I skipped morning report today to attend some of the Continuum of Health Sciences Education Conference over at the medical school. The keynote address was an engrossing talk by Dr. K. Anders Ericsson, a psychology professor at Florida State, titled “The Making of Superior Healthcare Professionals through Deliberate Practice: What Can We Learn from the Training of Chess Masters, Elite Athletes and Musicians?” Following are a few things I learned and some thoughts on how they relate to neurologic education.
Dr. Ericsson’s large body of work argues persuasively that it is deliberate practice, and not simply putting in your 10,000 hours, that leads to mastery in a field. What is deliberate practice? In this excerpt, he defines it as “activities designed . . . for the sole purpose of effectively improving specific aspects of an individual’s performance.” Think of hitting backhands over and over with a tennis coach as opposed to just playing a lot of tennis every weekend and expecting to get better over time.
One example he gave was in chess. What activity do you think correlates most strongly with mastery of the kings’ game: tournament play, non-tournament play, or analysis of positions and moves, such as the board shown above? One might guess that playing many, many chess games is the key to getting better, but it turns out that it’s the deliberate practice of studying chess positions and the moves made by top players that correlates most highly with performance. The more chess books on the shelf, the better the player.
Conversely, Dr. Ericsson argued that simply doing something for a long time is not sufficient to attain mastery. Length of education and experience are not correlated with teaching performance, for example. Accuracy of cardiac auscultation actually declines over time if physicians do not receive regular feedback. (On that note, check out this recent interactive feature in NEJM on pulmonary auscultation–quick, who can explain the difference between a ronchus and a wheeze?) In general, having a lot of experience leads to fast, automatic judgments (what Kahneman termed “system 1 thinking“) but not, by itself, to increased accuracy of judgment.
Dr. Ericsson made the very interesting observation that top musicians spend ~ 95% of their time practicing their instruments and only 5% in public performances, whereas physicians do the opposite. According to this study, we do so hardly at all, and here’s where I think some of the things we’re doing in this residency are relevant.
I’m biased, of course, but I believe that one of the best learning activities we have is the Monday stroke conference. I’ve always justified that view by asserting the superiority of case-based learning over topical lectures, the former being a more efficient way to construct knowledge of how to evaluate and manage stroke patients. It struck me today that our topic and the format are (or at least, can easily be) an exercise in deliberate practice. We can view a group of acute stroke cases as analogous to a series of chess boards. The pieces are arrayed in a certain pattern, and the challenge is to identify the best next move.
Mrs. Smith is 80 and has moderately severe dementia. She presents with a left MCA syndrome; NIHSS score is 23. She has atrial fibrillation but her warfarin was held for polypectomy 5 days ago. INR today is 1.7. Non-contrast head CT shows nothing acute. What’s the next move: IV tPA? CT angiogram and neuroendovascular consult? Conservative care?
Mr. Jones is 62. He presents with acute and isolated aphasia, which is moderate. NIHSS score is 2. tPA or conservative care?
Mrs. Miller presents with hyper-acute stroke symptoms, but CT shows a hypodensity of indeterminate age. tPA?
We go through hundreds of such cases throughout the course of a 3 year residency, on top of the hundreds more that each of you sees in real-time. I’ve had several residents tell me over the years, “Why would I do a stroke fellowship? That’s one of my strongest skills!” (Don’t let that dissuade you, Joe–I think we still have something to offer!) It is likely that this repetition, this deliberate practice of selecting the best diagnosis and treatment in many stroke cases over several years, with guidance and immediate feedback from the stroke faculty, is what leads to your high level of skill in this particular domain of neurology.
And I think we can and should expand this practice to other areas of neurology. Neuromuscular disease is an obvious example. We could run through a few cases in an hour-long session, discussing the clinical presentations and exams, and then reviewing the electrodiagnostics. After reviewing hundreds of such cases, you’d all be excellent at peripheral neurology and EMG interpretation (although this wouldn’t help as much in the performance of EMG–you still need to put in your time in the lab!). Liveson’s Peripheral Neurology: Case Studies is such a great book because it essentially helps you do just this.
MS is another area ripe for this type of practice. We, and especially our MS specialists, see case after case of patients with various symptoms, often normal exams, and MRIs showing lesions of uncertain significance. By reviewing case after case over the years, we’d get very good at quickly figuring out who’s got MS, who has an MS mimic, and whose picture doesn’t currently warrant a neurological diagnosis. We could look at hundreds of movement disorder videos, neuro-ophthalmology videos and fundus photographs, neuro-oncology clinic MRIs. Given enough time, we could theoretically employ deliberate practice in many areas of neurology and thereby build our clinical skills to a degree not attainable simply by being on call for 3 years (or 14, in my case). And to supplement the case conferences, you could take advantage of the AAN’s new publication, Neurology: Clinical Practice that, a little like the New England Journal, has regular case studies.
Of course in residency, there’s always a tension between the ideal pedagogical and curricular structure and the need to take care of the patients–deliberate practice needs to yield to clinical practice. I’m under no illusion that it is feasible or even desirable to be like musicians, with a practice:performance ratio of 95:5. But after attending Dr. Ericsson’s talk, I’m convinced that 0:100 in favor of clinical practice is not ideal either.
Finally, I’d like to make a point about how we can present cases in morning report and other venues in order to make the most out of the exercises. Often, a resident (sometimes even faculty) will start a case with the chief complaint and essentially begin a game of twenty questions: “Mr. Smith is a 55 year old man presenting with headache. What else do you want to know?” There ensues an haphazard flurry of questions from other residents and med students alike:
- “When did it start?”
- “Is he nauseated?”
- “Any numbness”
- “How about his pupils?”
- “Any sick contacts?”
- “What did his MRI show?”
Which is tantamount to asking:
- “Is there a black knight at E5?”
- “Is there a white rook at G3?”
- “Is there a black pawn at C3?”
I think it’s much more educational to get all of the facts that are reasonably expected to be known out on the table at the outset: Chief complaint, brief HPI, pertinent PMH, meds, etc. and even the exam. I think we sometimes draw out the presentation so much that there’s little time for the best parts–localization, differential diagnosis, and plan of investigation and management. If there is information that wouldn’t normally be known at the outset but only came to light due to your Holmesian powers of deduction then OK–keep us in the dark for a bit, and then we can discuss the cognitive process that led you to ask the crucial questions. But otherwise, let’s not waste time playing twenty questions–let’s deliberately practice neurology!