Why I teach emergency medicine residents to present cases "in reverse."
It's a cognitive strategy that refines thinking and potentially identifies cognitive traps.
As an “attending physician,” I supervise resident physicians and teach them how to take care of patients in the emergency room. There’s a lot to teach. We do it one case at a time. In a sense, fate decides what we learn each day. If a patient has a headache, we talk about how to think about dangerous headaches, and how to distinguish those from bothersome but benign ones.
Here’s the workflow. The resident signs up to see a new patient. They review any medical records and talk to the patient or the family. A physical examination is done. The resident does all of this without me, unless the patient is extremely unwell and needs immediate life-, limb-, or organ-saving attention.
After completing the initial evaluation, the resident comes to me and “presents” the case. Depending on the experience of the resident (and whether we’ve worked together before), the verbal presentation could be formal, highly structured, and lengthy, or it could be less formal and brief. It’s completely situational.
In medical school, we were taught how to “give case presentations” which describe a patient’s condition, ultimately leading to an assessment of what’s going on. It starts with the patient’s relevant medical history, the details of why they’re seeking care, their symptoms, etc. This is called the HPI (“history of present illness). Next comes the physical examination findings. After that, the resident provides a summary—an “assessment” of the situation, and a “plan” of action.
A good presentation is like a story. It starts with an interesting hook. “The patient has chest pain.” Ooh, chest pain? I’m interested! From there, the presenter should give me information about the HPI or exam that is consciously and conspicuously intended to tell a story that is either “concerning" or “reassuring.” When done well, by the time we get to the assessment and plan portion of the show, I should basically already know where everything is headed. A good presentation takes the listener down a path and, if accurate, the correct one.
The art of the presentation is learned and honed over time. The more experienced and refined the story-teller is, the earlier I can see where we’re headed. This is not supposed to be a suspense thriller with a surprise ending. Spoilers are a good thing.
Sometimes, though, I hear a case presentation that’s less focused and I can’t quite tell where we’re headed. By the time we get to the last act of the play—the assessment and plan—it seems as though our narrator doesn’t know the ending.
The reasons for this are variable. Sometimes, the resident hasn’t had time to think it all through. (Taking a few extra minutes to synthesize everything in their own mind before presenting to me might have fixed that). Other times, the case is just confusing and the presenter is just telling it like it is—that is, the story is murky, because the case is murky. In those cases, a conversation at the end of the presentation might replace a squeaky-clean summary and plan of action.
There’s a cognitive game here, and it’s at the core of what emergency doctors do. It’s more important than any procedure I can teach. I love this part. Taking brilliant young minds and pushing them a bit further.
One of the the ways I teach the residents how to organize their thoughts and refine their assessments is to do what I call “presenting in reverse.”
“Give me the end of the story first,” I tell them. “Then, tell me how we got there.”
That one change makes a huge difference. It also forces the resident to lead with conviction, rather than try to read my reaction as they proceed in the usual way.
Let’s use a case to demonstrate the difference.
The Usual Way:
“Mr. Smith is a 50-year-old male with a history of high cholesterol who is here with 3 days of intermittent chest discomfort. The pain is constant and nothing makes it better or worse. He has no symptoms of any infections. He has no difficulty breathing. He had a normal stress test 5 years ago and a clean CT of his coronary arteries 1 year ago. He is pain free now. His vital signs are pristine and physical exam is normal. There are no lung findings. His neurologic exam is normal and there are no signs of swelling or circulation problems in his arms or legs. His ECG has no sign of a heart attack or anything concerning. My assessment is that this patient is unlikely to be having even a mild heart attack. I also do not have any concerns for other things that these symptoms could mimic that we need to diagnose in the ER, such as a problem with the aorta, dangerous blood clots in the lungs, or other issues related to the heart or the surrounding tissue. We’ll draw some blood tests (including a troponin to look for signs of damage to the muscle of the heart), and get a chest x-ray. If everything is normal, we can discharge him and he can follow up with his primary care doctor (PCP) this week. I don’t think he needs to be admitted or placed in observation because he already had a negative CT coronary about a year ago and combined with our evaluation, that should suffice in terms of ruling out any impending heart attacks.”
It’s a good presentation. But sometimes I worry that my own reaction to the story as it comes alters where the resident goes with it. If I cast a wayward glance—because I think they’re overplaying or underplaying something—they’ll adjust on the fly. By the time we get to the end, they’re telling me what I want to hear.
By making them present in reverse, I essentially force them to lay out their cards early. From there, their spiel has to justify what they’ve announced up top. If the story doesn’t add up, they start to rethink it—which is what I want them to do in some cases, but not in others.
So here it is again, this time…
“I’m going to discharge Mr. Smith with close PCP follow-up after we rule out even a mild heart attack with our usual highly sensitive testing strategy. He’s at low risk for any other emergencies, including a blood clot, a problem with the aorta, an infection, or some other cardiac or lung problem because his ECG looks good and his physical exam is perfect, including normal vital signs. His symptoms have been ongoing for a few days, so it really doesn’t sound like a typical story for a mild heart attack, nor anything else that should scare us. There are no neurologic symptoms, breathing problems, or signs of infection or poor circulation. His only risks are his age and his high cholesterol. He actually had a clear CT coronary artery scan a year ago, and a negative stress a few years before that.”
In a way, each version has its benefits and its pitfalls. So, sometimes I ask for both. I’ll have the resident present any way they want—and then I’ll say, “Okay, now go in reverse and see if you’re still convinced.”
Usually my residents are so good that they know their patients backwards and forwards. (You had to know that was coming!) But once in a while, this little exercise reveals a cognitive lapse—which leads to a change of plan. It’s rarely a major shift. (Did I mention how good they are?) Still, those little details can matter once in a while.
My job is to make sure nothing gets missed—and to teach the next generation how to do that without going overboard with testing and treatments that are unnecessary.
Sometimes the best way to ensure that happens is just a simple change in perspective.
My high school geometry teacher taught us to do our proofs backwards too. Interesting article. Thanks
A good strategy I’ll try this sometime. I’ve don’t do much teaching these days as I am 100% clinical, and adding a student from our affiliated med school would be just more unpaid work on top of a day’s already unmanageable work load in primary care. But I digress from your thought provoking topic here...
I really love narrative medicine, and I use this intensely, mostly eschewing the EMR templates that chop narratives into incohesive checkboxes. But I have also learned the dangers in following a patient’s story too closely, and being led astray from the diagnosis
Have you ever tried just focusing on the chief complaint? For example I might listen intently to a long and nuanced story, but once I hear chest pain, I add that to the visit problem list. I’ll do a standard chest pain work up, explaining that while I was listening to the fact that it seems worse when the weather is cloudy, in parallel we have to rule out prime suspects in order of catastrophic to mundane.
The art is appreciating at once our patients’ complex stories AND their underlying architectural symptom-based simplicity.
Usually the patient ends up being right with their hunch, but often denial and rationalization and good story telling can be dangerous, too.
In terms of reverse history taking presented here, I think it’s another valuable way to break stories apart before we put them back together in diagnostic and therapeutic ways