Most note-taking advice assumes the hard part is understanding. In medical school the material is rarely conceptually difficult. There is simply too much of it, it must be recalled cold, and it must still be there in two years for boards.
That changes what a good note is. Notes that are pleasant to write and nice to look at are, in this context, mostly a way of feeling productive. What you need is a system that produces retrieval attempts at scale.
The uncomfortable consensus
Ask senior students and you will get a version of the same answer: flashcards do the retention, notes do the understanding, and the two are different files.
This is not app tribalism. It follows from the arithmetic. If a block contains several thousand testable facts and you will be examined on them months later, the only method that has ever worked at that volume is scheduled retrieval. See spaced repetition and the testing effect for the evidence base, which is unusually strong.
The mistake is concluding that notes are therefore worthless. They are not. They are just not the retention mechanism.
What notes are still for
| Purpose | Format that works |
|---|---|
| Understanding a mechanism the first time | Prose or a diagram, written once, in your own words |
| Seeing how things relate | A table, or a one-page map of a system |
| Distinguishing things that look alike | A comparison table. This is where most exam mistakes live |
| Remembering discrete facts | Not notes. Cards |
| Clinical reasoning patterns | A running document of cases and the thing that surprised you |
The highest-value note in preclinical years is the differentiation table: two or three conditions that present similarly, side by side, with the discriminating features in their own column. Those are the questions that get missed, and prose does not solve them.
The second highest is the one-page system map: everything about the kidney on a single sheet, hand-drawn if possible. It is worthless as reference and excellent as a thing to redraw from memory, which is active recall with a lot of retrieval per minute.
Lectures: stop transcribing
Recorded lectures at 1.5x plus a transcript makes transcription pointless and tempting at the same time. Write nothing during the first pass except questions and confusions.
A workable pattern:
- First pass, fast, no notes. Just watch or listen. You are finding out what you already know.
- Second pass, paused. Write only what you did not already know, and only in your own words. Turn each item into either a card or a line in a table.
- Same day, close everything and recall. Three minutes. What were the five things in that lecture?
Step three is what makes the rest stick, and it is the step people cut when they are behind. See how to take lecture notes and read-recite-review for the general form of this.
Cards: the rules that matter
Whether you write your own or use a shared deck, a few things determine whether the system works:
- One fact per card. A card asking three things fails as one, and you cannot tell which part you did not know.
- Write cards from your own confusion, not from the whole lecture. A card for something you already know is a tax you pay every review for years.
- Cloze the discriminating word, not a random one. If the card would still be answerable by someone who does not know the material, it is testing recognition.
- Do not skip days. The scheduler assumes you come back. A week off does not delay the work, it multiplies it.
Premade decks are a legitimate shortcut and a real trade-off: you get coverage immediately and lose the encoding that comes from making the card. The common compromise, using a shared deck and writing your own cards only for what you keep getting wrong, is sound.
Clinical years are a different problem
Once you are on the wards the bottleneck changes. There is little time to write and the material is cases, not facts.
What works is small and fast: a pocket note or phone note per patient with the presentation, what you thought, what it turned out to be, and the one thing you did not know. Write the last part as a question, look it up that evening, and answer it in one line.
That habit produces the single most valuable document of your training, because it is a record of your own gaps in the order you found them. It is also nearly effortless compared to preclinical note-taking. See how to keep a work journal for the same idea outside medicine.
Be careful with patient details. Identifiable information does not belong in a general notes app, and the rules on this are not advisory.
The traps
Rewriting notes to look good. This feels like studying and is copying. See the illusion of competence.
Highlighting entire pages. Consistently one of the least effective techniques studied, and the most popular.
Building an enormous reference document. The document nobody ever reopens is the classic med school artefact. If it is not generating retrieval, it is not studying. The collector's fallacy covers why it feels otherwise.
Switching systems mid-block. The cost of migration is always higher than the benefit of the better app. Pick something adequate and stay.
What to do this week
Keep one notes app for understanding, one card system for retention, and one running list of things you got wrong. Write less than you think you should, recall more often than feels necessary, and protect the daily review before anything else.
More in how to study from notes, how to take notes for exams, and the best notes app for medical students.