Carrizo 2009.

Study characteristics
Methods	Study design: Randomized controlled trial Study grouping: Parallel group How were participants recruited: Patients were invited at the outpatient centre Type of RCT: Double‐blind
Participants	Baseline characteristics Clozapine + metformin Sex (N, % female, % male): female 4 (17%), male 20 (84%) Age (mean, SD): 39.6 (SD 9.7) years Number of participants: 24 Clozapine + placebo Sex (N, % female, % male): female 7 (23%), male 23 (77%) Age (mean, SD): 38.3 (SD 8.7) years Number of participants: 30 Overall Sex (N, % female, % male): female 11 (20%), male 43 (80%) Age: not reported Number of participants: 54 Ethnicity: mixed Hispanic (93%), white (7%) Inclusion criteria: Inclusion criteria were to be under clozapine treatment for at least 3 consecutive months; to be older than 18 years, free of hormone replacement and have normal physical and laboratory tests (kidney, thyroid and liver). Exclusion criteria: Elevated serum glucose and lipid levels were not considered as exclusion criteria. No participants had symptoms of type 2 diabetes. Pretreatment: No significant differences were observed between the groups at baseline in any variable.
Interventions	Intervention characteristics Clozapine + metformin Class of drug: Antidiabetic, ET (GlucophageXR, Merck, Venezuela) Dose: 500 mg the first 2 weeks and then1000 mg daily (n=31) Frequency: Daily Duration: 14 weeks Who delivered the intervention? Psychiatrist and study coordinator. Type of intervention: Pharmacological Clozapine + placebo Class of drug: Placebo Dose: ‐ Frequency: Daily Duration: 14 weeks Who delivered the intervention? Psychiatrist and study coordinator. Type of intervention: Placebo
Outcomes	Fasting blood sugar Outcome type: Continuous Reporting: Fully reported Scale: mg/dl Direction: Lower is better Data value: Change from baseline BMI Outcome type: Continuous Scale: Kg/m2 Direction: Lower is better Data value: Change from baseline Cholesterol Outcome type: Continuous Data value: Change from baseline Waist circumference Outcome type: Continuous Unit of measure: cm Diastolic blood pressure Outcome type: Continuous Data value: Change from baseline Systolic blood pressure Outcome type: Continuous Unit of measure: mm/hg Data value: Change from baseline Drop‐outs Outcome type: Dichotomous Reporting: Fully reported Direction: Lower is better Data value: Endpoint
Identification	Sponsorship source: This study was supported by FONACIT, Caracas, Venezuela, Grant 2005‐000‐384. (The study Thanked to Valmorca Laboratories, Mérida, Venezuela, for providing the placebo pills.) Country: Venezuela Setting: Outpatient centre (Center for the Attention for Schizophrenia Outpatients and their Families) Comments: ‐ Authors name: Trino Baptista Institution: Department of Physiology, Los Andes University Medical School, Email: trinbap@yahoo.com Address: Department of Physiology, Los Andes University Medical School, P.O. Box 93, Mérida, 5101‐A, Venezuela
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Judgement comment: patients were randomly assigned by a computer programme to either MET or placebo.
Allocation concealment (selection bias)	Low risk	Judgement comment: Trino Baptista (TB) was the study coordinator. Only the study coordinator knew the treatment assignment.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Judgement comment: Only the study coordinator (TB) knew the treatment assignment. TB was in coordination with the treating psychiatrists of participants, who controlled their clozapine and other psychotropic drug doses and who periodically informed him. TB according to the intolerances adjusted the MET or placebo dose. But there are no more details on the blinding, that who was/were blinded in this study. Only the study coordinator (TB) knew the treatment assignment.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: Before starting the study, each patient was evaluated by a psychiatrist trained in clinical research and also later assessed the mental state for the follow up visits. "The same psychiatrist assessed the mental state in every visit (baseline, weeks 7 and 14) and maintained contacts by phone with patients and caregivers every two weeks."
Incomplete outcome data (attrition bias) All outcomes	High risk	Drop out 7 in metformin group, 0 in placebo group.
Selective reporting (reporting bias)	Unclear risk	Judgement comment: The study protocol is not provided. but in the Methods section, the outcomes were pre‐specified and later addressed.
Other sources of bias	Low risk	Quote: "Role of the funding source. No intervention in the study accomplishment, results analysis and manuscript preparation."