Background: Surgical wound infections can be significantly reduced by appropriate administration of prophylactic antibiotics (ProAbx). The purpose of this study was to determine the effect of interventions, specifically, forced functions, designed to improve administration of ProAbx in a university hospital setting. Study Design: We retrospectively reviewed charts from 1,622 consecutive patients, seen between March 2005 and June 2007, for evaluation of ProAbx administration to determine correct choice, appropriate timing of administration, and appropriate postoperative cessation of antibiotics. Processes designed to improve compliance were devising orders to limit ProAbx choice; requiring those orders in preadmission testing; administering ProAbx in the preparation and hold area, subsequently, at the time of the universal timeout; and standardizing postoperative orders in the computerized physician order entry. Results: Specialty-specific standardized orders improved compliance from 76% to 91% (p < 0.001) concerning ProAbx choice. Regarding timing of preoperative administration, a baseline compliance of 55% improved to: 78% (p < 0.001) on requiring orders in the preadmission testing area; 90% (p < 0.008) on administering ProAbx in the preparation and hold area; and 95% (p < 0.07) when ProAbx were administered at the universal timeout. Standardization of postoperative orders in the computerized physician order entry improved compliance with cessation from 60% to 86%, p < 0.001. Conclusions: Despite hospital-wide education, improving compliance with evidence-based recommendations for ProAbx required processes that "forced" physician behavior, specifically: specialty-specific, preprinted physician orders limiting ProAbx choice; linking administration of preoperative ProAbx administration to the universal timeout; and standardization of the postoperative ProAbx order in the computerized physician order entry. © 2008 American College of Surgeons.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below