Aims/hypothesis The primary aim of this completed multicentre randomised, parallel, double-blind placebo-controlled study was to elucidate the mechanisms of glucose-lowering with colesevelam and secondarily to investigate its effects on lipid metabolism (hepatic de novo lipogenesis, cholesterol and bile acid synthesis). Methods Participants with type 2 diabetes (HbA 1c 6.7-10.0% [50-86 mmol/mol], fasting glucose <16.7 mmol/l, fasting triacylglycerols <3.9 mmol/l and LDL-cholesterol >1.55 mmol/l) treated with diet and exercise, sulfonylurea, metformin or a combination thereof, were randomised by a central coordinator to either 3.75 g/day colesevelam (n=30) or placebo (n=30) for 12 weeks at three clinical sites in the USA. The primary measure was the change from baseline in glucose kinetics with colesevelam compared to placebo treatment. Fasting and postprandial glucose, lipid and bile acid pathways were measured at baseline and post-treatment using stable isotope techniques. Plasma glucose, insulin, total glucagon-like peptide-1 (GLP-1), total glucose-dependent insulinotropic polypeptide (GIP), glucagon and fibroblast growth factor-19 (FGF-19) concentrations were measured during the fasting state and following a meal tolerance test. Data was collected by people blinded to treatment. Colesevelam increased cholesterol and bile acid synthesis and decreased FGF-19 concentrations. However, no effect was seen on fractional hepatic de novo lipogenesis. Conclusions/interpretation Colesevelam, a non-absorbed bile acid sequestrant, increased circulating incretins and improved tissue glucose metabolism in both the fasting and postprandial states in a manner different from other approved oral agents.Trial registration: ClinicalTrials.gov NCT00596427 Funding: The study was funded by Daiichi Sankyo.