Consultant evaluation of a hospital medication system: implementation and evaluation of the new system.

The effects of consultant-recommended modifications in a hospital medication system on medication-error rates and response times were evaluated. Fourteen recommendations for improving the medication system in a large teaching hospital were implemented to varying extents over a period of 1.5 years. The response times and medication-error rates of the newly implemented system were then measured using pharmacist-observers as was done in the old medication system. Medication-error rates were then compared on one medical-nursing unit and two surgical-nursing units both before and after implementation of the new system on the medical unit alone. Response times for routine medication orders decreased by 55% compared with the previous system; for "now" and "stat" orders, response times were reduced by 57% for orders filled from the central pharmacy and by 70% for orders filled from medication carts on the nursing units. No significant differences in medication-error rates were found when the old and new medication systems were compared. This finding was attributed primarily to the hospital's failure to implement recommendations related to computer printing and sorting of orders, use of dispensing envelopes for delivering medications in true unit dose form, and packaging of all medications in unit dose form. This study illustrates how the benefits of a unit dose system can be compromised in implementation and the need for close monitoring to ensure that performance standards are maintained.