Why People Don’t Change Their Minds Despite The Data

Has this happened to you?

Your (emergency physician / neurosurgeon / orthopaedic surgeon) colleague wants to (get rib detail xrays / administer steroids / wait a few days before doing a femur ORIF). You question it based on your interpretation of the literature. You even provide a stack of papers to them to prove your point. Do they buy it? Even in the presence of randomized, double-blinded, placebo-controlled studies with thousands of patients (good luck finding those)?

The answer is generally NO! Why not? It’s science. It’s objective data. WTF?

Sociologists and psychologists have shown that there is a concept that they call the Backfire Effect. Essentially, once you come to believe something, you do your best to protect it from harm. You become more skeptical of facts that refute your beliefs, and less skeptical of the items that support them. Having one’s beliefs challenged, even with objective and authoritative data, causes us to hold them even more deeply. There are plenty of examples of this in everyday life. The absence of weapons of mass destruction in Iraq. The number of shooters in the JFK assassination. President Obama’s citizenship.

Bottom line: It’s human nature to try to pick apart a scientific article that challenges your biases, looking for every possible fault. It’s the Backfire Effect. Be aware of this built in flaw (protective mechanism?) in our psyche. And always ask yourself, “what if?” Look at the issue through the eyes of someone not familiar with the concepts. If someone challenges your beliefs, welcome it! Be skeptical of both them AND yourself. You might just learn something new!

The Trauma Professional’s Blog Is 8 Years Old!

Hey everyone! This blog turned 8 years old this week!

Stay tuned for some very interesting stuff this year. The Trauma MedEd newsletter will start cranking out regularly again later this month. I’m considering broadening the social media presence. There will be more stuff on YouTube. And lastly, could there be a book in the future? Stay tuned, and as always, your support and comments are welcome!

Michael

EAST 2018 #11: Prehospital Cervical Spine Clearance

More and more often, I am receiving trauma activation patients after blunt trauma with no cervical collar in place. Up until a year ago or so, literally everyone with even a hint of blunt trauma had one in place. Now, it is becoming a rarity. It seems that there has been a shift in the philosophy and practice of prehospital providers and the guidelines they follow. 

The group at SUNY Stony Brook reviewed their experience with prehospital spine clearance (meaning non-placement of a collar by EMS) over a 6 year period. They analyzed trends in prehospital spine immobilization during this period.

Here are the factoids:

  • Over 5,000 patients were analyzed, and the incidence of cervical spine injury remained constant at 9% over the study period
  • Placement of prehospital cervical immobilization decreased from 54% to 35%
  • The incidence of spine injury in patients without immobilization  increased from 4% to 6%
  • Of those without immobilization, 15% had a major spine injury (AIS > 3), and 19% had multisystem injuries
  • Factors significantly associated with “inappropriate” prehospital clearance included fall mechanism, elderly, functional dependence, dementia, and presence of comorbidities

Bottom line: This study is intriguing, but I worry that the study population is a bit too small to draw the best conclusions. I say this because the incidence of cervical injury is significantly higher in this study that in a larger one with 34,000 patients. This may indicate either a small sample size or some type of sample bias. I’m unclear about what data the prehospital agencies used to relax the immobilization criteria, and whether or not the criteria are being applied appropriately. It does appear, however, that the elderly are at higher risk for having an injury and not being immobilized.

Here are some questions for the authors to consider before their presentation:

  • How did you define cervical injury, and why is the incidence in your study so much higher?
  • Do the prehospital agencies delivering patients to your center utilize the same clearance guidelines?
  • Big picture question: What should we do to make sure that cervical immobilization is applied appropriately?

Reference: EAST 2018 Podium abstract #34.

EAST 2018 #10: Fresh Whole Blood And Survival

Decades ago, our blood bank system began disassembling units of donated blood, ushering in the era of component therapy. Now, it seems, we are seeing the light and starting to re-look at the concept of using fresh whole blood. To see the difference between fresh whole blood and “rebuilt” whole blood from components, read this post.

