This week: Newbies may now doc around the clock, three studies that show women and minorities don’t get a fair shot, increasing the use of jet-injected lidocaine in pediatric EM, and predicting death is now far more than a guessing game. Join in as our editors discuss the week’s headlines.
Starting July 1, new doctors can start working up to 24 hours per shift, as long as they don’t go beyond the 80 hours per week cap for all residents
Some say fix the handoff problem and give inexperienced docs more rest, while others say shorter shifts don’t necessarily improve care. Original Article by The New York Times.
E. Paul DeKoning, MD, MS: I just knew this would be on the list this week. We are reaping what was sown years ago–shorter isn’t always better, either for patients or for doctors in training. The public will be hot and bothered by the return to the 24 hr cap for interns. The press will see to that: a travesty of epic proportions! They will ignore, however, that the weekly 80hr cap hasn’t changed. The inhumane (yea tortuous!) increase will get the press and clearly it has, since we are talking about it. But, this has been the case for all other levels of training–providers who, in general, carry more patients and more responsibility for critical decisions than interns–the very essence of graduated responsibility. As a program director, I know the value of being part of a care team that sees a patient through more extended aspects of their care. I tend to agree with the ACGME that the lower cap on intern work hours disrupts the care these patients receive as well as the ability for interns to learn from that care. As EMPs we know that handoffs are the most dangerous time for patients (the public SHOULD care about that)–lower work hour caps mean more handoffs. Emergency Medicine Residents, however, have considerably stricter work hour guidelines–that has NOT changed. I love my specialty! Now, if they would just do away with the naps. Or at least figure out a way for me to get one.
Ryan McKennon, DO: Glad to see the return of the 24 hour shift. I learned the most when I was an intern on those shifts. I would probably extend it a few more hours to be able to participate in morning rounds which were also very helpful to talk out what happened through the night.
A trio of JAMA studies show gender and racial bias in physician training
Yale and Central Connecticut state found Asian and Black students less likely to be chosen for honor society. Universities of Chicago and Pennsylvania showed suspicious drop in female resident scores over time, and University of Pittsburgh and UCSF found women less likely to be chosen for the “TED Talks” of medicine. Author Bruce Yee says it’s time for trainers to examine and correct personal biases. Original Article by Forbes.
E. Paul DeKoning, MD, MS: I’m not sure what to do with this article. I get annoyed at a base level when articles start to lecture me and I get that feeling here. I get it that we all view others through the lenses of our experiences, upbringing, preferences, etc. My mom and dad taught me that one. The author makes some fair points on the need to look past outward appearances–something I endeavor daily to do, ESPECIALLY as a program director. I hope others give the same fair shake to me. Take a look at my Crash Cart Mugshot and start listing your biases. I guarantee you’ll be wrong…especially since the picture doesn’t look like me. Now, that’s a profound statement…
Ryan McKennon, DO: Bias exists in society, why would anyone think a subset like medicine would be exempt. This could easily be a study on bias in teaching, law, childcare, or any other profession. Of the three, I think the second study was the most relevant. It looked at female resident evaluations from first to third year. While the female residents scored slightly higher in the beginning of residency, they scored significantly lower at the end of the third year suggesting an implicit bias. Something to look further into.
Nicholas Genes, MD, PhD: I remember those dramatic studies on bias in the field of music, where “blinded” auditions were more likely to lead to the hiring of women musicians, compared to when the recruiters could see the performers. I remember thinking: medicine (particularly my area in academic medicine) surely makes these decisions based on merit and achievement. These studies suggest otherwise. It’s a problem. The lesson, as always, is to be recognize this as much as possible and try to counteract it, one decision at a time. I wish there was more research showing us ways to do that, effectively.
Use JIL to reduce pain during IV placement in children
Quality improvement methods can increase the use of jet-injected lidocaine (JIL) in pediatric emergency departments, with robust results. Original Article by Medscape.
