Diagnostic Skill Application II

EatingDisorderSymptomSeverityScale-ANewClinicianRatedMeasure.pdf

Home >Psychology homework help >Diagnostic Skill Application II

Eating Disorder Symptom Severity Scale: A New Clinician Rated Measure

KATHERINE A. HENDERSON and ANNICK BUCHHOLZ Regional Eating Disorders Program, Children’s Hospital of Eastern Ontario Research

Institute; and Department of Psychiatry, Carleton University, Ottawa, Ontario, Canada

JULIE PERKINS Regional Eating Disorders Program, Children’s Hospital of Eastern Ontario, Ottawa, Ontario,

Canada

SARAH NORWOOD Department of Psychology, York University, Toronto, Ontario, Canada

NICOLE OBEID Department of Psychology, University of Ottawa, Ottawa, Ontario, Canada

WENDY SPETTIGUE Regional Eating Disorders Program, Children’s Hospital of Eastern Ontario Research

Institute; and Department of Psychiatry, University of Ottawa, Ottawa, Ontario, Canada

STEPHEN FEDER Regional Eating Disorders Program, Children’s Hospital of Eastern Ontario Research

Institute; and Department of Pediatrics, University of Ottawa, Ottawa, Ontario, Canada

This study describes the development and validation of the clinician-rated Eating Disorder Symptom Severity Scale (EDS3), created to address a gap in measurement options for youth with eating disorders. The EDS3 is modeled on the Childhood Severity and Acuity of Psychiatric Illness Scales (Lyons, J. S, 1998). Factor analysis revealed a 5-factor solution and accounted for 78% of the variance, and internal consistency within the subscales was good (Cronbach alphas: 0.69 to 0.93). The EDS3 is a valid and

This study was supported by grants from the Ontario Provincial Centre of Excellence for Child and Youth Mental Health and the W. Garfield Weston Foundation. The authors’ gratitude is extended to all of the experts and clinicians who participated in this study.

Address correspondence to Katherine A. Henderson, Regional Eating Disorders Program, Children’s Hospital of Eastern Ontario Research Institute, 401 Smyth Road, Ottawa, Ontario K1H 8L1, Canada. E-mail: [email protected]

333

334 K. A. Henderson et al.

reliable measure designed for clinicians to help assess the severity of a youth’s eating disorder and to facilitate outcomes research.

While rates of disordered eating in youth are rising (Jones, Bennett, Olmsted, Lawson, & Rodin, 2001; Reijonen, Pratt, Patel, & Greydanus, 2003), there continues to be little information on the effectiveness of treatment for eating disorders (le Grange & Lock, 2005; Wilson, Grilo, & Vitousek, 2007). This gap is especially notable for intensive treatment programs, such as day treatment and inpatient programs. There is also a lack of research on clinical pathway management, even though it is acknowledged that there is a wide range of severity of eating disorders requiring different levels of intervention (Striegel- Moore, 2005; Wilson, Grilo, & Vitousek, 2007).

The dearth of information on treatment efficacy is partly due to inad- equate follow-up measurement of patients (Zipfel, et al., 2002). Current measures are dominantly in a self-report or interview format [e.g., the Eating Attitudes Test (Garner, Olmsted, Bohr, & Garfinkel, 1982), the Eating Disorder Examination (Fairburn & Cooper, 1993), the Structured Interview for Anorexic and Bulimic disorders (Fichter & Quadflieg, 2000)], and are not well-suited to monitoring purposes. Interviews can require extensive training and be lengthy to complete (Carter, Stewart, & Fairburn, 2001). Self- report measures, though less time-consuming and better standardized, can be associated with underreporting due to the denial that is characteristic of eating disorders (Fairburn & Beglin, 1994). Furthermore, many measures have been created for adults and then modified or simply used with youth, without developmental considerations.

When attempting to track treatment progress for research and/or clinical purposes, the measurement tool must work seamlessly within the regu- lar clinical practice and provide data relevant to clinical decision making. Currently, the interview and self-report measures that are available are not conducive to regular use in tracking treatment. The time commitment for patients and clinicians alike usually prohibits the regular use of self-report or interview tools throughout treatment. It is essential that an adequate mea- sure of eating disorder behaviors and symptoms be developed to regularly monitor illness severity and treatment progress and thereby to enhance clin- ical decision making and to provide a sensitive tool to be used in outcome research, especially in youth.

