Measures

Well-validated self-report questionnaires were administered by a trained psychometrist. Demographic characteristics were collected using a self-report parent intake form, developed by members of the ED program. For the purposes of this study, parental employment, primary language spoken at home, and whether the child/adolescent has an identified learning disability or individualized education plan (IEP) were examined.

A patient’s height is taken at admission, and weight is taken daily, including at admission and discharge, by a nurse using a calibrated scale (patient is in a gown, post voiding, with back to the scale). Body mass index (BMI) is used as a marker of medical status and used to track weight gain and proximity to the patient’s treatment goal weight (TGW) range. Using the publicly available World Health Organization’s (18, 19) Child Health Growth References macros for the IBM Statistical Package for the Social Sciences (SPSS), BMI-for-age z-scores were calculated. Cutoff interpretations are as follows: normal ≤ +1 standard deviation (SD) to ≥ −2 SD; thinness < −2 SD; severe thinness < −3 SD (20).

The EDEQ-A was used to measure ED psychopathology. As the EDEQ-A was only administered to patients 13 years of age or older due to developmental appropriateness, children <13 years were excluded from analyses. The EDEQ-A is a self-report questionnaire with 36 items, that produces a global score (average of the four subscales) and four subscale scores (Restraint, Eating Concern, Shape Concern, and Weight Concern) and has been found to have strong psychometric properties (21).

The Eating Disorder Inventory-3 (EDI-3) (22) was administered to measure ED attitudes and cognitions. The EDI-3 discriminates between samples (clinical versus nonclinical) and diagnoses [AN, bulimia nervosa (BN), partial AN/BN], and is documented to have good internal reliability for ED patients (α = 0.76–0.92) (23). The drive for thinness and body dissatisfaction subscales were used in this study.

Depressive symptoms were measured using the CDI (24) and levels of anxiety with the MASC (22). The CDI total score is reported to have excellent internal consistency (Cronbach’s α) in a clinical sample of females diagnosed with an ED (α = .93) (23). The MASC total score has high internal reliability in community samples (α = .90) and excellent internal consistency within clinical samples of children and youth with EDs (α = .92) (25). For the purposes of this study, only the total scores from the measures of depression and anxiety were used in our analyses. In addition, suicidality was assessed using a single item on the CDI (question #9) that asks respondents to indicate their level of suicidal ideation over the past 2 weeks.