan, and Woodside' s (1995) review of the literature estimated that, in western society, 1% to 1. 5% of young women have bulimia. An earlier study estimated approximately 1% of adolescents and young adult women to have bulimia (Fairburn & Beglin, 1990). Authors ofboth studies suggested that this percentage may be low because people with bulimia often refuse participation in such studies (Fairburn & Beglin, 1990; Garfinkel et al. , 1995). Researchers have only conducted a few outcome studies on bulimia nervosa. Outcome studies completed to date have shown varying results depending on each study' s definition of outcome, length of follow-up, and criteria for selecting subjects. Outcome studies are essential in helping us understand the recovery process and what facilitated recovery for the bulimic patient (Hsu, Crisp, & Callender, 1992). 2 Table 1 DSM-IV Diagnostic Criteria for Bulimia Nervosa* 1. Recurrent episodes of binge eating. Binge eating is characterized by the following: (a) eating in a discrete period of time (e.g., within any 2-hour period) an amount offood that is definitely larger than most people would eat during a similar period of time and under similar circumstances (b) a sense of lack of control over eating during the episode (e.g., a feeling that one cannot stop eating or control what or how much one is eating. 2. Recurrent inappropriate compensatory behavior in order 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 compensatory behaviors both occur. on average. at least twice a week for three months. 4. Self-evaluations unduly influenced by body shape and weight. 5. The disturbance does not occur exclusively during episodes of anorexia. * From the American Psychiatric Association. Diagnostic and statistical manual of Mental Disorders (fourth edition). APA Press, Washington DC, 1994 Very few patients abstain from bulimic behavior (binge/purge) following treatment. The American Psychological Association ( 1993) found that, of those patients who completed treatment 30% are still diagnosed with bulimia nervosa. Likewise, Hsu ( 1990) found that only two-thirds of patients who received psychotherapy improved by the one-year follow-up . Most clinicians consider treatment for bulimia successful when the patient reduces bingeing and purging, develops new coping ski!ls, decreases the time she spends thinking about food and body image, and improves self esteem, mood, and social skills. Rorty, Yager, and Rossetto (1993) asked bulimic women what features ofbulimia were the most difficult to change. Their subjects responded that body image and desire to be thin (80%), fear of getting fat (58%), obsessive or negative thoughts about food (55%), bulimic symptoms (48%), and awareness of hunger and satiety cues (23%). 3 A review of the literature revealed that there are numerous definitions of outcome. Johnson-Sabine, Reiss, and Dayson ( 1992) broke their outcome groups into four categories: cured, good outcome, intermediate outcome, and poor outcome. Most studies divided outcome into only two groups - good or poor. Many studies define good outcome as either abstinence from bingeing and purging, or less than monthly bingeing and purging (Johnson-Sabine et a!. 1992; Reiss & Johnson-Sabine 1995). Another study defined good outcome as four or fewer episodes ofbingeing or purging per four-week intervals (Davis, Olmsted, & Rockert, 1992). Purpose and Rationale In March 1993, the eating disorder clinic was established in Prince George, British Columbia (PGEDC). The clinic offers extensive multi-disciplinary treatment for bulimia nervosa. The clinic team includes two therapists, two physicians, and a nutritionist. The counselor tackles personal and family issues that contribute to the client's bulimia. The nutritionist works with the client on maintaining healthy eating and is involved in the re-feeding process when the client receives in-patient treatment. The physician acts at a case manager, coordinates over-all treatment, and does physical examinations with the client. Depending on the case, an external psychologist, a psychiatrist, and/or the client' s own general practitioner may also be involved with treatment. Because a thorough evaluation of the PGEDC has not been completed, the clinic deemed it important for the purpose of this study to evaluate the whole clinic and its various components (individual counseling, nutritional counseling, physical examination, group counseling, and family counseling). This study will allow the clinic staff to direct needed changes in the program and enhance those portions of the program that are ineffective. This project measured the global success rate of the clinic and the success of components of treatment. There are two major questions addressed in this study. The first question evaluates the effectiveness of the PGEDC and the second question hopes to determine discrepancies between the clinic' s evaluation oftreatment and the client's evaluation oftreatment. Research Questions 1. How effective is the Prince George Eating Disorder Clinic in treating bulimia nervosa? A. What proportion of participants have good outcome, defined as binge and or purging less than monthly? B. What proportion of participants have unsuccessful outcome, defined as bingeing or purging monthly or more? C. What do participants feel contributed to their changing of binge/purge ' behavior or lack of change? 