The Eating Disorder Symptom Severity Scale (EDS3) was created to address this gap in measurement options for youth suffering from eating disorders and to facilitate the outcomes evaluation of eating disorder treat- ment programs. The EDS3 is modeled on the Childhood Severity and Acuity of Psychiatric Illness Scales (Lyons, 1998) and the Child and Adolescent Needs and Strengths Measure (Lyons, Griffin, Fazio, & Lyons, 1999) which have been demonstrated to be effective in reforming decision making for residential treatment (Lyons, Libman-Mintzer, Kisiel, & Shallcross, 1998) and in quality improvement in crisis assessment services (Lyons, Kisiel, Dulcan,

Clinician Rated Eating Disorder Symptom Scale 335

Cohen, & Chesler, 1997). The EDS3 is the first and only measure of its kind designed to help achieve similar improvements in the field of eating disorders.

The scale was designed to measure four broad constructs: (1) eating dis- order behaviors, (2) eating disorder cognitions, (3) eating disorder anxiety, and (4) treatment progress. The constructs of eating disorder behaviors and cognitions ensure that the measure captures the standard diagnostic criteria and symptoms of an eating disorder. The eating disorder anxiety construct was included to tap into the anxiety and agitation inherent in patients strug- gling with an eating disorder. For example, many researchers and clinicians have attempted to treat the anxiety and agitation suffered in eating disor- ders with medications such as, serotonin specific reuptake inhibitors (Kaye, et al., 2001) and atypical antipsychotics (Dennis, le Grange, & Bremer, 2006; Dunican & DelDotto, 2007; Lock & Gowers, 2005), despite a lack of evi- dence for the efficacy of these interventions in youth (Norris, Spettigue, Buchholz, Henderson, 2007). Measures that capture the anxiety and agita- tion component of eating disorders are limited, obliging researchers typically to apply measures designed to capture psychotic behavior (Spettigue et al., 2008) or obsessive compulsive behaviors (Kaye et al., 2001). The EDS3 anx- iety subscale designed to assess the anxiety manifested specifically in eating disorders, will respond directly to this gap.

The fourth construct, treatment progress, addresses the motivation and self-efficacy that patients are experiencing with regard to their treatment and recovery process. An eating disorder is an egosyntonic illness, and thus a central issue in working with youth with eating disorders is their ambiva- lence about recovery. The transtheoretical model of change that offers a framework of different stages of readiness for change and different processes that are helpful at each stage (Geller et al., 2008; Gusella, Butler, Nicols, & Bird, 2003; Geller, Cockell, & Drab, 2001; Prochaska & Norcross, 2001) suggests that matching treatment to the youth’s stage of change facilitates engagement in treatment and provides a process for enhancing motivation and decreasing ambivalence. Furthermore, self-efficacy, which is a youth’s perceived ability to succeed in certain tasks, has been shown to affect the phases of change (Pinto, Guarda, Heinberg, & Diclemente, 2006). High lev- els of self-efficacy increase the amount of effort expended to achieve a goal and perseverance to continue in the face of challenges and set backs; while low levels of self-efficacy decrease effort and perseverance. This fourth construct of treatment progress that captures motivation and self-efficacy is included in the EDS3 because it is a critical component of engagement in treatment and recovery from an eating disorder.

The purpose of the current study was to develop and validate the Eating Disorder Symptom Severity Scale (EDS3), a measure designed for regular monitoring of illness severity in youth by a clinician. The goal was to cre- ate a measure that could be reliably completed by health care professionals (e.g. physicians, nurses, and psychologists) with minimal time requirements.

336 K. A. Henderson et al.

The measure would be sensitive to change across time and could be inte- grated seamlessly into the clinical care provided in a general outpatient clinic or by an intensive treatment team. It was developed to enhance decision- making in clinical care pathways and to facilitate outcomes management research for youth. Finally, it would be able to measure the constructs of eating disorder behaviors, eating disorder cognitions, eating disorder anx- iety, and treatment progress. The eating disorder behaviors subscale was designed to capture two separate types of behaviors: (1) behaviors consis- tent with a diagnosis of anorexia nervosa (AN-behaviors) and (2) behaviors consistent with a diagnosis of bulimia nervosa (BN-behaviors). The mea- surement of eating disorder anxiety is a construct hitherto unexplored and the EDS3 will represent a first attempt at capturing data to validate this new construct.

