This study is a randomized, double-blind, placebo-controlled trial. The ethics committee at the Children's Hospital of Eastern Ontario has approved the study protocol.
Adolescent females living within Eastern Ontario, Canada, who are between the ages of 12 and 17 and have been diagnosed with AN (either restricting or binge/purge type) or Eating Disorder Not Otherwise Specified (EDNOS) with a BMI of less than or equal to 17.5, will be offered participation in the study. Adolescents with a diagnosis of EDNOS and a low BMI are included as it is recognized that though not necessarily meeting all of the criteria for Anorexia Nervosa (see inclusion criteria), these two diagnoses are conceptualized similarly in terms of development and treatment.
A randomized, double-blind, placebo-controlled design will be used to determine the efficacy and safety of olanzapine in treating AN. Consecutive female patients diagnosed with a severe eating disorder, who meet eligibility criteria, and who consent to participate in the study, will be randomly assigned to one of two groups: olanzapine (OLA) or placebo (PLA). A medication schedule will be in place for a total of 14 weeks (see Table for Schedule of Events). Medication will be started at a low dose and increased up to a predetermined maximum dose, or until an intolerable side effect occurs. Both groups will receive either olanzapine or placebo for a total of 14 weeks. Both groups will continue to receive standard care that includes psychological, nutritional, and medical treatment.
Screening of participants for inclusion criteria and exclusion criteria will take place at the time of initial assessment. The assessment consists of a full day, during which patients complete psychological measures (see Table ) and undergo psychiatric, nutritional, and medical assessments. After completing a variety of psychological measures, the patient will be interviewed by members of the treating team. A psychiatrist or a psychologist along with a medical doctor will conduct the initial interview. Following the interview, the patient then completes a nutritional assessment with a registered dietician. During this time, the patient's parents are also interviewed. The medical assessment consists of a physical examination, along with documentation of vital signs (temperature plus lying and standing heart rate and blood pressure), and laboratory tests (see Table ). After the formalized assessment is complete, determination of the patient's eligibility for the outlined study is determined.
To be eligible to participate in this trial, candidates must meet the following inclusion criteria at screening:
1. Patient must be female
2. Patient and one parent must give written informed Consent or Assent.
3. Must be between ages of 12 and 17 (younger than 18) at beginning of Trial.
4. Based on the Diagnostic and Statistical Manual of Mental Disorders IV [1
] must have fulfilled the criteria for diagnosis of Anorexia Nervosa (of which there are two types: restricting type or binge-eating/purging type) or Eating Disorder Not Otherwise Specified, with a Body Mass Index of less than or equal to 17.5.
Candidates will be excluded from trial entry if any of the following exclusion criteria exist at screening:
1. Known sensitivity to any of the products to be administered.
2. Currently receiving treatment with any other anti-psychotic medication, mood stabilizer, or stimulant.
3. Treatment with medication known to interact with olanzapine, (eg. Fluvoxamine, Ciprofloxacin).
4. Known diagnosis of: diabetes, impaired glucose tolerance, hyperlipidemia, hepatic dysfunction, substance abuse, narrow angle glaucoma, paralytic ileus, or pancreatitis, or any other medical illness that would be considered to significantly impact treatment or recovery from AN
5. Inability to comply with trial requirements including lack of comprehension of English.
6. Pregnant or breast-feeding.
7. Other unspecified reasons that, in the opinion of the Investigator, make subject unsuitable for enrollment.
Laboratory exclusion criteria:
• Total white cell count < 2.5
• Neutrophil count < 1.0
• Liver function tests (AST/ALT > 2× normal)
• Positive pregnancy test
• EKG – QTc > 440 msec or arrythmia other than sinus bradycardia; conduction
abnormalities prolonged QTc or other
Consent and Enrollment
When the adolescent meets inclusion and not exclusion criteria, the physician or psychiatrist will provide the patient and her family with a brief description of the study, the extent and implications of adolescent and parental involvement, and the overall purpose of the research study. The adolescent and family will be asked to consider involvement in the study and permission will be requested for a follow-up phone call from the project coordinator within the next 72 hours. As well, a meeting with the study psychiatrist will be offered during which the nature of the study will be described in greater detail and questions will be answered. The family will be reassured that they will continue to receive standard care regardless of their involvement in the study.