2. Are there discrepancies between the clinic's evaluation of treatment and the client' s evaluation oftreatment? If so what are they? A. How does the clinical staffs' perspective compare to the clients' perspective of treatment? 5 CHAPTER TWO METHODOLOGY This study evaluated the separate components (ipdividual counseling, nutritional counseling, physical examination, group counseling, and family counseling) ofthe PGEDC and the clinic as a whole using structured questionnaires. Participants The sample included seventeen females who had the following characteristics : ( 1) completed treatment or the majority of their treatment at the PGEDC between March of 1993 to present day, and (2) met DSM-IV criteria for bulimia, or were diagnosed as having a eating disorder not otherwise specified (EDNOS) with binge/purge behavior. Patients diagnosed as EDNOS with binge/purge subtype engaged in eating disordered behavior, but at a lower frequency and severity than those clients diagnosed as bulimic. Research Instruments The research instruments consisted of a file review sheet, client questionnaire, the Eating Disorder Inventory-2 and the clinic questionnaire. File Review The PGEDC keeps files on all past and current clients. The files ofthe 50 potential subjects were reviewed before questionnaires were completed. The file reviews determined which specific components of treatment the client received or did not receive, and ensured that the client was an appropriate participant for this evaluative study (Appendix B). The typical file contains consents forms, consents to obtain and release information, health questionnaires, progress notes, and a closure report. 6 Client Questionnaire The structured client questionnaire collected demographic information, assess.ed current bulimic symptoms, and provided a means for the participant to evaluate the PGEDC (Appendix C). The demographic questions determined certain characteristics ' such as age, weight, and height ofthe participants. The bulimic symptoms assessed in the client questionnaire were binge eating, purging, thoughts about food, social relationships, and emotional functioning. The evaluative questions appraised the specific components ofthe clinic and the clinic as a whole. The specific components included individual counseling, group counseling, family counseling, nutritional consult/counseling and experiences with the PGEDC medical doctor. In accordance with other studies on bulimia, outcome was defined by frequency of bulimic symptoms. Good outcome included those who hinged and or purged less than monthly. Intermediate outcome was defined as bingeing and/or purging between 1 and 3 times a month. Poor outcome consisted ofbulimic symptoms occurring at a higher frequency. The client ' questionnaire asked the client specifically in (question #25 and #28) to determine her current amount ofbingeing and purging. Participants also completed the EDI-2 when filling out the client questionnaire. Eating Disorder Inventory-2 (EDI-2) The EDI-2 (Gamer, 1984) is a self-report measure that includes eleven subscales on typical symptoms of an eating disorder (Appendix D). This study used the EDI-2 to assess social functioning and self-esteem. Two subscales, interpersonal distrust and social insecurity, measured social functioning and another two subscales, body dissatisfaction and ineffectiveness, measured self-image. The EDI-2 has an internal 7 consistency above .80 (alpha) and subscale reliability coefficient (alphas) between .83 and .93 (Gamer, 1984). Clinic Questionnaire Employees ofthe clinic received a questionnaire for each subject participating in the study (Appendix E). The clinic questionnaire paralleled the client questionnaire in many respects. For example, each questionnaire had a question measuring the effectiveness of treatment for each participant. The clinic questionnaire elicited information on how the staff person worked with the participant and how effective he/she felt treatment was/is for that specific client. The clinic questionnaire was compared to the client questionnaire to determine perceptual discrepancies of treatment. Process Employees ofthe clinic contacted each potential subject with an informative letter which introduced the researcher, explained the purpose of the evaluation, and notified the participant that the researcher would contact her. All questionnaires contained an informed consent letter (Appendix F) that was to be signed and dated by the participant. The researcher phoned and set up times for completion of the two questionnaires. Participants completed the questionnaires at a local social service agency for youth and families (adolescents that received treatment at the PGEDC met with their counselor at this agency). If the potential participant could not make it to the prescribed location to complete the questionnaires the researcher offered to send the questionnaires to her home address. But in most cases, participants completed the client questionnaire and the EDI-2 under the supervision of the researcher. The researcher answered questions and ensured that questionnaires were fully completed. The researcher followed up with participants 8 who were mailed the questionnaire package (informed consent form, client questionnaire, and EDI-2) with a phone call two weeks later. All files ~ reviewed and the file review sheets completed. From these sheets it was determined what type oftherapy was used and who worked with each client. Employees of the clinic then were asked to fill out a questionnaire for each person participating in the study. Information from the questionnaires was transcrjbed and analyzed. Each subject was given a code that replaced her name. The key for the codes was kept in a secure place away from the questionnaires as a further safe guard in preventing identification of participants. Questionnaires were destroyed following the analysis. Microsoft Excel was used for the statistical analysis. 