Finally, psychometric properties of the EDS3 will be evaluated. It is expected that the measure will demonstrate good face validity, inter-rater reliability, test-retest reliability, and internal consistency. The theoretically derived subscales will be statistically evaluated using principal component analysis. Convergent validity will be demonstrated with an existing validated measure of eating disorders, the Eating Disorder Inventory (EDI-2) (Garner, 1991) and anxiety, the Multidimensional Anxiety Scale for Children (MASC; March, 1997). The EDI-2 (and the more recent EDI-3) is currently one of the most commonly used self-report measures for eating disorder cognitions and behaviors. It is a 91-item self-report scale with 11 clinical dimensions. The MASC is one of the most common self-report measurement tools for anxiety in children. It is a 39-item self-report questionnaire with 9 subscales. Total symptom severity and scores on the Eating Disorder Cognitions subscale are expected to be positively related to the Body Dissatisfaction and Drive for Thinness subscales of the EDI-2 (Garner, 1991) while the AN-behaviors of the Eating Disorder Behaviors subscale are expected to be positively correlated with the EDI-2 Drive for Thinness subscale, and the BN-behaviors component to be correlated with the Bulimia subscale of the EDI-2. The Eating Disorder Anxiety subscale is expected to be positively correlated with the Multidimensional Anxiety Scale for Children (MASC) Total score and the MASC Social Anxiety subscale score.

METHOD

Scale Development

The development of the individual items and factor definitions was based on a review of the literature on current measures for assessing eating disorder sypmptomatology, a review of the Childhood Severity and Acuity of Psychiatric Illness Scales (Lyons, 1998), focus groups with health care

Clinician Rated Eating Disorder Symptom Scale 337

practitioners specialized in eating disorders, and expert feedback from experts in eating disorders. Dr. John Lyons, the author of the CAPI and CANS, provided expert consultation throughout the development of the EDS3. A thorough review of the literature on available measures for assess- ing eating disorder symptoms was conducted so that potential items for the measure could be identified. Also, three focus groups were held with health care practitioners working in the field of eating disorders in order to generate feedback and to further refine item development. Following this work a draft of an item booklet and a rating sheet were developed. This version of the EDS3 was then circulated to eating disorder experts for their consultation and comment on the content of the measure, the constructs chosen, the wording, the likelihood that they would use such a measure, and other comments that they had. The rating sheet and item booklet were modified according to the feedback from the group of experts who responded (for an example of one item as described in the item booklet and rating scale, please see figure 1).

Item Booklet: Motivation for Treatment

This item explores the extent to which the youth wants to get better. It also looks at feelings of guilt, which are commonly associated with recovery from an eating disorder. If the patient has not yet begun treatment please respond to the following questions by considering their potential motivation for treatment.

0 Consistent motivation. Consistently motivated for change. Wants to get better and fully recover from the eating disorder. Feels happy and free to move towards health and away from the eating disorder (i.e., no guilt whatsoever associated with recovery).

1 Frequent motivation. Often motivated for change. More often than not, wants to get better. Sometimes feels guilty about moving towards health and away from the eating disorder.

2 Occasional motivation. Sometimes motivated for change. Sometimes wants to get better. Feels a lot of guilt about moving towards health and away from the eating disorder.

3 No motivation. No motivation for change. Does not want to get better. Feels comfortable with the eating disorder and does not want to move away from the eating disorder.

Rating Scale 0 1 2 3

Motivation for Treatment

Consistent motivation

Frequent motivation

Occasional motivation

No motivation

FIGURE 1 Example of the item booklet and rating sheet entry for motivation for treatment.

338 K. A. Henderson et al.

Participants Clinician participants. Health care practitioners working on the eating

disorder team at the Children’s Hospital of Eastern Ontario were invited to participate in the training and reliability testing of this measure. Participants were 2 males and 15 females with a range of 1 to 15 years experience working in the field of eating disorders.

