Type 2 diabetes is associated with a higher cardiovascular risk and there has been a growing interest in using dietary intervention to improve lipid profile and glucose control. The present work aims at analysing the effects of the enrichment of a normal diet with β-glucan (3·5 g/d) in free-living type 2 diabetic subjects for 2 months, using a palatable soup. This trial was a parallel, placebo-controlled, double-blinded randomised study performed in fifty-three type 2 diabetic subjects. During a 3-week run-in period, subjects daily consumed a ready meal control soup (without β-glucan). For the following 8 weeks, subjects were randomly assigned to consume daily either a control soup or a β-glucan soup. Changes in lipid profile (total cholesterol (TC), HDL- and LDL-cholesterol (HDLc and LDLc), apo B and TAG) and in glucose control (HbA1c and fasting glucose) were measured. There was no significant alteration in lipid profile in the two groups (TC, HDLc, LDLc and apo B). TAG decreased significantly in the β-glucan group compared with the control group ( − 0·12 (sd 0·38) v. 0·12 (sd 0·44) mmol/l, P = 0·03). HbA1c and fasting glucose were not reduced in any group. A single daily ingestion of 3·5 g β-glucan, as required by official dietary recommendations, for 8 weeks did not change the lipid profile and HbA1c in type 2 diabetic subjects. To improve the metabolic profile of type 2 diabetic subjects in the long term, the quantity, the food vectors and the tolerability of β-glucan products may be re-evaluated.