peer reviewed ; Initiating or intensifying insulin therapy is often considered as a challenge in general practice. The observational prospective Belgian study InsuStar was performed in 2011-2013 among 150 representative general practitioners, who were invited to initiate or intensify insulin therapy when necessary in 523 patients with type 2 diabetes (mean age : 65.5 years; mean HbA1c : 8.8 %). The initiation of insulin therapy (glargine in > 50 %) was justified by insufficient glycaemic control (96 %) and its intensification (replacement of insulin NPH or premixed insulins by insulin glargine, eventually with the addition of a short-acting insulin analogue) aimed at improving glucose control (58 %), avoiding hypoglycaemia (17 %) or both (17 %). After a follow up of 6 ± 1 months, HbA1c level decreased from 8.79% to 7.52% (-1.27 %; 95 % confidence interval : -1.43,-1.11; p<0.001). Overall 27.6 % of patients reached an HbA1c < 7% versus 5.9 % at inclusion (p<0.001), with rather few hypoglycaemia and a high physician confidence level regarding insulin therapy. These results should encourage general practitioners to initiate insulin therapy at an earlier stage and to intensify it when necessary in patients with insufficiently controlled type 2 diabetes.