9 CHAPTER THREE ANALYSIS The descriptive analysis suggests that the sample was representative of the population. The participants were 17 females with a mean age of 28.4 (SD=9. 1). Height ranged from 61 to 70 inches (M=64.6, SD=1.9), and weight ranged from 105 pounds to 240 pounds (M=154.3, SD=48 .1). The majority ofthe participants were single (64.7%) with only 23.5% married, 5.9% divorced, and 5.9% separated. It is typical for women with eating disorders to be well educated since they are often overachievers. The sample for this study was well educated; 29.4% had a university or college degree while 35 .3% had a few years of college, university, or technical school. Approximately 23 .5% ofthe sample had only high school education, but these participants were adolescents who were currently attending high school. Of the 50 potential participants, 17 were successfully recruited for this study, 11 others refused to participate, 11 did not return the questionnaire or phone calls, and 1 1 could not be located. Table 2 displays the differences between potential participants and the sample. As shown in Table 2, the sample shows slightly lower frequencies on all variables presented. For example, the sample has a slightly lower percentage ofbulimic clients and is approximately one year younger than the population. The differences between the sample and the population are relatively small; therefore, the sample is considered to be similar to the population on all characteristics presented in Table 2. Results should reflect the entire population. 10 Table 2 A Comparison of the Population and Sample Percentage Mean Bulimic Age Months in Treatment Number of Contacts Months Out of Treatment Population N=50 76.5% 29.5 17.2 18.12 19.4 Population less Sample 78.8% 30 18.5 18.2 21.6 Sample n=17 70.6% 28.4 14.8 14.1 16.7 General Evaluation The majority of participants learned about the clinic from their family physician and/or from a counselor outside of the PGEDC. Participants had also learned of the clinic through other social service agencies, their schools, advertisements, friends, and their families . When initially contacting the clinic, participants expected their bulimia to be cured, they hoped to learn new coping skills, increase their knowledge of bulimia and nutrition, and get help with personal issues through counseling. Approximately 24% (n=4) did not have any expectations when first contacting the clinic. The majority of participants felt the clinic met their expectations (70.5%, n=12) while 29.5% (n=5) felt that the clinic did not meet their expectations. Figure 1 compares responses to a question from the client questionnaire (#I 5) to a question on the clinic questionnaire (#6). A comparison could not be made for three of the participants because either the participant or the clinic staff member did not answer the question. The two questions rate the success of treatment. Question # 15 asked clients to rate the overall effectiveness of the clinic on a scale of I (not successful) to 6 II (very successful). The clinic staff member who had the most contact with a client completed a parallel clinic questionnaire for that client. In question #6 (clinic questionnaire), the clinic staff member rated the effectiveness of treatment for that client on the same scale as the client did in question # 15 (client questionnaire). The graph shows that the clinic staff rated the effectiveness of treatment lower than the client did . There are only two cases where a clinic staff member rated effectiveness of treatment higher than the client. ~ ftl ~ ~ ~~~~ 2 3 4 ~ ~~~~~ 5 6 7 8 9 10 ~ ~ 11 12 ~~ 13 14 Participant Figure 1. A Comparison of client and clinic staff evaluation of treatment success. When clients commented on their rating, the majority reported that the clinic assisted recovery by encouraging them to work through personal issues, by providing group counseling, and by prescribing medication. Another 31 .3% (n=S) felt that they did not receive the treatment needed and were still suffering from bulimic symptoms. These participants mentioned that the clinic did not have enough funding to give needed support and felt that after their treatment they did not receive essential medical follow-up . When clinic staff members commented on their ratings in question #6 (clinic questionnaire) they reported that the majority of participants moved toward recovery by working through issues relevant to their eating disorder. The clinic staff also recognized problems that may have prevented successful treatment for certain participants. For example, clinic 12 staffwere concerned that some participants avoided certain emotions, had difficulty establishing rapport, had