The consumption of metformin in Norway is low. Most probably this is caused by poorly based impressions of small effects and frequent and serious side effects. A review of the literature shows that metformin and sulfonylurea lower average blood glucose values equally well, both in obese and non-obese type 2 diabetic patients. In sulfonylurea failure, addition of metformin lowers the glucose values to the same extent as a shift to insulin monotherapy does. Metformin is anti-hyperglycaemic only, and therefore does not cause hypoglycaemia. The risk of drug-related deaths is no higher with metformin than with sulfonylurea. Metformin improves the lipid profile, and has other effects which could lower risk of cardiovascular complications. The place of metformin in the treatment of type 2 diabetic patients should be reconsidered.