Aim: To determine whether gastrointestinal (GI) tolerability of metformin monotherapy varies according to baseline BMI or at doses >1500 mg/day in patients newly diagnosed with type 2 diabetes. Methods: We performed a sub-analysis of the safety population from a prospective, multicenter, Phase IV open-label study in which 371 Chinese patients with type 2 diabetes received extended-release metformin monotherapy for 16 weeks. The incidence, severity and duration of GI adverse events (AEs) were compared between normal-weight (BMI <25 kg/m(2), n = 155) and overweight/obese (BMI >= 25 kg/m2, n = 216) patients. The primary objective was to determine whether baseline BMI affect the incidence, severity and duration of GI AEs, using Fisher's exact test and Student's t-test. Secondary objectives were to compare these factors according to final metformin dose (<= 1500 mg/day versus 2000 mg/day). Results: The proportion of patients who reported >= 1 GI AE did not differ significantly between BMI groups (25.2% of the normal-weight group versus 21.3% of the overweight/obese group; p = 0.3840). Patients who reported GI AEs in the two BMI groups experienced similar GI AE severity (p = 0.5410), mean duration (p = 0.3572) and duration distribution (p = 0.1347). There was no significant difference in GI AE severity and duration between metformin dosage groups (<= 1500 mg/day versus 2000 mg/day). Conclusions: Newly-diagnosed Chinese type 2 diabetes patients of normal weight are no more likely than overweight/obese patients to suffer from increased incidence rates, severity or duration of GI AEs when treated with first-line extended-release metformin monotherapy. Doses of 2000 mg/day did not increase the severity or duration of GI AEs. (C) 2016 Diabetes India. Published by Elsevier Ltd. All rights reserved.