Farmer A, Wade A, Goyder E et al on behalf of the Diabetes Glycaemic Education and Monitoring Trial Group. BMJ online doi:10.1136/bmj.39247.447431.BE
Are we wasting money prescribing blood glucose testing strips? That was the question for this three-arm, open, randomised study in 48 practices in Oxfordshire and South Yorkshire.
The study involved 453 patients with a mean age of 65.7 years, mean duration of diabetes of three years and mean HbA1c of 7.5%. The patients were randomised to three interventions. The control group was standard care with HbA1c monitoring every three months. The less intensive group had blood glucose monitoring with GP support, while the more intensive group self-monitored blood glucose, with additional education and support from a research nurse.
All patients had three-monthly visits and feedback on HbA1c results, tested two weeks before the visit. This gave the doctor an opportunity to adjust therapy in line with NICE guidelines.
The primary outcome, HbA1c at 12 months, was designed to show a 0.5 per cent difference in HbA1c, assuming a 1.25 per cent standard deviation and 10 per cent drop-out rate. The groups were well matched, with only a 12.6 per cent loss to follow-up.
There was no evidence of significant differences in HbA1c between groups, with a reduction of only 0.2 per cent. It was interesting that patients in the less intensive monitoring group were more likely to persist with using meters at least twice a week than the more intensive group (67 per cent versus 52 per cent).
Similar numbers received an increase in hypoglycaemic medication, about 30 per cent in all groups. However, patients in the more intensive groups were more likely to report hypoglycaemic symptoms and more patients in both intensive groups were taking insulin by one year.
The trial can be criticised for its size and problems related to blinding. It is also difficult to separate the benefits of education and self-management from monitoring alone. The lack of benefit may be related to the moderate control at the start and poorly controlled patients may benefit more from monitoring. Overall, this trial does not support the expense of blood glucose monitoring in patients with stable type-2 diabetes and this approach has been adopted in the All Wales Diabetes Guidelines.
- Dr Kathryn Griffith is a GPSI in cardiology, in York.