Patient participants. Clinicians completed 1 to 24 EDS3 measures on 105 different female patient participants with a mean age of 15.06 years (SD = 1.77). Clinicians completed the measures on the same patient partici- pant across 2 to 16 weeks of treatment. The data was collected on patients who were participating in either a comprehensive assessment for an eat- ing disorder (22.0%), attending an outpatient medical-nutritional clinic for eating disorders (38.4%), receiving intensive treatment in a day program (31.0%), or inpatient program (8.2%) for youth with eating disorders. Of the 105 patients, 48.6% had a DSM-IV diagnosis of AN, 16.8% had a DSM-IV diagnosis of BN, 24.8% had a DSM-IV diagnosis of Eating Disorder Not Otherwise Specified, 1.0% had a diagnosis of selective eating, and 8.9% were assessed as not meeting criteria for a clinical diagnosis of an eating disorder.

Measures Demographic information and clinical indices. Demographic informa-

tion of age and education was collected for each patient participant at presentation to the eating disorder treatment program. Eating disorder diag- noses were made by experienced clinicians using a semi-structured clinical interview and based on a consensus diagnosis between two interviewers (pediatrician and a psychiatrist or psychologist).

Training vignettes. Five training vignettes were developed by a team of two psychologists and one psychology graduate student. The vignettes were developed based on the characteristics of a combination of real patients. Each vignette was designed to portray patients with different eating disorder diagnoses and varying levels of severity.

Eating Disorder Inventory–2 (EDI-2). The EDI-2 (Garner, 1991) is a self-report questionnaire of disordered eating symptoms. This 91-item scale follows a 6-point Likert scale format with responses ranging from 0 (least pathological) to 3 (most pathological). It is divided into 11 clinical dimen- sions, 3 of which are relevant to this study: Body Dissatisfaction (e.g., “I think that my stomach is too big”), Drive for Thinness (e.g., “I am terrified of gain- ing weight”), and Bulimia (e.g., “I stuff myself with food). The subscales of the EDI-2 have been found to be internally consistent with Cronbach alphas ranging from 0.80 to 0.91 (Eberenz & Gleaves, 1994).

Multidimensional Anxiety Scale for Children (MASC). The MASC (March, 1997) is a 39-item self-report questionnaire designed to assess anx- iety in children and youth. Only the MASC Social Anxiety subscale and

Clinician Rated Eating Disorder Symptom Scale 339

Total Score were utilized in the current study. March (1997) demonstrated excellent internal reliability for this measure (Cronbach alpha = 0.90).

Procedure

The current study was approved by the institutional ethical review board of the Children’s Hospital of Eastern Ontario. All clinician participants partici- pated in a group two-hour training session or an individual 2-hour training session. The training session reviewed the purpose of the measure and the factor structures. Each participant received instruction in how to com- plete the measure and was encouraged to ask any questions. Participants completed one practice vignette and received feedback on their ratings. Participants then completed an EDS3 rating sheet on a training vignette. Participants were certified to complete the EDS3 when they had a spearman correlation of .70 with the standard scoring for a vignette.

All clinician participants were then asked to complete the EDS3 as part of the standard assessment battery and clinical interview, following outpa- tient visits on a small sample of their regular patients, and on new and graduating patients from the eating disorder day treatment program.

Patient participants completed the standard assessment battery and day treatment entrance and exit packages. They were not required to com- plete any additional measures than the standard clinical measures already in use and approved by the research ethics board for program evaluation and research purposes.

Statistical Methods

Data were analyzed using SPSS 15.0. Inter-rater reliability was examined using a Spearman correlation between clinician ratings on the vignettes and the scoring key on the training vignettes. Face validity was determined through the focus group and expert consultation process. The factor struc- ture of the EDS3 was examined using a Principal Component Analysis (PCA) with Varimax rotation. The internal consistency of the EDS3 was examined by evaluating the item-total correlations and the Cronbach alpha coefficients for the whole scale and the subscales. Convergent validity was evaluated by comparing the EDS3 to validated self-report scales (EDI-2 and MASC) using Pearson Product Moment correlations. Test-retest reliability over a two-week period was established using correlation analyses. All tests of significance were based on alpha set at 0.05.

RESULTS

Inter-Rater Reliability

All raters were trained to a reliability of greater than 0.70 using spearman cor- relation to the scoring key on the training vignettes (Mean = 0.865, n = 16).

