Trends and influences on use of antidiabetic drugs in England, 1992-2003

To describe trends in prescribing for glycaemic control in diabetes mellitus in England, 1993-2003, and to examine the effects of new clinical evidence or national guidance on these.

Observational study of prescribing adjusted for extrapolated changes in epidemiology of insulin and non-insulin treated diabetes.

Primary care in England, 1992-2003.

Three monthly prescribing data from the Prescription Pricing Authority.

Defined daily doses/patient/quarter of oral hypoglycaemics and of insulin.

There was an increase in the prevalence of both insulin treated (85%) and non-insulin treated (80%) diabetes. There were increases in all drug use, which were largely explained by this: for example, sulphonylurea use increased by 101% in total but only 11% after correction for change in prevalence and population; insulin use increased by 140% in total, but 33% after correction. Neither guidances from the National Institute of Clinical Excellence nor new clinical evidence had any discernible effect on this. An exception was metformin use, which increased by 495% in total, and by 277% after correction. This increase occurred mainly from early 1998 and seemed to coincide with the publication of evidence of the specific benefits of metformin.

There has been a trend towards more aggressive treatment of diabetes. National guidance may have reinforced this but has not altered the trend. Specific evidence around metformin may have been very influential in increasing its use.