Abstract
OBJECTIVES: To identify patients with discrepantly high clinic blood glucose concentrations compared with self reported values and to assess whether such patients have errors in self monitoring technique. To determine whether, in patients with good technique, the discrepancy is a transient phenomenon related to clinic attendance. DESIGN: Prospective study of diabetes clinic patients recruited over six months. SETTING: Outpatient diabetes clinic of a teaching hospital. SUBJECTS: 34 consecutive patients with non-insulin dependent diabetes who had had at least two consecutive clinic blood glucose concentrations more than 5 mmol/l higher than the mean self reported concentration. MAIN OUTCOME MEASURES: Assessment of monitoring technique; presence of cognitive or physical impairment; serum fructosamine concentration; home and clinic blood glucose concentrations. RESULTS: 15 of 34 patients had errors in monitoring technique, 12 of whom had cognitive or physical impairment. In the remaining 19, the mean (SD) blood glucose concentrations of capillary and venous samples taken at home (10.2 (0.6) and 12.2 (1.1) mmol/l respectively) were significantly lower than in those taken at the clinic (16.8 (1.6) mmol/l, p < 0.0002). The fructosamine concentration was significantly higher in patients with monitoring errors than those without (2.4 (0.4) v 1.8 (0.4) mmol/l, p < 0.0001). CONCLUSIONS: "White coat" hyperglycaemia was detected in about half the patients but errors in technique accounted for the rest of the discrepancies. Patients' ability should be assessed before teaching self monitoring and the technique checked regularly.