stopped follow-up sessions, and were not working through underlying issues. Clinic staff worked with clients to establish appropriate goals during the initial phase oftreatment. Treatment goals are an important part oftherapy. According to question #32 (client questionnaire), 42.9% (n=6) ofthe participants' felt that they reached their treatment goals, 42.9% (n=6) partially met their goals, and 14.3% (n=2) did not meet their goals. Participants that met treatment goals felt that the PGEDC helped them accomplish their goals. Participants remembered setting such goals as "normalizing eating," stopping the binge/purge cycle, regaining self-esteem, learning new coping ' skills, gaining confidence with other people, and getting in touch with emotions. On question #31 some participants (15 .4%, n=2) responded that they could not remember setting treatment goals with the clinician. Clinic staff members remembered setting similar goals with their clients including: eating healthy, stopping the binge/purge cycle, increasing assertiveness, learning self acceptance, and dealing with emotions. Sixteen participants completed question #18 ofthe client questionnaire. Question #18 asks: "Ifyou had it all to do over again would you come to the PGEDC?" To this question 100% (n=16) felt that they would attend the PGEDC ifthey had it all to do over again. Only one participant did not answer the question; this participant rated the clinic 1 (not successful) . Participants cited numerous suggestions for improving the clinic. The majority of participants recommended making treatment more available by having a drop-in center, opening in the evenings, having a full-time clinic, making more frequent counseling 13 ' sessions, and having longer sessions. Other suggestions included having more groups, having follow-up sessions, allowing clients to do more talking than staff, making resources more available, having more aggressive treatment, and having counseling sessions outside in the fresh air. Evaluation According to Symptoms Figure 2 presents the participants' mean scores for questions 19A to 19AA (client questionnaire) and the clinic staff members' corresponding mean scores for questions 8A to 8AA (clinic questionnaire). Participants were asked to rate changes in their behavior following their treatment while the clinic staff was asked how much they emphasized certain topics in treatment. In question 19 A to 19 AA (client questionnaire) the ' participants were asked, "since treatment has there been a change in" followed by list of 27 items (A, B, C .... AA) related to bulimia nervosa. Participants could respond much worse, worse, no change, better, or much better. The responses were weighed from 1 (much worse) to 5 (much better). In corresponding question 8A to 8AA (clinic questionnaire) the clinic staff members were asked "how much did you emphasize the following with this client" followed by the same list of items. The clinic staff member could answer one ofthe following none, very little emphasis, some emphasis, significant emphasis, and very significant emphasis. The responses were weighed from 1 (very little) to 5 (very significant). The mean for each response is presented in Figure 2. The graph shows that the clinic staff consistently scored lower than the client. In general the client and clinic mean response curves are similar indicating that clients see changes in their behavior following treatment even if the clinic staff members feel that they did not highly emphasize certain topics during the client's treatment. The comparison made in 14 Figure 2 suggests that attending therapy and learning new skills may help clients with other issues related to their eating disorder. 5 4.5 4 3.5 Cl .5 7a a: 3 2.5 2 -+-Client ---staff 1.5 0 ' ~ ~ 0 0 0 ~ Question 19 (Client) and 8 (Staff) Figure 2. A comparison of client and clinic mean score for (question 19A to 19AA of the client questionnaire and question 8A to 8AA of the clinic questionnaire). Figure 3 shows the method of purging used before treatment and the method of purging used in the past three months. Each method of purging has dropped since treatment. Using vomiting as a means of losing weight dropped dramatically following treatment from 100% to 29.4%. After treatment, participants used exercise and dieting as preferred means to losing weight. Exercise and dieting are unhealthy when used for purging. Interesting enough, those two methods of purging are socially acceptable ways for people to lose weight. The use of laxatives and diet pills dropped to 11. 8%. Not one participant used water pills and/or enemas in the past three months. 15 100% 90% 80% ---------i • 70%> 60% 50% 40% 30% 20% 10% 0% before treatment em pastthree months Figure 3. Method of purging before treatment and during the past three months. Figure 4 and Figure 5 display the rate of bulimic behavior before treatment and in the past three months. In Figure 4, notice that 100% (n=17) ofthe participants were purging daily or more than daily. Following treatment there are significant changes in the rate of purging for most participants. 100% .---------------------------------80% -t-- - - - - 1 • before treatment mpast three months 60% 40% +---------------------------------20% 0% never less than once a month between 1- about once daily or 3 times a a week more than month daily Figure 4. The rate of purging before treatment and in the past three months. 