Objectives: The primary aim of this study was to assess residual beta cell function at diagnosis of type 2 diabetes and identify accessible laboratory markers that best estimate it. The secondary objective was to evaluate the change in beta cell function 6 months after starting different therapeutical regimens. Materials and methods: Forty seven subjects were included in the study and each performed a 75-g oral glucose tolerance test (OGTT) at baseline and after 6 months. Metabolic and immunologic parameters were determined from fasting samples. According to the degree of metabolic decompensation, specific therapy was started: metformin, metformin plus gliclazide or insulin therapy (with/out metformin). Early and total beta cell function was evaluated by the disposition index (DI) calculated for 30 minutes and 120 minutes, respectively. Results: At diagnosis, fasting blood glucose (BG) and HbA1c varied largely (129-521 mg/dl and 5.5-14%, respectively). The DI30 and DI120 decreased with more severe glycemic decompensation. For both DI30 and DI120 significant negative correlations were found for glycemic markers (HbA1c, 2-hour BG and maximal BG amplitude) and positive correlation for 2- hour C peptide (p<0.0001 for all). HbA1c value of 7% discriminated an important decrease of DI30 and DI120. Insulin and combined therapy significantly improved DI120 at 6 months (p: 0.0062 and 0.01, respectively), while DI30 was improved only with insulin therapy (p: 0.0326). Conclusions: Beta cell function at onset correlated with HbA1c, 2-hour BG and C peptide during OGTT. Thus OGTT and HbA1c are pivotal for evaluation of beta cell function. Insulin therapy improved early and total insulin secretion at 6 months.