Sources of glucose variability in insulin-treated type 2 diabetes: the Diabetes Outcomes in Veterans Study (DOVES)

OBJECTIVE Glucose variability can be a significant barrier to glycaemic control for diabetic patients on insulin. The study identified clinical and behavioural factors associated with glucose variability in type 2 diabetes. DESIGN Observational study. SUBJECTS Randomly selected veterans with stable, insulin-treated type 2 diabetes. MEASUREMENTS Baseline data included demographics, clinical characteristics, diet, activity and haemoglobin A1c. Subjects self-monitored blood glucose before all meals and at bedtime for 8 weeks. Monitoring data were used to derive an overall glucose mean and glucose coefficient of variation (100*SD/mean) for each subject. Univariate and multivariate analyses were used to evaluate the relationship between glucose coefficient of variation and clinical factors. RESULTS Two-hundred and four subjects completed the protocol. The 8-week mean (+/- SD) glucose was 9.9 +/- 2.2 mmol/l, and the average overall coefficient of variation was 35.6 +/- 7.7%. Univariate analysis showed a higher variation in older subjects and in those with longer duration of insulin treatment, greater consumption of sugars and greater confidence in their self-care abilities. Variation was lower in subjects who were obese, more compliant with medications and receiving larger insulin doses. Multivariate analysis showed that treatment duration, sugar consumption, medication compliance and insulin dose were independently associated with glucose variation. Medication compliance was more influential for fasting variation, while body mass index was more influential for before lunch variation. CONCLUSIONS Numerous behavioural, clinical and treatment factors were independently associated with glucose variability. Reducing sugar consumption and improving medication compliance might offset the greater glucose fluctuation observed in lean patients with longstanding diabetes.