The project coordinator will read through the assent and consent forms with the patient and her parent(s), have all members sign, and will provide them with their first scheduled visit (see Table for Schedule of Events). It is expected that most patients who consent will be enrolled into the study within two weeks of their initial assessment (screening visit). If, however, more than two weeks have elapsed since the screening visit, their baseline blood work, urinalysis, weight, height, vital signs and EKG will be repeated (but not bone mineral density).
The consent of only one parent will be required. Adolescents whose parent(s) consent to the study, but who themselves do not assent or consent, will not be included in the study and in no way will this compromise their care.
Participants, having provided the appropriate consent, are to be enrolled at Visit 1, after all baseline evaluations have been completed. Participants will be randomized (1:1 ratio) to olanzapine or placebo using a computer generated block-size randomization schedule. To help ensure adequate allocation concealment, the computer-generated sequence will be kept centrally at the Chalmers Research Group (CRG) on computer and can only be activated when there is an eligible participant to randomize. The research coordinator will complete an eligibility criteria checklist using a 24/7-telephone randomization system. Upon receipt of an affirmative answer of the participant eligibility, the system will release a randomization number that will correspond to a bottle of medication stored in pharmacy.
All personnel at the investigational site will be blinded to the treatment except for the Pharmacist (or designee). The participant and her parents will also be blinded in a similar fashion. The look-a-like placebo will be dispensed in identical bottles to the experimental medication. The research pharmacist will be responsible for labelling the medication with the participant's randomization number and placing the required dose for the 14-week treatment period. Only the pharmacist will be aware of the randomization sequence and that sequence will not be broken until analysis of the results is completed. The sequence may only otherwise be broken if requested by the independent Data Safety and Monitoring Committee (DSMC). The statistician will also be blinded to the groups when managing data sets.
The recommended dose of olanzapine for treating psychosis in adult patients with schizophrenia is 5–20 mg/day, with doses of 30–40 mg/day not uncommon in clinical practice [27
]. A study currently underway at the Ottawa Hospital (General Campus) to evaluate the efficacy and safety of olanzapine for the treatment of adults with AN used 5 mg of olanzapine as a starting dose, and increased to 10 mg/day (Dr. H. Bissada, personal communication, September, 2004).
The dosing regimen to be used for the outlined study has been derived from both current clinical practice parameters and available literature. Low doses of olanzapine (2.5–5.0 mg daily) are typically prescribed for the treatment of children and adolescents with severe eating disorders. In older or more agitated patients the dose may be increased to 7.5 mg or even 10 mg per day. These doses are typically well tolerated, with sedation as the main noted side effect. In anticipation of this trial, members of the American Academy for Eating Disorders were polled through the online email service ("AED listserv") regarding their dosing of olanzapine in clinical practice. The results of this poll revealed a wide range in the dosing of olanzapine, from 1.25–20 mg/day.
With the above guidelines in mind, olanzapine will be initiated at the lowest dose available by tablet formulation (1.25 mg). The medication will be slowly increased over a two-week period assuming patient tolerance to a maximum dose of 7.5 mg/day (1.25 mg/day for three days, followed by 2.5 mg/day for four days, 5 mg/day for one week, and then the target dose of 7.5 mg/day for 10 weeks). After a period of 10 weeks at 7.5 mg, the medication will be tapered and discontinued over a further two weeks. Table outlines the exact dose schedule as part of the trial schedule.