E. Paul DeKoning, MD, MS: Don’t you find it at least a little amusing that the title of the article is “Simple Steps to Reducing Pain During IV Placement in Children” but “the authors did not measure changes in pain perception associated with use of JIL.” They figured “increased use of JIL seems a good proxy for reduced pain, given the evidence of its efficacy shown in other studies. They also did not study the cost implications of using JIL, although its cost efficacy has been demonstrated by other investigators. A third limitation was assuming that the number of JIL devices ordered accurately reflected the use of these devices, when in fact, “we did not have a reliable measure of actual administration,” the authors state. So what did they actually measure and how is it news? Good to know that at least the assumptions they didn’t measure in their setting are generalizable to mine: “Despite these limitations, the “results of this project are likely generalizable to other institutions and clinical settings,” they conclude. These findings are promising, because the adverse effects of “pain in the acute care setting are being increasingly acknowledged, and pain management needs to be urgently addressed.” Right. Next week I expect we will be discussing new measures to combat the epidemic of topical anesthetic use in children as a consequence of responding to the urgent need noted above. BTW, I’ve never heard of JIL. The way the article is written, she sounds very nice. Maybe I can get her to come place IVs in my ED.
Ryan McKennon, DO: Great synopsis Paul. I have no idea what to do with this “robust” data. This was more of a study on if use increased if you made it easier to use. Not sure why we need that study. How did they get away with “installation of easily accessible wall refrigerators so JIL devices could be stored in multiple locations in the ED?” I can’t even stock band-aids unlocked in a room or have cotton balls in the department due to regulations. The best quote though has to be “increased use of JIL seems a good proxy for reduced pain.” Because we use it more, it must result in less pain. Well, no need to study that one further…
Nicholas Genes, MD, PhD: This is a blurb about an implementation study – it’s not meant to demonstrate the efficacy of jet-injected lidocaine because there’s already good evidence for it. The researchers were just trying to study the best way to get staff to change practice. Here’s a short video that shows the tech and summarizes some recent literature (for those of us that hate videos: JIL means pressurized lidocaine is propelled into the skin without a needle, it’s shown to reduce IV pain better than sham saline, and doesn’t distort the tissue so placing an IV is just as easy.
Big data analytics may soon help doctors and patients make better medical decisions based on more accurate prognoses
The same technology used to predict your Amazon purchases and next Netflix choice can predict an individual’s mortality. And there’s a “warmer, fuzzier” side to number crunching that people are overlooking. Original Article by Slate.
E. Paul DeKoning, MD, MS: This is rich: “I think there’s that warmer, fuzzier side of algorithms that we overlook. They don’t miss anything. And that can be very reassuring.” This is frighteningly reminiscent of HAL 9000 in “2001: A Space Odyssey” or Count Tyrone Rugen, the six-fingered man in The Princess Bride: Count Rugen says “Beautiful isn’t it? It took me half a lifetime to invent it. I’m sure you’ve discovered my deep and abiding interest in pain. Presently I’m writing the definitive work on the subject, so I want you to be totally honest with me on how the machine makes you feel. This being our first try, I’ll use the lowest setting … As you know, the concept of the suction pump is centuries old. Really that’s all this is except that instead of sucking water, I’m sucking life. I’ve just sucked one year of your life away. I might one day go as high as five, but I really don’t know what that would do to you. So, let’s just start with what we have. What did this do to you? Tell me. And remember, this is for posterity so be honest. How do you feel?” More proof that The Princess Bride is in the top 3 best movies ever.
Ryan McKennon, DO: Would you want to know they day or week you were going to die? Even if you had, say a year left? I’m not sure I would. I’m all for having better end of life discussion and realistic expectation but even with the best algorithms, all it can tell you is the likelihood of dying at or by a certain time. I think this would be more destructive to the human spirit than helpful.
Nicholas Genes, MD, PhD: I love the phrase “ritualization of optimism” and the stats about doctors over-estimating survival rings true. Predicting mortality within a year sounds like a worthy goal – I think we’d all make better decisions about how (and where) we’d spend that time. I tend to agree with Obermeyer that when the end-of-life conversation happens in the ED, it’s too late. Still, the article misses out that the best mortality predictors are already making their way into the ED and inpatient settings, in the form of advanced clinical decision support. The popups and alerts are still often clunky as hell, and we need to do a lot better refinement (too many false positives and missed cases)… Slowly but surely, however, we’re getting better at identifying (and treating) occult sepsis, delirium, and other killers, thanks to analysis of EHR data and natural-language processing of physician notes. There’s already evidence-based triggers for identifying inpatients who’d likely benefit from a palliative care consult (based on advanced chronic diseases plus low functional status and other factors) – that seems a good deal of the way there, toward “machine prediction of death.”