Creating a diabetes foot reminder-based registry using the electronic medical record.

Objective We created a new diabetes foot examination clinical reminder to directly populate a foot risk registry and examined its accuracy versus administrative data. Methods A pre- and post-test design assessed accuracy of coding foot risk and clinician acceptability. The intervention hospital's reminder was replaced with a dialogue tick box containing the International Diabetic Foot Classification System to populate risk using health factors. Results There were no hospital agreement differences for each foot condition except diabetes and peripheral neuropathy, demonstrating higher agreement at the intervention hospital. There were no differences in service agreement adherence or consulting rates although both demonstrated significantly lower consulting rates at study end. The intervention hospital had a significantly lower patient cancellation rate (1% v. 5%, P=0.01) and better coding for grade 3 patients. The new reminder demonstrated high acceptability. Conclusions The registry system resulted in improved discrimination of the highest foot risk. Further testing is recommended.