War Games!

One of the problems with medical education is that while you are intellectually trained to deal with medical problems and emergencies, actual experience with how to respond to emergent clinical situations is difficult to teach and usually only comes with experience. Further, real clinical experts make medical decisions almost by reflex. You see this in medical school that while you as a medical student have to actively think about what is going on in any given situation, medical experts act more by pattern recognition and have an instant reflexive response to clinical situations. And how do you teach reflexes?

Here at UVA, Jeff Young, a trauma surgeon and researcher in clinical decision making has published on a new strategy of assessing and improving the response of doctors in training to high-risk medical situations. His strategy is rather than stressing cognitive experience, which much of medical school and resident training emphasizes, the goal is to build reflexive responses to critical situations. In an emergency, the ability to generate differential diagnoses and recall complex information is secondary to knowing how to acutely assess patients, resuscitate and stabilize them. Clinical experts do this without even thinking about it. Young’s goal is to train medical students, interns and residents by simulation of critical care situations so that when they end up involved in charge of a critical patient they will reflexively perform the correct actions to resuscitate and stabilize patients. After all, practice makes perfect.

The result is what Dr. Young calls “War Games” – simulations in which students and residents are drilled in their responses to medical emergencies. By putting students under some stress and making them think fast about critical care, reflexive responses to emergent situations are drilled into the subjects, and hopefully when the situations are encountered in real life they’ll know what to do without even thinking about it.

So enough talking about it. Here’s what one looks like – me being drilled by the chief resident on a patient presenting with hypotension.

You notice that rather than going for diagnosis the goal is to start with the basics. First you evaluate the airway, breathing, and circulatory status, resuscitate the patient as necessary, gain IV access, get basic vitals and check tests. Only after you’ve stabilized a patient should you start thinking about what the exact diagnosis is, whether you need to operate etc. It also emphasizes things you don’t necessarily learn in class, like the need to call the attending when some disaster has occurred. It seems like things like this should be obvious (they probably are to EMTs and paramedics), but the reality is that these kinds of practical skills are difficult to relate in a classroom setting. You also quickly realize that when you are under pressure, it’s completely different from all those sessions you remember from 2nd year where you sat around thinking about differential diagnosis with 5 other people in the room. I clearly screw up a few times during the simulation, but hey, that’s why I’m in training and why I appreciate these sessions.

This also demonstrates something I think we can appreciate about evidence-based medicine. Not only do we emphasize a scientific basis for the treatments we use, but we also actively use science to figure out the best ways to train doctors to be better clinicians. I found this strategy to be incredibly useful, and I hope other medical schools around the country also adopt War Games to help train their students to be better docs.


  1. I clearly screw up a few times during the simulation,

    Good! That’s the best time. I imagine a lot of your screw-ups were things everyone does at some time, so I’d be happier if you did them before you got to me. If you’re going to screw up my life in an emergency, at least do it in an original way!

  2. Mark,

    Thanks for posting this.

    I am a long-expired NREMTP. We used to use scenario-based training also.

    I experienced two benefits: 1) as you said, it’s an inoculation against pressure/nerves and 2) it helps maintain situational awareness skills. EMTs get what used to be referred to as “big eye” where we would focus on the first problem we found. It’s bad for patients to deliver them to the hospital with perfectly splinted bilateral wrist fractures having totally missed the stroke symptoms that precipitated the fall.

  3. I dig it! They should do this for clinical psychologists as well!

    You didn’t mess up that much, It’s still pen and paper, you’re not there with the patient. I sincerely doubt you would have administered 2L of IV if you got results back that said you should have backed off.

    Here’s a question though. In a hospital setting, how much time could you spend with this patient? How in communication are you with the nursing staff, and how much of that can they do themselves with your orders?

    It would seem like there’s not enough doctors to go around.

    Also on a side note, new blog, nifty. Have you seen Sicko?

  4. The goal of “building reflexes” is similar to how I am learning plant identification. A plant taxonomy class teaches you the use of keys and recognition of a few families, but going out in the woods with a knowledgeable person is how you learn sight ID. A good teacher (in my case, a forest manager I’m friends with) starts with mnemonics and learning telltale species or genus characters and supports you in gaining the experience needed to recognize the plant without consciously relying on the mnemonics and signs that are so useful early on.

  5. Does this look like a D&D session to anyone else?

  6. As a medical technologist of many years it is very interesting to see how the ‘numbers’ we provide you fit into delineating the nature of the ‘problem(s)’ you and your patient face and your treatment response.

    Very interesting, thanks for having the . . .’guts’ (and it surely must have been a bit intimidating to make public those minor ‘screw ups’, to quote you…) to blog this great example of the science of medicine.

    Very nicely done.


  7. Actually giving fluids wasn’t a big mistake, in a patient who appears to be going into shock you should give fluids until proof of a contraindication appears. I guess I could have suspected cardiogenic shock with bradycardia in the clinical picture and been more cautious, but it wasn’t what I consider the big slip up.

    Those were, failing to mention that I would do an abdominal exam on a patient post-op day 3 from abdominal surgery, and suggesting pacing pads before giving the very obvious chronotropic/dromotropic pharmacologic agent helpful for bradycardia – atropine. I mentioned chronotropes but there is an order to these things that should be observed. First the chronotrope, then pacing either via pads or by dropping a wire. It’s still not so clear cut, atropine can be contraindicated in certain ischemic heart blocks and other agents like epi may have been better suggestions. But either way I should have gone pharmacologic first.

Leave a Reply

Your email address will not be published. Required fields are marked *