Evaluation of treatment adherence in type 1 diabetes: a novel approach

Abstract

Intensified insulin therapy requires outstanding compliance but no measure of therapy adherence has been agreed upon. The aim of the current study was to test the hypothesis that treatment adherence, as described by a novel multiple regression model, relates to glycosylated haemoglobin and hypoglycaemia frequency in type 1 diabetes. Furthermore, we sought to analyse the complex diurnal patterns of therapy adherence. Thirty type 1 diabetes patients (20 females and 10 males), treated with intensified insulin therapy, were studied in a retrospective manner. Patients were trained to follow treatment algorithms for adjusting regular insulin dosage which took into account the actual blood glucose, food intake and the time of the day. By means of multiple linear regression analysis, with regular insulin dosage as the dependent variable, blood glucose and food intake as the independent variables, the insulin treatment algorithms actually used by the individual patient were retrieved. The correlation between prescribed and implemented insulin therapy served as a measure of adherence. Metabolic control was assessed by glycosylated haemoglobin and hypoglycaemia frequency. Median glycosylated haemoglobin was 7.7% (range: 6.3-10.8); median monthly hypoglycaemia frequency was 3.8 (range: 0-9.8). Patients with good metabolic control (glycosylated haemoglobin < 7.7 and/or hypoglycaemia frequency < 3.8 per month) adhered to prescribed insulin dosing algorithms more frequently than those with poor metabolic control. In patients with type 1 diabetes on intensified therapy a positive relationship between adherence to the therapy prescribed and metabolic control exists.