From the May 1, 2003 issue

PROBLEM: A pharmacist shared a story that conveys just how error-prone a new medication protocol can be when planning and testing are sidestepped in favor of quick, widespread adoption. A few physicians had obtained an insulin infusion protocol from another hospital. Right away, they began copying the protocol and placing it on patient records whenever an insulin infusion was prescribed. In response, hospital staff quickly reproduced the protocol in their own format, pushed it through the Pharmacy and Therapeutics (P&T) Committee, and encouraged the entire medical staff to use it for all patients who required an insulin infusion. But then the problems began.

According to the protocol, the standard concentration for the insulin infusion was 1 unit per 5 mL. But this proved to be too much fluid for some patients, so physicians sometimes changed the concentration on the protocol to 1 unit per mL while still using the protocol for dose changes. This led to enormous confusion. Another catch with the protocol was that it presented the insulin dose in units per hour, while nurses were accustomed to handwritten orders that prescribed the infusion in mL per hour. Inevitably, nurses sometimes entered the infusion rate as the dose per hour, not the mL per hour, in which case the patient received just 20% of the intended dose. Also, the protocol called for bolus doses and dose adjustments every 2 hours, which often resulted in labile blood sugars. Eventually, many physicians began altering the frequency of bolus doses, or eliminating them, while still using the protocol for dose adjustments. It later was discovered that the insulin infusion protocol from the other hospital had been developed specifically for patients who were receiving tube feedings and TPN.

SAFE PRACTICE RECOMMENDATION: Standardized protocols are invaluable tools to guide safe use of medications, but their adoption requires groundwork to be successful. Before implementing any new protocol, gather an appropriate interdisciplinary team to perform a failure mode and effects analysis (FMEA) to determine potential pitfalls in the protocol and plan their remedies before implementation. In the case above, for example, FMEA would likely have led to many changes, including use of a 1 unit per mL concentration to avoid patient harm from fluid overload or accidental entry of the dose per hour as the rate per hour (dose and rate per hour are the same with a 1 unit per mL concentration). Right up front, the interdisciplinary team also should determine process and outcome measures to help evaluate success with the protocols. Examples include compliance with different parts of the protocol (process measures), fluctuations in glucose levels, and episodes of hyperglycemia and hypoglycemia (both outcome measures).

After approval of the protocol’s concept, ask several enthusiastic physicians and nurses to test it on the next few admissions while gathering information on its ease of use and effectiveness. Have the interdisciplinary team address any problems identified before spreading its use throughout the organization. Of course, staff education also is required. Finally, don’t let this example discourage you from using tools that have been created by other health systems, or from sharing your tools with others. This sharing of tools – really, sharing of innovation – has caught fire in healthcare and is truly making an impact on safety and quality. The wisdom of adopting innovation that has been developed by others should not be in doubt as long as you consistently follow a process as described above, all of which can be accomplished in a short period of time.