340 K. A. Henderson et al.

Test-Retest Reliability

Test-retest reliability was established in order to ensure that the EDS3 was a stable measure over time. Seven raters completed an EDS3 on Vignette #1, and re-completed the measure again 2 weeks later. Results of a correlation analysis (r = .902) revealed excellent test-retest reliability for the EDS3.

Factor Analysis of the EDS3

In order to determine the sampling adequacy of the dataset, the Kaiser- Meyer-Olkin (KMO) test was performed, indicating that the data were suited for a factor analysis (KMO = .922). The 18-item EDS3 was subjected to a PCA with Varimax rotation. Following the authors’ theoretical construction of the scale, the items were forced into five factors. An examination of the factor loadings revealed that the items “Other Compensatory Behaviors” of the Eating Disordered Behaviors subscale, and “Agitation Behaviors” of the Eating Disorder Anxiety subscale, loaded poorly (less than 0.40) on their hypothesized subscales. Therefore, a second PCA was performed with those two items removed and the items forced into 5 factors. Items which had factor loadings of greater than 0.40 were retained (Tabachnick & Fidell, 2007). The results of the second PCA revealed a 5-factor structure which accounted for 77.75% of the total variance (see Table 1). The 5 compo- nents were identified as Eating Disordered Behaviors, composed of two subsections AN-behaviors (2 items) and BN-behaviors (2 items), Eating Disordered Cognitions (5 items), Eating Disorder Anxiety (3 items), and Treatment Progress (4 items). Four of the items cross-loaded strongly on more than one factor (i.e., items 1, 10, 12, & 16). Further inspection of these items indicated that they were theoretically important to the EDS3, and that they had strong item-total correlations in their respective sub- scales. Therefore, these 4 items were retained in the original hypothesized subscale.

Internal Consistency Reliability of the EDS3

In order to determine that the EDS3 was a reliable measure, item-total cor- relations were calculated, revealing that each item had a strong correlation (0.51 to 0.87) with the total score (of their respective subscale). Cronbach alpha for the EDS3 was strong (0.93), indicating that the total scale is a reli- able measure. Reliabilities and descriptives for each subscale were calculated (see Table 1). The Cognitions, Anxiety, and Treatment Progress subscales demonstrated good reliability (0.80 to 0.92), while the AN-Behaviors and BN-behaviors subscales demonstrated adequate reliability (0.69 and 0.77 respectively).

Clinician Rated Eating Disorder Symptom Scale 341

TABLE 1 Factor Loadings and Descriptive Statistics of the Items on the EDS3

Item F1 F2 F3 F4 F5

Food restriction 0.52 Excessive exercise 0.92 Binge eating 0.88 Vomiting 0.87 Body image dissatisfaction 0.83 Body image distortion 0.76 Fear of gaining weight and/or becoming “fat” 0.79 Importance of appearance to self-esteem 0.71 Preoccupation with food/weight 0.76 Food anxiety 0.41 Eating rituals 0.84 Social anxiety related to eating/body image 0.45 Motivation for treatment 0.81 Cooperation during treatment 0.83 Distorted beliefs about treatment 0.72 Ability/hope for recovery 0.74 % Variance 9.03 10.36 28.20 8.82 21.35 Mean 2.20 .79 8.50 3.81 4.50 S.D. 1.83 1.44 3.92 2.55 3.20 Cronbach alpha 0.69 0.77 0.91 0.80 0.92

Note. F1 = AN-Behaviors; F2 = BN-Behaviors; F3 = Cognitions; F4 = Anxiety; F5 = Treatment Progress. Factor loadings < 0.40 are not included.

Convergent Validity

In order to establish the EDS3 as a valid measure of eating disorder symp- tomatology, convergent validity was assessed by examining the relationship between the individual subscales of the EDS3 and a general/validated mea- sure of eating pathology and anxiety. Pearson Product Moment Correlations were performed to assess the relationship between the EDS3 subscales and total score, three subscales of the EDI-2 (i.e., Drive for Thinness, Body Dissatisfaction, and Bulimia), and the MASC Social Anxiety subscale and total score (see Table 2).

