Vishnupriya 2016.

Study characteristics
Methods	Randomisation: randomised, computer‐generated Blinding: open‐label, no other details Duration: 24 weeks Country: no details
Participants	Diagnosis: first episode schizophrenia patients (DSM‐IV) N = 96 Age: 18‐40 years Sex: male and female Setting: no details History: on treatment with risperidone 2 mg twice/d for ≥ 2 months Excluded: unco‐operative and aggressive patients; patients with suicidal tendency; pregnant and lactating women; patients with history of liver disease, renal disease, cardiovascular disease, diabetes mellitus, hypertension, dyslipidaemia, substance abuse,seizure disorder, malignancy; patients with diagnosis other than schizophrenia; patients with mental retardation; patients taking other drugs that may affect body weight (carbamazepine, lithium, and topiramate, antidepressants, valproate and hormone replacement therapy); patients on a special diet and who do exercise for weight loss
Interventions	Group 1: participants were given risperidone 2 mg alone, orally, twice daily after food; N = 48 Group 2: participants were given metformin 500 mg orally, twice daily after food + risperidone; N = 48
Outcomes	Able to use: Primary outcomes Proportion of patients is developing metabolic syndrome at the end of 6 months in both groups Secondary outcomes Weight measures Waist circumference BMI Physiological: laboratory measures Fasting blood glucose TGS
Notes	We contacted study authors to obtain additional data on body weight and laboratory measures that were not included in the published paper. We did not receive a response from them.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants who were initiated on risperidone 2 mg orally twice daily for ≤ 2 months for first‐episode schizophrenia were randomised using computer‐generated table into 2 groups
Allocation concealment (selection bias)	High risk	No allocation concealment mentioned, however, the open‐label nature of the study puts it at high risk.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcome unlikely to be biased by blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	All the participants completed the study, and the results were analysed.
Selective reporting (reporting bias)	Low risk	Study protocol unavailable, but all outcomes indicated in the methods section were reported on.
Other bias	Low risk	No obvious bias