16 Figure 5 shows that the rate ofbingeing has dropped significantly since treatment. Before treatment, all participants were bingeing either "about once a week" or "dail y or more than daily." Now, 50% have taken huge steps by abstaining from bingeing for 3 months or by bingeing less than monthly in the past three months. 90% .--------------------------------------, 80% +------------------------------70% • Before Treatment 60% 0 Past Three Months f - - - - - - - 50% +------------------------------40% +------------------------------30% ~ ~~ 20% 10% 0% ~~~ ~ ~ ~ never less than between 1once a 3 times a month month about once a week daily or more than daily Figure 5. The rate ofbingeing before treatment and in the past three months. The following study defined good outcome as participants bingeing and/or purging less than monthly, intermediate outcome as bingeing and/or purging between 1 and 3 times a month, and poor outcome consisted of bulimic symptoms occurring at frequency of weekly or more. The intermediate outcome group was important, because it considers participants who have taken huge steps in their recovery even though they have not decreased bulimic symptoms enough to be considered part of the good outcome group. According to this study, 35 .3% (n= 6) had good outcome, 23 .5% (n=4) intermediate outcome, and 41 .2% (n= 7) had poor outcome. According to question #15 (client questionnaire) the good outcome group rated the effectiveness of treatment (M=4.3), the intermediate outcome group rated the effectiveness oftreatment (M=3 .3) 17 and the poor outcome group rated the effectiveness of treatment (M=4.5). Both the poor and good outcome groups rated the clinic highly. The intermediate outcome group may have been low because it only contained 4 participants and one participant had given the clinic a rating ofl (unsuccessful). Ifwe discard the extreme rating of 1, the intermediate outcome group would rate the clinic as a 4.0. Participants in the poor outcome group may have seen progress in treatment as something other than the total elimination of the binge/purge cycle. There also could be a higher ~ of poor outcomes, because over-exercising and/or crash dieting are considered methods of purging when used for weight loss. In question# 25 participants over-exercising and crash dieting at a rate of "about once a week" or "daily or more than daily" would have been considered to have poor outcome. These are socially acceptable ways to lose weight in western society. The participants, in general, felt that the clinic helped them manage their bulimic symptoms. However, a few participants reported ways that the clinic may have increased their bulimic symptoms. These comments included: (1) the client felt that her emotions were getting out of control causing an escalation in the binge/purge cycle; (2) the client was feeling that the binge/purge cycle got worse at the beginning of treatment; (3) the clinic was failing to notice when the client got worse; (4) the client moving from bulimic . behavior to anorexic behavior; (5) the client learning new techniques for purging; (6) the client felt that the counselor told her how she was supposed to feel. In addition to binge/purge symptoms it is important to recognize other features of bulimia such as poor social functioning and low self-esteem. This study used two subscales ofthe EDI-2, Interpersonal Distrust and Social Insecurity, to measure social functioning, and another two subscales, Body Dissatisfaction and Ineffectiveness, to 18 measure self-esteem. In Table 3 the mean raw scores ofthe current sample are compared to the mean scores of a bulimic sample and to the mean scores of a normal female college population. For the current sample the mean raw scores for the subscales were interpersonal distrust (4.0), social insecurity (6.0), body dissatisfaction (16.2), and ineffectiveness (5.9). The mean raw scores for the current sample were below that of the bulimic sample, but were higher than mean raw scores for females without an eating disorder. The body dissatisfaction subscale remains in the eating disorder range; however, in a previous study 80% ofthe sample found body image one ofthe most difficult features ofbulimia to change (Rorty, Yager, & Rossetto, 1993). Without preliminary scores for each participant, it is difficult to estimate changes in social functioning and self esteem; however, the average raw scores for interpersonal distrust, social insecurity, and ineffectiveness show improvements over others who have eating disorders. 19 Table 3 A Comparison of Scores on the EDI-2 Subscales * Social Skills Selfesteem Current Sample After Treatment Bulimic Sample from the EDI-2 Normal College Comparison Group from the EDI-2 Interpersonal Distrust M=4.0, SD=4.0 M=5 .3, SD=4.5 M=2.0, SD=3 .1 Social Insecurity M=6.0, SD=5 .0 M=8.2, SD=4.5 M=3 .3, SD=3.3 Body Dissatisfaction M=16.2, SD=8 .8 M=l7.9, SD=7.9 M=5 .5, SD=5 .5 Ineffectiveness M=5 .9, SD=7.4 M=ll.O, SD=7.5 M=2.3, SD=3 .6 * Information from two columns (Buluruc Sample and Normal College Companson Group) are from the Eating Disorder Inventory-2 : Professional Manual. Psychological Assessment Resources. Odessa. Florida: 1984 Evaluation ofthe Therapist There are two therapists working at the clinic.