# Why Medical Students Fall Behind on Spaced Repetition > Spacing is well proven in medical education. What breaks is the daily review load, and it breaks in predictable places. Source: https://getmemset.app/blog/memset-for-medical-students Published: 2026-10-13 --- A recent survey of first-year medical students at one US school found that 94 percent of them used the same flashcard app, Anki. Among those users, 82 percent said it overwhelmed them. I keep coming back to those two numbers because they describe medical school better than any study guide I have read. Almost everyone has settled on the same method, and the method mostly works. What fails is the load that builds up around it. I am not a doctor. I build [a review planner](https://getmemset.app) for people who are learning a lot of material at once, and medical students are the most extreme version of that person I can think of. So I spent a few weeks reading how they study, where the research says their systems break, and which of those breaks a scheduling tool can actually do something about. Some of them it can. A couple it cannot, and I will say which. ![A glossy anatomical heart model made of translucent blue glass resting on a stack of closed textbooks, with a thin ribbon of light circling it in widening loops against a deep blue background](https://getmemset.app/blog/memset-for-medical-students.webp) ## The method is not the problem It helps to start with what is going right, because medical students are unusually well served by the evidence. In a 2007 randomised trial, third-year students finished a urology rotation and then received a short case question by email each week, covering two of the four topics they had studied. At the end of the year they were tested on all four. [The spaced topics held up significantly better](https://doi.org/10.1111/j.1365-2929.2006.02644.x), and the gap was widest for students whose rotation was six to eleven months behind them. The longer ago they learned it, the more the spacing mattered. A later study looked at board scores. Among 72 students at one school, the number of unique Anki cards a student had seen was [one of the independent predictors of their USMLE Step 1 score](https://doi.org/10.1007/s40037-015-0220-x), alongside the number of practice questions they had completed. It is a small, single-school correlation. It still points the same way as the trial. So medical students do not need convincing that spacing works. The trouble starts when a spacing habit meets a medical curriculum. ## The volume arrives faster than it settles The first struggle is the obvious one, and it is worth being precise about. Very few facts in the preclinical years are hard on their own. The brachial plexus is learnable. So are the coagulation cascade, the antiarrhythmic classes, and the foramina the cranial nerves pass through. The problem is that they arrive several lectures a day, in parallel, for two years, and every one of them is supposed to still be there at the end. Left alone, they will not be. A review of retention research in medical education estimated that students keep [roughly two-thirds to three-quarters of basic science knowledge after one year](https://doi.org/10.1007/s10459-008-9101-y), and a little under half after two. That is [the forgetting curve](https://getmemset.app/learn/the-forgetting-curve-explained) doing what it always does, applied to about the largest pile of material a student can be asked to hold. Step 1 going pass/fail in 2022 did not make any of this optional. The material comes back on Step 2 CK, on every shelf exam, and on the wards, where an attending asks why the patient's potassium matters and expects an answer before the team moves to the next bed. Very few degrees are this cumulative. **Nothing you learn is allowed to expire.** ## The review pile grows while you look away This is where the overwhelm in that survey comes from. Spaced repetition has a property that is easy to miss in the first month. Every new item you add creates future reviews, and those reviews stack. A student who adds a few hundred cards a week is not signing up for a few hundred reviews a week. They are signing up for a daily total that keeps climbing as long as new material keeps arriving, which in medical school is always. Then a day gets skipped. An exam, a cold, a family weekend. The next queue holds two days of work, and a student who was already at their limit has to choose between catching up and learning anything new. In the same survey, [75 percent of the Anki users said they never or rarely kept up](https://doi.org/10.7759/cureus.95674) with reviews from earlier modules, and 68 percent felt frustrated whenever they could not finish a day's reviews. > Nobody quits spaced repetition because they stopped believing in it; they quit because the number on the screen stopped looking possible. This is the problem Memset is built around most directly. Related material sits in a Category, and each Category has a daily limit. If Pharmacology is capped at four items a day and nine come due, four show up and the rest roll forward. The day stays a size you can actually finish. A review that lands two days late still does most of its job. A review that never happens because the list looked impossible does nothing at all. Missed days work the same way. Each memset has a grace period, and a repetition left undone past it makes the memset overdue. That resets your current streak, though your longest streak stays on record, and nothing else collapses. You pick up the overdue items and carry on, instead of returning to a wall that grew while you were gone. I wrote about the mechanics of this in more depth in a guide to [spaced repetition without the burnout](https://getmemset.app/learn/spaced-repetition-without-burnout). ## Your material lives in six places The second struggle gets talked about less. A medical