Eight Percent of Housestaff Unable to Diagnose Tombstone MI on EKG

Here's a frightening paper. It says eight percent of internal medicine and emergency medicine housestaff could not diagnose a straightforward “tombstone” MI on an EKG. Six percent couldn't diagnose ventricular tachycardia, and a whopping 58 percent couldn't diagnose complete heart block. In these latter cases, about a quarter of the residents were sure they were correct.FigureTo be fair, 30 percent of the residents were PG-1s and 35 percent were PG-2s, but this is still a very distressing report. To make it worse, at least for the emergency medicine residents, they didn't seem to do significantly better as they progressed through their training. Given the fact that EKGs are an integral part of the day-to-day practice of emergency medicine, it would appear that better efforts at more formal training in EKG interpretation should be mandatory, and this is supported by the fact that most of the residents felt their EKG training was inadequate. Competency in Electrocardiogram Interpretation Among Internal Medicine and Emergency Medicine Residents Berger JS, et al, Am J Med 2005;118(8):873 BACKGROUND: The 2001 guidelines of the American College of Cardiology/American Heart Association cite supervised interpretation of 500 EKGs as a goal for achieving competency, but current guidelines are not evidence-based. METHODS: The authors from Beth Israel Medical Center in New York examined the interpretation of 12 representative EKGs by 87 internal medicine residents and 33 emergency medicine residents at two large teaching hospitals (30% PG-1, 35% PG-2, 30% PG-3, 5% PG-4). RESULTS: Median scores for self-rated proficiency in EKG interpretation, on a scale of 0–10, were 6.0 for the internal medicine residents and 5.0 for the emergency medicine residents. Median scores for certainty in the interpretation of the 12 study EKGs on a scale of 0–48, were 35.0 for the internal medicine residents and 35.5 for the emergency medicine residents. More than half the participants (58%) felt that their EKG training was inadequate. The overall accuracy in interpretation of the 12 EKGs was 62.5 percent for the internal medicine residents and 58.3% for the emergency medicine residents. Three EKGs were representative of cardiac emergencies. Rates of incorrect interpretation were eight percent for myocardial infarction, six percent for ventricular tachycardia, and 58 percent for complete heart block. (For heart block, 35 subjects who noted 100 percent confidence in their diagnoses were incorrect.) There was a statistically significant positive correlation between accuracy of EKG interpretation and postgraduate year for the internal medicine group but not the emergency medicine group. Factors that were predictive of EKG competency included career interest in cardiology and level of training. CONCLUSIONS: These findings demonstrate a low level of competency in EKG interpretation among internal medicine and emergency medicine residents. Crashes and Cell Phones I think that motor vehicle crashes that occur under the influence of mobile phone usage should result in a fine that is four times the usual one. Why? The following study indicates that the risk of a crash resulting in an ED visit is four times higher if a mobile phone has been used. I must admit that sometimes I make calls while I'm driving. I also must admit that I'm not as attentive to driving when I'm on the phone. Sometimes I have little recollection for the trip when on the phone, like I was driving on autopilot. I've resolved for the new year that I will not make nonessential calls while driving. I think if most people were honest, they too would acknowledge diminished capacity when under the influence of cell phones. And you can be sure that if I get hit by someone who was driving and talking, I will be angry. This paper makes clear that this is not about “hands-free” vs. holding the phone; it is about engaging in any phone conversation. Role of Mobile Phones in Motor Vehicle Crashes Resulting in Hospital Attendance: A Case-Crossover Study McEvoy SP, et al, Brit Med J, 2005;331:428 BACKGROUND: Studies have found that the use of mobile phones impairs driving performance. METHODS: This Australian study examined the relationship between mobile phone use and motor vehicle crashes resulting in hospital attendance. Interviews were conducted with 456 drivers 17 or older who presented to an ED following a motor vehicle crash and who reported owning or using a mobile phone and consented to analysis of his mobile phone records. Mobile phone use during the 10 minutes prior to the motor vehicle crash was compared with use on self-reported driving trips at the same time of day 24 hours, 72 hours, and one week prior to the crash. Information was available for all 456 crashes and 801 “control” episodes. RESULTS: Nearly all of the subjects (93%) sustained at least one injury (predominantly mild to moderate in severity). Just more than half the subjects (52%) reported that they had some type of hands-free mobile phone device in their vehicles, and nearly all of these individuals who reported using a phone while driving stated that they used a hands-free device at least some of the time. According to mobile phone records, mobile phones were in use up to 10 minutes prior in nine percent of the motor vehicle crashes compared with three percent of the corresponding control intervals. The odds ratio [OR] for a crash during mobile phone use was 4.1, and was not statistically influenced by the age or gender of the subject or the type of mobile phone used (ORs for a crash: 4.9 with use of a handheld phone and 3.8 with use of a hands-free phone). CONCLUSIONS: Use of a mobile phone while driving was associated with a fourfold increase in the risk of a motor vehicle crash. This risk was not statistically reduced by ownership of a hands-free device. P-Values and Clinical Significance Finally, a paper you would never expect to see in an orthopedic journal. It comes to the same conclusion that a growing number of other authors have: Clinical studies should abandon the use of p-values because they too easily lead readers to conclude that statistically significant results are, in fact, clinically significant when this is frequently not the case. The Undue Influence of Significant P-Values on the Perceived Importance of Study Results Bhandari M, et al, Acta Orthopaed, 2005;76(3):291 BACKGROUND: It is often (and sometimes erroneously) believed that study results that achieve statistical significance have a comparable clinical significance or produce effects that are important to the patient. METHODS: In this study, 12 orthopedists (three residents, five fellows, and four attendings) at St. Michael's Hospital in Toronto were presented with the results of 40 comparative clinical trials with and without reporting p-values for the primary outcomes. Participants were blinded to the studies' authors and affiliations, and were presented with questionnaires eight weeks apart with some variation in the material presented to decrease the possibility of recall. Participants were asked to rate the clinical importance of the study results with final scoring that ranged between 1 (definitely/probably unimportant) to 3 (definitely/probably important). RESULTS: For studies in which actual p-values were statistically significant (p<0.05), the average clinical importance score assigned by the participants was 1.9 when p-values were not presented compared with 2.4 when p-values were presented (10 of 12 reviewers assigned higher clinical importance scores when presented with statistically significant p-values). For studies in which results did not achieve statistical significance, the average clinical importance scores assigned by the participants when p-values were available or unavailable were 1.5 and 1.6, respectively. CONCLUSIONS: The provision of statistically significant p-values to reviewers was associated with the perception of increased clinical importance of the results of comparative studies. The authors suggest that reliance on p-values might interfere with clinical judgment and feel that use of p-values in the medical literature should be abandoned.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.