Recent but limited short term studies have shown that Metformin can slow down weight gain in obese children and in children with psychotropic-induced weight gain, two distinct pediatric populations that are at risk for obesity related co-morbid conditions. The purpose of this study is to conduct a long term prospective pilot cohort study to investigate the use of Metformin to prevent or decrease weight gain in two cohorts of children: 1) children with psychotropic induced weight gain on Metformin and 2) children with BMI above the 95th percentile on Metformin. Both study populations will be enrolled in a lifestyle weight management program

Approximately 21 percent of children, 12-17 years old are diagnosed with DSM IV disorders, with 11 percent exhibiting severe impairment and 5 percent severe emotional difficulties. By 18 years, 1-5 percent of children are diagnosed with bipolar disorder and up to 20 percent of children with depression. As greater numbers of children and adolescents have been diagnosed with these disorders in the last 10 years, the use of psychotropic drugs in the pediatric populations has increased. Many of the drugs prescribed are the newer antipsychotic drugs olanzapine, risperidone, and quetiapine, referred to as atypical antipsychotics. Compared to the older drugs, such as haldol and thorazine, atypical antipsychotics boast an improved safety profile, with fewer side effects such as tardive dyskinesia, extrapyramidal symptoms and hyperprolactinemia. This advantage has led to providers prescribing antipsychotic more frequently not only for psychotic conditions, but also for other behavioral problems, eg., oppositional defiant disorder, mood disorders, and autism spectrum disorders. In many ways, these medicines are life saving. They protect children from the fate of psychosis, unchecked rage and agitation, allowing the them a chance to grow up more normally.

Our study will provide preliminary evidence for the feasibility of using metformin as an adjunct for weight management in two vulnerable pediatric populations. We will apply for external funding for a large scale randomized clinical trial that will test efficacy of metformin in both our study populations with appropriate comparison groups. In addition, results from our exploratory analysis of patient characteristics eg., insulin level, eating behaviors) that may affect response to treatment will provide a basis to generate further hypothesis for mechanism of action.

Primary objective: Describe and compare the pattern of changes in weight trajectory in the (PIW) and (OME) group.

Secondary Objective: To conduct a preliminary investigation of factors(Baseline BMI, adherence, presence of gastrointestinal side effects, HOMA-IR, eating patterns) that influence the response to metformin.

Inclusion Criteria:

1. Children aged 10-17
2. Currently prescribed one of the following psychotropic medications: Haloperidol, perphenazine, clozapine, olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, thioridazine, chlorprothixene, loxapine, mesoridazine, thiothixene or trifluoperazine.
3. Documented weight gain while on prescribed medications
4. Either >5% weight increase from the start of medication through 3 months on, or crossing into the 95th percentile for BMI, or crossing into the 85-95th percentile plus one obesity related complication.
5. Children aged 10-17 years old with BMI >95th percentile and fasting insulin level>21.7U/L not currently on psychotropic medications

Exclusion Criteria:

1. History of liver disease
2. History of kidney disease
3. Abnormal creatinine
4. Abnormal liver function blood levels -