Interest in testing ChatGPT as a clinical tool is increasing. This study asked ChatGPT to provide a differential diagnosis, diagnostic testing, final diagnosis, and care management for 36 previously published clinical vignettes. ChatGPT had an overall accuracy of 72%, with the highest level of accuracy at the final diagnosis stage (77%).

… to implement and maintain. This research builds on a 2013 study , describing the quality improvement (QI) …

In this retrospective study, researchers used electronic health record and quality assurance issue (QAI) data to analyze risk factors associated with patient safety events in the emergency department (ED). Multivariable analyses showed several potential risk factors for safety events – including length of time in the ED, which increased the odds of a safety event by 4.5% for each hour spent in the ED.

… J Healthc Risk Manag … This retrospective cohort study based … are defendants. … Myers LC, Sawicki D, Heard L, et al. A description of medical malpractice claims involving advanced practice providers. J Healthc Risk Manag . Epub 2020 May 3. doi: …

… Jt Comm J Qual Patient Saf … This article describes the …

… of emergency medicine … Ann Emerg Med … Misdiagnosis of a neurologic emergency such as stroke can lead to serious morbidity or mortality. Using a large multi-state database, this study examined the … neurological event after ED discharge. Extrapolated to a national level, this translates to over 55,000 patients …

This retrospective cohort study of 16,801 patients examined whether patients in alternative care areas like hallways experienced more safety problems, such as intensive care unit transfer, readmission, and hospital-acquired infections. Investigators found that patients in alternative care areas were younger, had fewer comorbid health conditions, and no increase in risk for adverse safety outcomes. The results suggest that bed management takes into account patient acuity when assigning location to mitigate potential safety concerns.

Having multiple patient records open in the electronic health record increases the potential risk of wrong-patient actions. This randomized trial tested two different electronic health record configurations: one allowed up to four patient records to be open at a time, and the other allowed only one to be open. Among the 3356 clinicians with nearly 4.5 million order sessions, there were no significant differences in wrong-patient orders. However, the investigators noted that clinicians in the multiple records group placed most orders with just one record open. A post hoc analysis determined that the rate of errors increased when orders were placed with multiple records open. A related editorial highlights the tradeoffs between safety and efficiency and argues for examining the context of the two configurations, including throughput and clinician satisfaction. A previous PSNet perspective discussed assessing and improving the safety of electronic health records.

Mobile health care applications are increasingly being developed and marketed to patients for self-care and diagnosis, with little oversight as to their effectiveness or safety. This commentary outlines four key issues that must be addressed to improve the safety of medical applications.

Laboratory errors, such as mislabeling, improper collection, or specimen loss, can lead to delays in diagnosis and the need for repeat procedures or blood draws. In this single institution study, implementing a custom-built specimen collection module led to a significant decrease in the rate of lab specimen collection and handling errors for blood samples drawn by nurses in both the emergency department and inpatient settings.

In this study, better safety climates in emergency departments were associated with more intercepted near misses but not with fewer preventable adverse events.

Emergency department (ED) crowding was strongly associated with an increased risk of preventable medical error, in this study conducted in four Massachusetts EDs.

Adoption of computerized provider order entry (CPOE) remained low in this survey of emergency departments in four states, with only 30% of respondents having implemented CPOE.

… consultations in emergency departments. … Magid DJ, Sullivan AF, Cleary PD, et al. The safety of emergency care systems: Results of a survey of clinicians in 65 US emergency departments.  Ann Emerg Med . 2009;53(6):715-23.e1. doi:10.1016/j.annemergmed.2008.10.007 …

… Med … The authors describe the process used to implement a national epidemiological study of medical error in emergency departments. … Sullivan AF, Camargo CA Jr, Cleary PD, et al. The National … design. Acad Emerg Med. 2007;14(12):1182-1189. doi:10.1197/j.aem.2007.07.014 …