DISCUSSION

Psychometric Properties

Results from the current study indicate the EDS3 is a reliable and valid clinician measure to assess eating disorder behaviors (AN-behaviors and BN-behaviors), eating disorder cognitions, eating disorder anxiety, and treat- ment progress in youth suffering from eating disorders. It is a reliable tool for assessing the severity of an eating disorder in youth. It is time efficient, simple and offers an alternative to self-report and interview.

TA B

LE 2

C o rr

el at

io n s

A m

o n g

th e

E D

S3 , E at

in g

P at

h o lo

gy , an

d A

n xi

et y

M ea

su re

1 2

3 4

5 6

7 8

9 10

1. E D

S3 : A

N -b

eh av

io rs

1 .2

4∗ ∗

.6 0∗

∗ .6

5∗ ∗

.6 2∗

∗ .7

8∗ ∗

.6 3∗

∗ .2

1 .1

6 .1

8 .1

6 2.

E D

S3 : B

N -b

eh av

io rs

1 .1

7∗ ∗

.1 5∗

.2 0∗

∗ .3

5∗ ∗

.3 1∗

.7 9∗

∗ -. 02

.0 5

3. E D

S3 : C o gn

iti o n s

1 .7

3∗ ∗

.7 0∗

∗ .9

0∗ ∗

.6 3∗

∗ .4

3∗ ∗

.2 0

.2 9∗

.2 5∗

4. E D

S3 : A

n xi

et y

1 .6

9∗ ∗

.8 5∗

∗ .4

8∗ ∗

.2 7∗

.1 2

.2 1

.2 7∗

5. E D

S3 : T re

at m

en t p ro

gr es

s 1

.8 6∗

∗ .2

8∗ .2

7 -. 05

.0 7

.1 5

6. E D

S3 To

ta l

. 1

.6 2∗

∗ .4

4∗ ∗

.3 2∗

.2 7∗

.2 8∗

7. E D

I- 2:

D ri ve

fo r

th in

n es

s 1

.8 5∗

∗ .4

7∗ ∗

.3 8∗

∗ .4

5∗ ∗

8. E D

I- 2:

B o d y

d is

sa tis

fa ct

io n

1 .4

5∗ ∗

.4 8∗

∗ .5

7∗ ∗

9. E D

I- 2:

B u lim

ia 1

.0 5

.1 0

10 . M

A SC

: So

ci al

an xi

et y

1 .7

4∗ ∗

11 . M

A SC

: To

ta l an

xi et

y sc

o re

∗ p <

.0 5.

∗∗ p

< .0

01 .

342

Clinician Rated Eating Disorder Symptom Scale 343

Face and content validity were ensured through the use of focus groups and expert consultation regarding the design, wording, and factor structure of the measure. Principal Components Analysis statistically demonstrated that deleting two items resulted in a 16-item scale that accounted for 78% of the total variance and provided a 5-factor solution consistent with the the- oretically constructed 5 subscales. Inter-rater and test-retest reliability were good supporting the reliability of the scale.

The total scale and the subscales demonstrated adequate to good internal consistency reliability and good item-total correlations. The Eating Disorder Cognitions, Eating Disorder Anxiety and Treatment Progress sub- scales had good reliability, while the BN-behaviors and the AN-behaviors subscales had adequate internal consistency. The lower internal consistency scores for both AN and BN- behaviors is likely attributed to the fact that these subscales only include two items.

The scale also demonstrated good convergent validity with a cur- rent gold standard standardized self-report measure of eating disorders (EDI-2) and anxiety (MASC). The EDS3 AN-behaviors subscale was posi- tively correlated with the Drive for Thinness subscale of the EDI-2, while the BN-behaviors subscale demonstrated a strong, positive relationship with the Bulimia subscale of the EDI-2. Additionally, the Eating Disorder Cognitions subscale was significantly correlated in the hypothesized direction with Body Dissatisfaction and Drive for Thinness subscales of the EDI-2. The Cognitions subscale also demonstrated a significant relationship with social anxiety and overall anxiety. Finally, the Anxiety subscale of the EDS3 demonstrated a sig- nificant, positive relationship with the overall Anxiety score, and a positive, though non-significant, correlation with the Social Anxiety subscale of the MASC (see Table 2). Therefore, the results of our current study support the validity of the EDS3 as a measure of eating disorder severity. Furthermore, they also provide support for the new construct of eating disorder anxiety that was developed for the EDS3.

