Does rapid transition to insulin therapy in subjects with newly diagnosed type 2 diabetes mellitus benefit glycaemic control and diabetes-related complications? A German population-based study.

OBJECTIVE The aim of this study was to assess whether earlier transition to insulin in subjects with newly diagnosed type 2 diabetes improves glycaemic control and reduces diabetes-related complications. METHODS Subjects with newly diagnosed type 2 diabetes, and 2 or more recorded glycosylated haemoglobin (HbA1c) values, were identified from the Mediplus Germany database between June 1993 and May 2001. Subjects were stratified by treatment group: diet and exercise, sulfonylurea drugs, antihyperglycaemic drugs, insulin, or insulin plus sulfonylurea or antihyperglycaemic drugs. Treatment modifications were tracked over time and a rapid transition to insulin recorded if insulin was the initial therapy administered or the immediate treatment after diet and exercise. The area under the curve (AUC) for HbA1c was calculated and a linear regression model used to explain AUC as a function of rapid transitioning to insulin. A Cox proportional hazard model assessed the relationship between the time to first complication and AUC, rapid transition to insulin, and the number of treatment modifications. RESULTS Of the 3136 subjects who met the study entry criteria, just 151 (4.8%) were initiated on insulin; after 5 years only 811 (25.9%) subjects had received insulin therapy. In the regression model explaining AUC, rapid transition to insulin significantly improved glycaemic control (-0.20, p=0.03). The Cox proportional hazard model demonstrated that the time to first complication was negatively related to AUC (-0.05, p<0.01) and the rapid use of insulin (-0.27, p<0.01), and positively related to the number of treatment modifications (0.07, p<0.01). CONCLUSIONS Using actual real world clinical practice data, the present study found that the immediate use of insulin in patients with type 2 diabetes improved blood glucose control as measured by the AUC for HbA1c readings. This, in turn, reduced the risk of diabetes-related complications. In contrast, we observed that a stepwise transition treatment pattern (switching from diet and exercise to sulfonylureas and then perhaps to antihyperglycaemic agents and finally insulin) increased the risk of diabetes-related complications. Greater effort is required to remove some of the barriers currently preventing earlier initiation of insulin therapy in patients with type 2 diabetes.