The military has a keen interest in studying the practice of using whole blood, since combat locations have a considerable number of “walking blood banks” (i.e. soldiers) . An abstract being presented tomorrow at EAST was submitted by the US Army Institute of Surgical Research. They performed a straightforward study looking at mortality in combat casualties, comparing troops who received fresh whole blood (FWB) to those who received component therapy (kind of). They used regression analysis to try to identify and control for other variables, and also analyzed a subgroup who required massive transfusion.

Here are the factoids:

  • A total of 215 soldiers received FWB, and 896 did not. Of note, the non-FWB patients did not necessarily receive platelets.
  • Overall, survival was similar in both groups at about 94%
  • After controlling for physiologic injury severity and blood product/crystalloid volumes, the risk of death was twice as high in the group that did not receive FWB
  • Survival was higher in FWB patients who underwent massive transfusion (89% vs 80%), although this was only marginally significant

Bottom line: I see this an an interesting but preliminary study, with many unanswered questions. It’s not really a comparison of patients receiving fresh whole blood vs component therapy, because not all of the latter patients received platelets. It also did not take into account the specific anatomic injury areas, particularly critical ones such as brain injury. But this study should certainly stimulate some better designed projects for followup.

Here are some questions for the authors to consider before their presentation:

  • Did you do a power analysis to estimate how many patients would need to be enrolled to discover a real difference? If so, how many?
  • Have you performed a subanalysis on patients in the non-FWB group who received platelets? This would then be a comparison of FWB vs component therapy.
  • Any idea of the age of the components given vs the day 0 FWB?
  • Be sure to show and interpret your significance testing in the presentation

Reference: EAST 2018 Podium paper #15.

EAST 2018 #9: Occupational Exposure During ED Thoracotomy

ED thoracotomy is performed infrequently, under high stress circumstances, and with high stakes for the victim. Thus, it is a setup for mayhem. If not conducted properly, it can be noisy, disorganized, and dangerous due to the possibility of blood exposure. Unfortunately, we don’t know where these trauma patients have been. Previous data shows that the incidence of HIV, hepatitis, and other infectious agents is low but significant.

Occupational exposure of healthcare providers to these infectious agents via needlestick/cut, mucus membrane, open wound, or eyes can happen during any surgical procedure. But the possibility during the less controlled ED thoracotomy would seem to be greater. So the group at the University of Pennsylvania decided to perform a prospective, observational study at 16 trauma centers over a 2 year period. A total of 1360 participants were surveyed who were involved in 305 ED thoracotomies. They analyzed the data for risk of occupational exposure.

Here are the factoids:

  • Mechanism was 68% gunshot, 57% were undergoing prehospital CPR, and 37% arrived with signs of life
  • 22 exposures were documented, or a rate of 7% per thoracotomy and 1% per participant
  • There was no difference between Level I and II centers or hours worked at time of procedure
  • Those with exposures were typically trainees (68%) who sustained a percutaneous injury (86%) during the actual procedure (73%)
  • Full personal protective precautions were only utilized by 46% of exposed providers (!!)
  • Each additional piece of personal protective equipment reduced the risk of exposure by 32%

Bottom line: The authors concluded that the incidence of exposure to patient blood is the same as for other operative procedures. Hmm. They also state that the fear of occupational exposure should not deter providers from performing thoracotomy.

I certainly agree that one should always follow the accepted indications for performing ED thoracotomy. I’m not so sure about the comparison with non-emergent procedures, since the numbers are fairly low. However, of one thing there is no doubt: wear your personal protective equipment! You never know when you might be exposed!

Here are some questions for the authors to consider before their presentation:

  • What kind of power analysis did you do to ensure that you could draw reasonable comparisons between thoracotomy and non-emergent procedures?
  • Please provide detailed breakdown of how you sliced and diced your numbers in terms of type of provider, hours worked, trainee level, precautions taken, etc
  • I enjoyed this paper and look forward to hearing the details!

Home of the Trauma Professional's Blog