Utility and Implications of the EDS3

The EDS3 provides clinicians and researchers with a reliable and valid mea- sure for regular monitoring of eating disorder illness severity in youth. It is a measure that can be learned through a brief training session. The EDS3

is available through a ‘train the trainer model’ and has supporting training manuals. The EDS3 does not require a significant financial or time com- mitment from clinicians or facilities. The EDS3 manual, item booklet, and rating sheet are available at no cost from the authors at the following web- site: www.praedfoundation.org or by contacting Dr. Henderson, C.Psych. at [email protected]. The website outlines the usage agreement with the authors. It is hoped that the elimination of these obstacles will enhance the use of regular monitoring of illness severity and treatment progress for

344 K. A. Henderson et al.

the purpose of decision-making within and across treatment centers. The EDS3 can be seamlessly merged into regular clinical care, facilitating the col- lection of information on treatment effectiveness and helping to address the dearth of information in eating disorder outcomes research. It is essential that the international eating disorder community develop a better understanding of treatment effectiveness to improve patient care, treatment outcome, and secure funding for programs internationally. The EDS3 may provide one small step toward this important goal.

Limitations and Future Directions

The validity and reliability of the EDS3 has only been tested in one patient population in Canada. Future research should include reliability and valid- ity testing in a broader sample including males. Recommendations regarding sample size in factor analysis vary across researcher. While the sample size of this study meets the recommended sample size and participant per item ratio by some researchers (Gorsuch, 1983; Guadagnoli & Velicer, 1988), it does not meet those recommended by others; therefore, it would be helpful to repli- cate the factor structure with a larger sample size in the future (Tabachnick & Fidell, 2007). The authors plan to continue to evaluate the predictive util- ity of the EDS3 in decision-making in all levels of care in specialized eating disorder programs across the province of Ontario in Canada and evaluate the EDS3s sensitivity to change. Future research will also include the devel- opment of a youth self-report and a parent report version of the EDS3 to be used in conjunction with the clinician measure.

REFERENCES

Carter, J. C., Stewart, D. A., & Fairburn, C. G. (2001). Eating disorder examina- tion questionnaire: Norms for young adolescent girls. Behaviour Research and Therapy, 39, 625–632.

Dennis, K., le Grange, D., & Bremer, J. (2006). Olanzapine use in adolescent anorexia nervosa. Eating & Weight Disorders, 11, 53–56.

Dunican, K. C., & DelDotto, D. (2007). The role of olanzapine in the treatment of anorexia nervosa. The Annals of Pharmacotherapy, 41, 111–115.

Eberenz, K. P., & Gleaves, D. H. (1994). An examination of the internal consistency and factor structure of the eating disorder inventory-2 in a clinical sample. International Journal of Eating Disorders, 16 , 371–379.

Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16 , 363–370.

Fairburn, C. G., & Cooper, Z. (1993). The Eating Disorder Examination. In C. G.Fairburn & G. Wilson (Eds.), Binge eating: Nature, assessment and treatment (pp. 333–356). New York: Guilford Press.

Clinician Rated Eating Disorder Symptom Scale 345

Fichter, M. M., & Quadflieg, N. (2000). Comparing self- and expert rating: A self- report screening version (SIAB-S) of the structured interview for anorexic and bulimic syndromes for DSM-IV and ICD-10 (SIAB-EX). European Archives of Psychiatry and Clinical Neuroscience, 250, 175–185.

Garner, D. M. (1991). Eating Disorder Inventory-2. Odessa, FL: Psychological Assessment Resources.

Garner, D. M., Olmsted, M. P., Bohr, Y., & Garfinkel, P. E. (1982). The eating attitudes test: psychometric features and clinical correlates. Psychological Medicine, 12, 871–878.

Geller, J., Brown, K. E., Zaitsoff, S. L., Menna, R., Bates, M. E., & Dunn, E. C. (2008). Assessing readiness for change in adolescents with eating disorders. Psychological Assessment, 20, 63–69.

Geller, J., Cockell, S. J., & Drab, D. L. (2001). Assessing readiness for change in the eating disorders: the psychometric properties of the readiness and motivation interview. Psychological Assessment, 13, 189–198.