student's understanding of renal physiology is not in one place. It is spread across lecture slides, a few pages of First Aid, a Pathoma chapter, a dozen question-bank explanations, a diagram drawn in a margin, and whatever cards a shared deck has for it. Flashcards handle one slice of that very well. They are excellent for discrete facts: a drug's mechanism, a bug's virulence factor, the nerve that supplies a muscle. They are clumsy for the rest. A pathway, a whole lecture, a differential, or the reasoning behind a question you got wrong rarely survives being squeezed into a card. The same survey found that 97.6 percent of the students using Anki worked from pre-made cards, alone or alongside cards they wrote themselves. That makes sense, since nobody in first year has time to write a full deck from scratch. A shared deck covers what its authors judged card-shaped, and anything outside it gets scheduled only if the student adds it. Memset works one level above the card. A memset is a short reference to material you already have. "Cranial Nerves - Neuro" is one of the examples I use when I explain the app. Another could be "Diuretics: lecture 14 slides, First Aid renal pharm, my missed qbank questions." When it comes due, you close everything, try to rebuild the topic from memory, and only then open the sources to see what slipped. If you already run a deck and it is working, keep it. Memset is not a replacement for [Anki](https://getmemset.app/compare/memset-vs-anki) when it comes to drilling facts. It covers the material that never made it into the deck, and it keeps the daily list of those topics to a size you set. ## Rotations break every routine The third struggle arrives in the clinical years. A preclinical study system assumes a predictable day, with lectures in the morning and a block of review in the evening. Clerkships take that away. Surgery starts before six. Call nights run late. Some weeks you are free by four, and in others you get home and go straight to sleep. A system built around preclinical life may not survive this schedule, and the easy response is to drop it. Without reviews, the preclinical material is left to the retention numbers above. Two things help. The first is keeping each review small enough to fit into a gap, like ten minutes between cases or the ride home. A memset review is one topic, recalled and then checked, not a two-hour queue. Reminders bring it to you, so it does not depend on an evening block that no longer exists. The second is letting old material slow down instead of disappear. Memset has an endless mode that keeps a topic active after its last scheduled repetition and brings it back at an interval you choose, such as every two months. That costs almost nothing during a hard rotation, and it keeps first-year material from quietly expiring. The urology trial is a small model of the same idea. A single question a week after the rotation ended was enough to change what students remembered months later. ## A guilt loop you do not need The last struggle sits underneath the others. A meta-analysis covering more than 17,000 students put [burnout before residency at around 44 percent](https://doi.org/10.1016/j.eurpsy.2018.08.006). A study app cannot fix the causes behind that number, and I would be lying if I suggested otherwise. What an app can do is avoid adding to it. A queue that ends every day with an unfinished number is a small daily verdict on your discipline, and over a semester those verdicts pile up. In the first-year survey, about a third of the Anki users said they had lost sleep over their reviews, and a similar share said they had forgotten to eat. The design choice I care about most is that **a finished day should always be possible**. That means a capped list you can complete, a missed day that costs you at most a streak instead of the whole system, and a schedule that bends when a rotation does. It sounds modest. For a medical student, it decides whether a review habit makes it through four years or dies in the first exam block. ## Where a planner stops helping A review schedule protects what you have already learned. It will not make you a good clinician. It will not teach you to reason through a presentation, build a differential, or pick the next best step with a clock running. Question banks do that far better, and in the Step 1 study the number of practice questions completed predicted scores alongside the flashcards. It also cannot stand in for understanding a mechanism properly the first time. Review keeps what you understood, but it cannot create understanding that was never there. What it can remove is a narrower failure: knowing something cold in first year and losing it by the time a patient needs it. ## A setup built for bad weeks If I were starting medical school tomorrow, this is roughly how I would set things up. - One Category per organ-system block, like Cardio, Renal, and Neuro, plus a separate one for Pharmacology, because pharm cuts across everything and piles up fastest. - A daily limit on each Category that matches a bad day, not a good one. - A memset per lecture cluster or high-yield topic, pointing at the slides, the First Aid pages, and your own notes instead of copying any of them. - The Forgetting Curve Strategy for new material, and endless mode for anything you will need on the wards. - On every review, recall first and open the sources second. None of that is clever. It is a plan designed for the week when everything goes wrong, which in medical school is a lot of weeks. The students in that survey were doing spaced repetition, and by their own account it was helping. Many of them also reported paying for it in sleep and stress. If you want to keep the method and lose the pile, with your slides, decks, and notes staying exactly where they are, that is the job [Memset](https://getmemset.app) was built to do.