As they become more prevalent, electronic medical records (EMRs) are being used to improve safety in increasingly creative ways. Recent studies have shown that EMRs can be used to detect diagnostic errors and postoperative complications with accuracy. In this study, the investigators developed an automated method for detecting unapproved abbreviations (UAAs) within clinicians' notes, measured the incidence of UAAs over time, and fed back data to individual clinicians on their use of UAAs. This system resulted in a significant reduction in the use of UAAs over the 6-month study period. Since using UAAs is common and has been linked to serious adverse events, this study demonstrates another potential use of EMRs to improve patient safety.