Gorsuch, R. L. (1983). Factor analysis. (2nd ed.) Hillsdale, NJ: Erlbaum. Guadagnoli, E., & Velicer, W. F. (1988). Relation of sample size to the stability of

component patterns. Psychological Bulletin, 103, 265–275. Gusella, J., Butler, G., Nichols, L., & Bird, D. (2003). A brief questionnaire to assess

readiness to change in adolescents with eating disorders: Its applications to group therapy. European Eating Disorders Review, 11, 58–71.

Jones, J. M., Bennett, S., Olmsted, M. P., Lawson, M. L., & Rodin, G. (2001). Disordered eating attitudes and behaviours in teenaged girls: a school-based study. Canadian Medical Association Journal, 165, 547–552.

Kaye, W. H., Nagata, T., Weltzin, T. E., Hsu, L. K., Sokol, M. S. . .Deep, D. (2001). Double-blind placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type anorexia nervosa. Biological Psychiatry, 49, 644–652.

le Grange, D., & Lock, J. (2005). The dearth of psychological treatment studies for anorexia nervosa. International Journal of Eating Disorders, 37 , 79–91.

Lock, J. & Gowers, S. (2005). Effective interventions for adolescents with anorexia nervosa. Journal of Mental Health, 14, 599–610.

Lyons, J. S. (1998). The Severity and Acuity of Psychiatric Illness Scales Manual – Child and adolescent version. San Antonio, TX: The Psychological Corporation, Harcourt Brace & Company.

Lyons, J. S., Griffin, E., Fazio, M., & Lyons, M. (1999). Child and adolescent needs and strengths: An information integration tool for children and adolescents with mental health challenges (CANS-MH), manual. Chicago, IL: Buddin Praed Foundation.

Lyons, J. S., Kisiel, C., Dulcan, M., Cohen, R., & Chesler, P. (1997). Crisis assessment and psychiatric hospitalization of children and adolescents in state custody. Journal of Child & Family Studies, 6 , 311–320.

Lyons, J. S., Libman-Mintzer, L. N., Kisiel, C. L., & Shallcross, H. (1998). Under- standing the mental health needs of children and adolescents in residential treatment. Professional Psychology: Research and Practice, 29, 582–587.

March, J. S. (1997). Multidimensional anxiety scale for children technical manual. Toronto, ON: Multi-Health Systems Inc.

Norris, M., Spettigue, W., Buchholz, A., & Henderson, K. A. (2007). Challenges asso- ciated with controlled psychopharmacological research trials in adolescents

346 K. A. Henderson et al.

with eating disorders. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 16 , 167–172.

Pinto, M. A., Guarda, A. S., Heinberg, L. J., & Diclemente, C. C. (2006). Development of the eating disorder recovery self-efficacy questionnaire. International Journal of Eating Disorders, 39, 376–384.

Prochaska, J. O., & Norcross, J. C. (2001). Stages of change. Psychotherapy, 38, 443–448.

Reijonen, J. H., Pratt, H. D., Patel, D. R., & Greydanus, D. E. (2003). Eating disorders in the adolescent population: An overview. Journal of Adolescent Research, 18, 209–222.

Spettigue, W., Buchholz, A., Henderson, K., Feder, S., Moher, D. . . Ledoux, S. (2008). Evaluation of the efficacy and safety of olanzapine as an adjunctive treatment for anorexia nervosa in adolescent females: A randomized, double- blind, placebo-controlled trial. BMC Pediatrics, 8, 4–13.

Striegel-Moore, R. (2005). Health services research in anorexia nervosa. Inter- national Journal of Eating Disorders, 37 , S31–S34.

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. (5th ed.) Boston, MA: Allyn and Bacon.

Wilson, G. T., Grilo, C. M., & Vitousek, K. M. (2007). Psychological treatment of eating disorders. American Psychologist, 62, 199–216.

Zipfel, S., Reas, D. L., Thornton, C., Olmsted, M. P., Williamson, D. A. . . Beumont, P. (2002). Day hospitalization programs for eating disorders: A systematic review of the literature. International Journal of Eating Disorders, 31, 105–117.

multiple sites or posted to a listserv without the copyright holder's express written permission. However, users

may print, download, or email articles for individual use.