I was a student in a CCU in 1991 when an attending cardiologist asked, “Why did you order that echo?” Not how to interpret it. Not what it shows. Why I ordered it. I didn’t have a good answer — and I don’t think I was supposed to. (“Because echos are awesome and we get a ton of information from them?” probably wasn’t the answer he was looking for.) That question took about 45 seconds. The lesson it produced lasted 30 years.
What Reflection Actually Is
There’s a version of “reflection” that clinical education programs have gotten wrong for decades. It shows up as journaling assignments. End-of-rotation summaries. The portfolio requirement nobody reads twice. Reflective practice in that sense means write something down about your feelings and your growth, submit it by Friday. That’s not what I’m talking about.
What I mean is the Socratic moment — the pause between what happened and what you understood about why it happened. The attending who interrupts a presentation to ask “what were you thinking when you chose that medication?” is doing something different in kind from the one who corrects the choice and moves on.
Correcting the choice gives the student the answer. The Socratic question makes the student examine her reasoning. One deposits knowledge. The other builds the capacity for clinical judgment.
I learned a long time ago that learning doesn’t happen while you’re actively engaged in the experience. It only happens during the time you reflect on it. So we have to build that time into learning. Journaling can help that process. But just being curious enough to reflect on the experience while driving home is even more important.
Why Reflection Got Cut
Clinical throughput is the short answer. The longer answer involves a system redesigned around efficiency at the exact moment the learner-to-preceptor ratio was increasing.
An attending with four students, 20 patients, and a half-day clinic doesn’t have 45 seconds for “why did you order that?” She has the correct answer and a limited window to deliver it. She’s not being lazy. She’s trying to get home before 9 to see her kids.
The Socratic model requires slack. You need time for the learner to fail at answering, to search for the reasoning, to hit the edge of what they know and sit there for a moment. That silence is productive. It’s where the learning happens.
Modern clinical environments don’t produce slack. They produce throughput. Every minute with a learner is a minute borrowed from care delivery, documentation, or the list of tasks that already ran past the shift.
Nobody designed a training system that cuts reflection. They designed a clinical system that cuts whatever doesn’t move patients. Reflection doesn’t move patients.
What We Lose When It Goes
Clinicians who accumulate hours without reflective practice get experienced, and experienced and good aren’t the same thing. They’ve seen a lot. They’ve handled a lot. They can move fast, and they almost never stop to ask why because the pattern recognition is deep enough that the question seems unnecessary.
Good clinicians update their mental models. They notice when a pattern doesn’t fit a specific case and they ask why. They catch themselves making an assumption and interrogate it.
Anders Ericsson spent most of his career studying how expertise actually develops — musicians, chess players, surgeons. What he kept finding was that deliberate practice is not the same thing as repetition. Repetition builds fluency. Deliberate practice — which requires reflection on errors and near-misses with specific feedback — builds judgment. Flight simulators work not because pilots accumulate hours in them but because they accumulate feedback. Instant, specific, calibrated feedback on why the approach went wrong.
Clinical training at its worst is repetition without feedback. Students accumulate hours. They follow the workflow. They see enough patients to feel comfortable. What they don’t develop is the habit of examining their own reasoning — and that’s the habit the best attending builds with a 45-second question.
What Nobody Has Time to Fix
The attending who asks “why did you order that?” isn’t rare because she’s exceptional. She’s rare because she’s working in a system that’s eliminated the conditions for that question.
I’ve seen it come back in small pockets — residency programs with protected teaching rounds, simulation centers where the debrief runs longer than the scenario, preceptors with small enough panels to sit with a learner and think out loud. Those pockets exist, they’re expensive, and they don’t scale.
Part of what I’m trying to build into NP prep is that question — not just what’s right, but why, and what the reasoning looks like when it goes wrong. It’s not the same as 45 seconds with someone who’s been in a CCU for 20 years. Nothing is. But the question still matters, it mattered in 1991, and it’s still the step that gets skipped.