ABSTRACT
OBJECTIVES: Performance improvement programs in emergency medicine (EM) have evolved beyond peer reviews of referred cases and now encompass a large set of quality metrics that are measured proactively. However, peer review of cases continues to be an important element of performance improvement, and selection of cases tends to be driven by an ad hoc referral process based on concerns about problems with care in the emergency department (ED). In the past decade, there has been widespread hospital adoption of rapid response teams (RRTs) that respond to patients who decline clinically to reduce adverse outcomes. In an effort to cast a wider net, to take a more systematic approach, and to avoid "blind spots" from individual variability in criteria for referring cases, the institution instituted a new process for selecting cases for ED peer review based on RRT activations within 24 hours of admission from the ED. The hypothesis was that a formal process for review of these activation cases would increase the number of cases for peer review. METHODS: This was a prospective, observational study conducted from July 1, 2012, to June 30, 2013, at an urban, academic medical center with an EM residency program. A new automated monthly report was created, capturing all RRT activations within 24 hours of admission from the ED. All events were reviewed by three physicians from the ED performance improvement committee to examine for systems issues, individual provider issues, or both, that might yield opportunities for improvement. Cases with potential opportunities were reviewed by the full ED performance improvement committee. Cases were classified according to the indication for response team activation using the system outlined by the U.S. Agency for Healthcare Research and Quality. RESULTS: During the study period 61,814 patients were treated in the ED, and 13,067 were admitted to inpatient status. Thirty-two RRT activations within 24 hours of admission from the ED occurred among these admitted patients, representing 0.24% of admissions (95% confidence interval [CI] = 0.16% to 0.33%). Of the 32 cases, only one was also referred independently for ED performance improvement review via the traditional ad hoc process. During the same period of time, 85 cases were referred to the ED performance improvement committee via the traditional ad hoc referral process. Thus, the RRT cases added an additional 31 cases, or 36.5%, to the 85 cases reviewed in ED performance improvement. Of the 32 cases, two were determined by the performance improvement committee to have individual provider factors in their ED care, which contributed to the clinical decline triggering the response teams; none had system factors. Most of the response team activations were for neurologic changes (n = 13) and respiratory status changes (n = 12). In two cases there was long-term morbidity or mortality related to the team activation event; in neither of these cases were ED system or individual provider factors judged to have contributed. CONCLUSIONS: The review of RRT activations within 24 hours of admission from the ED significantly supplemented the typical ad hoc referral system for peer review of cases, highlighting cases that likely would not have received attention within the ED. This novel and unique case review process revealed opportunities for education and performance improvement. This and other systematic approaches to case detection may be useful adjuncts to traditional case referrals for review.
