Healthy Lifestyles for Mentally Ill People Who Have Experienced Weight Gain From Their Antipsychotic Medications

Not Applicable

Completed

Conditions: Diabetes
Weight Gain
Obesity

Interventions: Behavioral: Behavioral Weight Loss Program

Registration Number: NCT00344500

Lead Sponsor: VA Office of Research and Development

Brief Summary: This program aims to help Veterans who take antipsychotic medications lose weight. The investigators use a program based on the American Diabetes Association's "Diabetes Prevention Program," and the investigators have modified it to fit the lifestyles of people with mental illness. All participants are educated about nutrition and cutting down fat intake, how and when to exercise, and the causes of diabetes and how to prevent it. Participants must be Veterans who live within one hour of the West Los Angeles VA hospital.

Detailed Description: Rationale: The focus of this project is to develop a strategy to combat medication associated weight gain, the most problematic side effect of the newer antipsychotic medications. Improvements in long-term health outcomes might then be expected to change quality of life, promote treatment adherence, rehabilitative potential, and decrease resource utilization.

Procedures:

Half of the patients will randomized to the behavioral weight loss program (Lifestyle Balance Program) and do the following: Meet with their psychiatrist and a nutritionist who will go over diet recommendations with the patient Given a 7% weight loss goal Assisted in obtaining a 500 calorie reduction per day Asked to exercise for at least 30 min/day, at least 5 days a week Maintain weekly food and exercise diaries Be quizzed on their knowledge of healthy eating habits and nutrition

The other half of the patients will be randomized to "Usual Care" and will:

Receive pamphlets about Lifestyle Balance, starting exercise, and general nutritional information regarding food pyramids and the amount of calories in fast foods.

Be encouraged to exercise and eat a healthy diet Not receive classes on nutrition or exercise Be seen on the same schedule as the experimental group, in order to equalize contact time.

After six months, if patients in the "usual care group" wish to participate in the more rigorous behavioral weight loss program they will be given that opportunity.

Visits:

Time Line:

Each subject will be followed for one year. Healthy Lifestyles courses will take eight weeks to complete. Visits will be monthly for the remaining 12 months Enrollment: 120 patients Course of study: October 2005-October 2008

At Beginning and Termination:

Informed Consent (at start of study only) Psychiatric interviews and assessments Physical Examination Framingham risk assessment for cardiovascular disease Electrocardiogram Lipid Profile, fasting glucose, Hemoglobin A1C

At Each Visit:

Vitals, weight, waist circumference, BMI, and % Fat Food and exercise diaries will be reviewed

At Month One and Month Six:

Lipid Profile, fasting glucose, Hemoglobin A1C Electrocardiogram Psychiatric assessments

At Month Nine:

Lipid Profile, fasting glucose, Hemoglobin A1C Electrocardiogram

Clinical Care:

All patients will continue their clinical care with their current psychiatrist and general physician. Medications will not be manipulated for the purpose of this study.

Incentives:

The group participating in the behavioral intervention program will receive rewards for achieving weight loss goals and be given incentives such as a pedometer and Slimfast meal replacements.

Recruitment & Eligibility

Status: COMPLETED

Sex: All

Target Recruitment: 122

Inclusion Criteria

Must be a Veteran
Diagnosis of psychotic disorders, schizophrenia, schizoaffective disorder and bipolar illness
Age 18-70
Clinically determined to require ongoing treatment with Second Generation Antipsychotics (SGA) such as olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, clozapine
Experienced weight gain since treatment with SGA's
Inpatient or outpatient at the West Los Angeles VA
Competent to sign informed consent

Exclusion Criteria

Have recently been diagnosed with schizophrenia (less than 1 year)
Are pregnant or breast feeding a baby
Have a medically unstable condition

Study & Design

Study Type: INTERVENTIONAL

Study Design: CROSSOVER

Arm && Interventions

Group	Intervention	Description
Lifestyle Balance	Behavioral Weight Loss Program	Behavioral Weight Loss Program

Primary Outcome Measures

Name	Time	Method
Mean Weight	Weekly/Monthly, up to 1 year	Average weight of subjects attending each of the first 8 weekly visits and the 10 monthly visits which followed, per study group.
Change in Predicted Trajectory of Mean Body Fat Percentage Per GLMM Analysis	12 months	Computed as % body fat at 12 month - % body fat at baseline. General Linear Mixed Model (GLMM) is a full information maximum likelihood approach that permits inclusion of all available data and provides unbiased parameter estimates even if there are missing data under the condition that data are missing at random. The GLMM approach assumes that every patient is on a specific trajectory over time and that both the slope and the shape of this trajectory are a potential function of group membership or other person-level covariates. Using a likelihood ratio test, we found a linear model, assuming the same rate of change throughout the study, provided a good fit to the data compared to other models. We used a linear model of the average rate of change over time (slope) for all comparisons. To illustrate the magnitude of difference between slopes for major outcomes, we report the estimated difference at 12 months between two hypothetical participants with identical baseline characteristics.
Change in Predicted Trajectory of Mean BMI Per GLMM Analysis	12 months	General Linear Mixed Model (GLMM) is a full information maximum likelihood approach that permits inclusion of all available data and provides unbiased parameter estimates even if there are missing data under the condition that data are missing at random. The GLMM approach assumes that every patient is on a specific trajectory over time and that both the slope and the shape of this trajectory are a potential function of group membership or other person-level covariates. Using a likelihood ratio test, we compared different options to model these trajectories and found a linear model, which assumes that the same rate of change is maintained over the whole study, provided a good fit to the data. We used a linear model of the average rate of change over time (slope) for all comparisons. To illustrate the magnitude of difference between slopes for major outcomes, we report the estimated difference at 12 months between two hypothetical participants with identical baseline characteristics.