Search

Introduction: While emergency physicians initially resuscitate patients arriving at a hospital, resident physicians attend the majority of in-hospital resuscitations. The objective of this study was to perform a formal needs assessment establishing essential in-hospital resuscitation knowledge and skill sets for resident physicians. Methods: All cardiologists, intensivists, and internal medicine attending staff at 4 teaching hospitals were electronically surveyed using a modified Dillman method in a blinded fashion. A broad list of knowledge and skill sets was gathered from recent resuscitation guidelines. Using a 6-point Likert scale, respondents ranked items in terms of importance for in-hospital resuscitation. Responses were collated using descriptive statistics, including median and interquartile ranges for each domain surveyed. Institutional ethics approval was granted. Results: The response rate was 75% (n = 93) with the majority (52%) of respondents internal medicine attendings. The top 5 resuscitation knowledge sets (in order) were: cardiac rhythm assessment, discussion of code status, delivery of bad news, management of wide complex tachycardia, and management of bradycardia. The top 5 resuscitation skills (in order) were: cardiac defibrillation, airway assessment, bag-mask ventilation, central venous access, and synchronized cardioversion. The bottom 5 resuscitation knowledge sets (in order) were: whole bowel irrigation for GI decontamination, neuromuscular blockade for intubation, ventilator management, charcoal for GI decontamination, and procedural sedation. The bottom 5 resuscitation skills (in order) were: open cardiac massage, cricothyrotomy, compartment pressure measurement, tube thoracostomy, and gastric lavage. Conclusion: This study has prioritized in-hospital resuscitation knowledge and skill sets. Resuscitation education for resident physicians should be tailored by these findings, particularly as they attend the majority of in-hospital resuscitations in teaching institutions. Keywords: resuscitation, needs assessment, educational research

Introduction: While emergency physicians initially resuscitate patients arriving at a hospital, resident physicians attend the majority of in-hospital resuscitations. The objective of this study was to evaluate the effectiveness of an in-hospital resuscitation course for resident physicians. Methods: A 2-day, low-fidelity simulation, case-based curriculum was designed based on a needs assessment involving cardiologists, intensivists and internal medicine attendings. A pilot version was modified based on participant feedback. Course participants were electronically surveyed in a blinded fashion 1 month pre- and 1 month postcourse using a modified Dillman methodology. Responses were collated using descriptive statistics, including median and interquartile ranges for each domain surveyed. Institutional ethics approval was granted. Results: The response rate was 93% (n = 27) pre- and 85% (n = 23) postcourse. Precourse only 24% agreed that residents received adequate in-hospital resuscitation training, while 28% felt prepared to lead a resuscitation. Only 4% agreed that attending physicians supervised in-hospital resuscitation. Postcourse 45% of participants self-reported using course knowledge and skills during an in-hospital resuscitation. Significant self reported changes in median confidence scores pre- to postcourse included: management of bradycardia (p < 0.01); management of narrow complex tachycardia (p = 0.02); management of septic shock (0.03); management of anaphylactic shock (p < 0.01); management of status epilepticus (p = 0.01); defibrillation (p = 0.02); electrical cardioversion (p < 0.01); and central venous access (p = 0.02). Conclusion: This study suggests an educational need for resident physicians. Precourse respondents felt unprepared and unsupervised to lead an in-hospital resuscitation. A low-fidelity simulation course improves self-reported confidence in resuscitation knowledge and skills. Keywords: inhospital resuscitation, low-fidelity simulation, curriculum evaluation

Introduction: The objective was to determine the effect of cognitive forcing strategies (CFS) training to decrease diagnostic error among senior medical students during an emergency medicine rotation. Methods: This was a prospective, cross-over study with consecutive enrollment. Ethics approval was granted. Fifty-six subjects were exposed to an interactive case-based teaching session on CFS. Subjects were evaluated on 4 satisficing bias computer-based cases (1 near-transfer, 1 far-transfer and 2 'false positive' cases with a single diagnosis) and 2 availability bias cases (near- & far-transfer). Thirty-seven subjects were immediately tested. Nine subjects were tested 1 week later. Data was analyzed using descriptive statistics and a McNemar [chi square] test. Near- v. far-transfer was defined based on the type of diagnostic test. If CFS teaching used an ECG as an example, ECG testing was near-transfer and x-ray testing was fartransfer, and vice versa. Results: Satisficing bias--On immediate testing right after instruction, only 64% (near) and 55% (far) of subjects searched for a second diagnosis. The difference between nearand far-transfer was not statistically significant (NS). In the 'false positive' cases, a second diagnosis that was not present was identified in 43% of cases. Availability bias--Only 30% (near) and 17% (far) of subjects made the uncommon correct diagnosis. The difference between near- and far-transfer was NS. Delay in testing--When tested 1 week later, only 30% of subjects searched for a second diagnosis in the satisficing bias cases. In the availability bias cases, only 5.5% made the uncommon correct diagnosis. Conclusion: This preliminary data suggests that small group teaching to decrease cognitive error is ineffective. Even with immediate testing, cognitive forcing strategies are not employed by a significant proportion of students, and this worsens with delayed testing. Larger studies are required to determine if the identified trends are statistically significant. Keywords: diagnostic errors, cognitive forcing strategies, education research

Introduction: This study aimed to review the published reports of CT interpretation error rates by radiology trainees and their clinical consequences on patient care. The outcomes of interest were the overall interpretation error rates, the potential to change patient care management, actual adverse clinical outcomes resulting from errors, and stratification of error rate by level of training. Methods: An electronic search of the published literature was conducted using the PubMed, CINAHL and OVID databases. Abstracts, article reference lists and meetings proceedings were also manually searched. Relevant data was abstracted for outcomes above. Results: Eleven published studies met the inclusion criteria for this review, representing 44 527 patient scans (58% were CT scans of the head). There were 1499 major interpretation errors (3.4%, 95% CI 3.2%-3.6%) that had the potential to change patient management. Of the studies that reported patient management changes, there were 94 changes (0.3%, 95% CI 0.2%-0.4%). There were 2 adverse patient outcomes reported (0.007%). Senior residents (PGY 3,4,5) made more errors (14.3%, 95% CI 11.7%-16.9%) than did their junior (PGY1,2) counterparts (2.3%, 95% CI 2.1%-2.5%). The number- needed-to-harm for adverse outcome was NNH = 14 285. Conclusion: This systematic review suggests that although radiology trainees make errors in interpreting CT scans, the subsequent rate of patient management change and adverse clinical outcome is low. The error rate was higher among senior radiology trainees compared to junior trainees, for reasons not reported. The included studies are limited by variable reporting rates for all outcomes of interest for this study. These results could have significant patient care implications for emergency medicine and other clinical practitioners in academic teaching centres. Keywords: interpretation errors, CT scan, systematic review