Development and implementation of a novel decision support tool on physical restraint use in critically ill adult patients.

Aim To investigate whether a novel decision support tool would effectively minimize physical restraint use in critically ill adult patients. Design A nonequivalent quasi-experimental design combined with a descriptive qualitative approach was used. Methods A Restraint Decision Tree was developed based on a qualitative study that explored the barriers to employ the Restraint Decision Wheel. During the quasi-experimental study, patients admitted to the unit between October 2017 and March 2018 were enrolled as the control group receiving the Restraint Decision Wheel (n = 528), whereas patients between April 2018 and September 2018 were enrolled as the intervention group receiving the Restraint Decision Tree (n = 564). The physical restraint rate, accidental catheter removal rate and nurses' satisfaction were compared. Results The Restraint Decision Tree significantly decreased physical restraint use. No significant difference in the rate of accidental catheter removal was found. Nurses reported increased satisfaction with the restraint decision-making. Conclusions The Restraint Decision Tree could minimize physical restraint use. Physicians' involvement in the restraint decision-making and nurses' competence in delirium assessment may be essential for successful implementation of the Restraint Decision Tree.