Computerisation of a paper-based intravenous insulin protocol reduces errors in a prospective crossover simulated tight glycaemic control study

Background: Paper-based continuous intravenous insulin protocols for tight glycaemic control (TGC) are typically complex, error-prone, time-consuming and burdensome. Little is known about the errors that occur as a result of misinterpretation and whether computerised protocols reduce errors. Objective: To compare the errors resulting from protocol misinterpretation, time required to manage insulin infusions and nursing satisfaction between a computerised insulin protocol and a paper-based protocol. Methods: In a crossover study, 62 ICU nurses completed 10 TGC simulated scenarios for the computerised and paper protocols. Scenarios evaluated three phases of insulin management: initiation, titration and transition. Scenarios response errors, time to completion and user satisfaction were examined. Results: A total of 620 responses were recorded using both protocols. The computerised protocols were associated with higher user satisfaction, as well as: fewer errors in the titration (13 vs. 113 errors, p= .0001) and transition phases (9 vs. 23 errors, p= .001), fewer dosing errors, although not statistically significant (p= .096), in the initiation phase, and less time to complete in the titration phase (6 vs. 9.5. min, p= .0001). Conclusions: In a simulated environment, a computerised protocol for TGC resulted in significant insulin dosing error reduction, saved time and improved nurse satisfaction.