Clements, Casey M.

Violence in the emergency department (ED) remains underreported. Patient factors are often cited as a source of confusion in determining the culpability of perpetrators and whether to proceed with incident reporting. This study’s objective was to determine how ED staff at one academic medical center perceive certain clinical scenarios and how this compares to local law enforcement officers (LEO). An anonymous survey with 4 scenarios was sent to multidisciplinary ED staff at our academic medical center, as well as local LEO and inquired whether respondents considered any of the scenarios to be reportable as a crime. Chi-square analysis was used for comparison. The study was deemed exempt by the Institutional Review Board. A total of 261 ED staff and 77 LEO completed the survey. Both groups were equally likely to believe that a reportable crime occurred in Scenario 1, where a patient with dementia punches a nurse (LEO: 26.0% vs. ED: 31.4%, p = 0.44), and in Scenario 2, where an intoxicated patient spits at a phlebotomist (LEO: 97.4% vs. ED: 95.0%, p = 0.56). However, the two groups differed in Scenario 3, in which a patient with delirium makes verbal threats to a doctor (LEO: 20.8% vs. ED: 42.9%, p < 0.001), and Scenario 4, in which a patient’s parent throws a chair at a medical student (LEO: 66.2% vs. ED: 81.2%, p = 0.009). As health systems seek to improve workplace safety, it is important to consider the barriers to reporting violent incidents, including staff’s understanding of what acts may constitute reportable violence, as well as LEO understanding of the unique ED environment and patient responsibilities.

Abstract Background Several high-sensitivity cardiac troponin (cTn) assays have been cleared by the United States (US) Food and Drug Administration (FDA) for clinical use. Although some of them have been used outside the US for some time, there is limited experience thus far across US sites. Following FDA clearance of the Roche 5th Gen cTnT assay in January 2017, a multicenter US regional healthcare system introduced this more sensitive cTnT assay into clinical practice in September 2018. Purpose To examine cTn testing patterns, incidence of acute myocardial infarction (MI), and resource utilization before and after implementation of the 5th Gen cTnT assay across a large US healthcare system. Methods Using electronic-health records reporting software (Cogito SlicerDicer, EPIC), administrative data was examined to evaluate the transition (4-months before and after implementation) from the 4th to 5th Gen cTnT assays across 16 hospitals but not the major hub hospital of the system that transitioned at a separate time. Adult patient visits (emergency and hospital encounters), cTnT testing, incidence of chest discomfort and acute MI, and resource utilization, including hospital admissions, were examined during the transition period. Results 98,558 adult ED patient visits occurred during an 8-month period across 16 hospitals, including 50,485 and 48,073 patient visits before (5/12/18 – 9/11/18) and after (9/12/18 – 1/11/19) implementation of the 5th Gen cTnT assay respectively. cTnT testing occurred in 20% (range 8.9–34.8%) of all ED visits, with testing performed in the absence of chest pain in 54% (range 37–67%) of cases. Acute MI was diagnosed in 5.6% (n=561) of cases using the 4th Gen cTnT assay as compared to 6.6% (n=608) of cases using the 5th Gen cTnT assay (absolute difference 1.0%, 95% CI: 0.32–1.68, p=0.004). Much of the increase in MI diagnoses were in women (2.2% vs. 2.8%, p=0.008). The proportion of ED patients requiring hospital admission did not differ between pre- and post-implementation periods (24.1% vs. 23.6%, p=0.4); however, among the subset of patients with chest pain, fewer admissions occurred post-implementation (17.9% vs. 15.8%, p=0.006). There was no increase in echocardiography (9.3% vs. 8.4%), coronary angiography (2.9% vs. 3.1%), or cardiac consultations (7.3% vs. 7.3%) utilization post-implementation. Conclusions Cardiac troponin testing occurs in nearly 20% of ED visits in a large multicenter regional healthcare system. About half of all tests are ordered in the absence of chest pain, with a low MI rate observed among the patients undergoing testing. Following implementation of the 5th Gen cTnT assay, testing rates/patterns remained unchanged, with a small but significant increase in MI diagnoses, primarily due to more women being diagnosed with MI. Fewer ED patients with chest pain required admission. Despite using a more sensitive cTn assay, hospital admissions and resource utilization did not increase post-implementation.