Prospective, open-label, sequential period, pilot study of 86 patients with poorly controlled type 2 diabetes (admission HbA1c ≥ 7.0%) and a blood glucose concentration > 10mmol/L requiring admission to the Intensive Care Unit (ICU) and administration of insulin. The 'control' patients (n = 53) were consecutively admitted during a 6 month period and the 'intervention' patients (n = 33) were admitted during a subsequent 6 month period. During the 'control' period blood glucose was targeted between 6-10mmol/l, whereas during the 'intervention' period the target was 10-14mmol/l. Time weighted mean glucose was calculated and recorded blood glucose concentrations < 4.0mmol/l considered as an hypoglycaemic episode for each patient. Data are mean (SE) or median [IQR] and analysed using independent samples t-test, Mann-Whitney test, Chi-squared test, Linear and Logistic regression as appropriate.