Diabetes Prevention Program in Schizophrenia [DPPS]

Inclusion Criteria: People who are at least 18 years old, diagnosed as having schizophrenia, confirmed through a structured clinical interview (SCID-P) for DSM IV, treated with antipsychotic drugs at least for 2 years and deemed clinically stable. Clinical stability is operationally defined as absence of a relapse warranting hospitalization in the preceding six months. People who are deemed as "pre-diabetics" in accordance with the diagnostic criteria established by the American Diabetic Association (ADA, 2004) as following: impaired fasting glucose (IFG) indicating fasting plasma glucose ranging between 100-125 mg/dl or 5.6 - 6.9 mmol/l.; and impaired glucose tolerance (IGT) indicating 2-h post-load glucose ranging between 140-199 mg/dl or 7.8 - 11.0 mmol/l. People who have gained > 10% body weight since??? Or body mass index > 24 kg/m², except Asian decent at 22 kg/m² or greater?? Competent to provide informed consent to voluntarily participate in the study. Exclusion Criteria: People who meet the criteria for diabetes, (i.e., repeat fasting blood glucose (FBG) >7 mmol/l or, 2 hr. post-load glucose >11.1 mmol. as determined by a 2 hr. glucose tolerance test (GTT)). People with evidence of clinically significant liver disease, renal or gastrointestinal impairments, as suggested by clinical history and liver and kidney functions tests. Any impairment deemed clinically significant would be a relative contra-indication for the use of metformin. Women in the child bearing age, who are not willing to use contraceptive measures. People with other comorbid disorders such as clinically significant heart or lung disease that may prevent participation in various physical activities or disorders of glucose metabolism (e.g., Cushing's Syndrome, Acromegaly, and chronic pancreatitis). People with treatments that would interfere with participation or completion of the protocol (e.g., in shared care and planning to be discharged shortly from the clinic), or having a confounding effect on the measurement of the primary outcomes of the study (prescription weight loss drugs, lipid lowering agents?) People with weight loss >10% in past 6 months