Patients' psychological well-being will be measured utilizing the following six scales:
Eating disorder symptoms will be measured using the Eating Attitudes Test
], the EAT-26 is a 26-item self-report questionnaire with a 6-point scale that measures 3 domains of disordered eating: dieting, bulimia and food preoccupation, and oral control. The EAT-26 is a widely used, well-validated instrument in this clinical population. In this study, the total EAT-26 score will be used.
Psychiatric Symptoms will be measured using the Children's Acuity of Psychiatric Illness Scale-Child and Adolescent
], The CAPI is an outcome measure for children and adolescents that is designed to monitor change deriving from contact with mental health services. The CAPI yields a total raw score and four subscale scores, including Risk Factors, Symptoms, Functioning, and Systems support. The scale has good inter-rater reliability (.78 to .85) and good internal consistency (.87) across a variety of settings [29
Depressive symptoms will be measured using the Children's Depression Inventory
]. The CDI is a 27-item questionnaire, which asks respondents to endorse statements about themselves reflecting cognitive, behavioural and somatic symptoms of depression. Items are rated on a 3-point scale indicating symptom severity. Total scores on the CDI can range from 0 to 52, with higher scores indicating higher levels of depressive symptomatology. The measure has been shown to have acceptable internal consistency (alphas = .71 to .89) as well as good discriminant validity when classifying children with no significant psychopathology versus depressed children [30
]. The total score of the CDI will be used in this study.
Anxiety symptoms will be measured using the Multidimensional Anxiety Scale for Children
]. The MASC is a 39-item 4-point Likert-style self-report which demonstrated a satisfactory to excellent range for the intraclass correlation coefficient for the three subscales: physical symptoms, harm avoidance, and social anxiety in a sample of youth (ICC = 0.79 to 0.92) [32
]. The total score on the MASC will be used in this study.
Clinicians will be asked to complete the Eating Disorder Symptom Severity Scale
] which will examine their observations of the youth's eating disorder symptoms such as: dieting behaviours, body image preoccupations, agitation related to eating/body image, fear of fat, food anxiety, as well as overall motivation for treatment. Clinicians involved in the study have attended a workshop meant to improve inter-rater reliability for this scale (r = .75 to r = .80 was deemed acceptable)
Parents will be asked to complete the Child Behavior Checklist
] in order to evaluate their perceptions of their child's mood and anxiety. The Child Behaviour Checklist (CBCL) has been demonstrated to be a reliable measure when completed by parents, with a mean alpha 0.80 being reported [34
]. The CBCL is a widely used, well-validated instrument in pediatric populations.
The psychological measures will be completed at baseline, weeks 6, 12, 15 and 40 (see Table ). The patient will complete the EAT, CDI and the MASC; the parent will complete the CBCL; and the treating physician and therapist will complete the CAPI and EDS3.
Medical Monitoring (see Table for Schedule of Events)
The baseline medical assessment will include a physical examination, monitoring of side effects/symptoms (utilizing the side-effect checklist), completion of the Acquired Involuntary Movement Scale (AIMS) to monitor extrapyramidal symptoms, taking of vital signs (temperature plus lying and standing heart rate and blood pressure), an EKG, bone mineral density, and comprehensive lab work (see Table ). An EKG will be repeated at Weeks 2, 4, and 15. At each clinic visit, a urine sample (urinalysis and pregnancy test) will be taken, vital signs completed, and weight recorded.
The AIMS and an adverse event checklist are also completed at each clinic visit.
Primary Outcome Measures
The primary outcome measure will be the change from baseline on the total score of the EAT-26 score measured at week 12 prior to topping off the medication, and BMI change over the first 12 weeks of treatment.
Secondary Outcome Measures
Change from baseline in the EAT-26 score measured at week 6, 15 (i.e. off medication) and at the end of the maintenance period (week 40), as well as BMI change measured at the same time points will be analyzed as for the primary outcomes. Change from baseline in the CAPI, the CDI, the MASC, the EDS3 and the CBCLwill be calculated for weeks 6, 12, 15 and 40.