A concept can feel clear immediately after an explanation and still be difficult to retrieve later. This is familiar during medical revision: a question makes sense on Tuesday, but a related vignette on Friday produces the same uncertainty.
Spaced review makes that later encounter part of the plan. The aim is to retrieve the idea after some time has passed, correct what is missing and apply it in a different example. A useful review therefore asks more than whether you recognise the original answer.
The method below combines findings from learning research with a practical worked example. The suggested schedule is an illustration, not a scientifically established optimal interval for every learner, topic or examination.
What the research supports
In a controlled experiment, Karpicke and Roediger found that continued retrieval after initial learning improved later recall more than continued study under the tested conditions. The study used vocabulary learning, so it should not be presented as direct proof of improved performance in a particular medical examination. Its relevant contribution is the distinction between recognising material during study and retrieving it independently later.
Cepeda and colleagues examined how spacing between learning events related to later retention. Their findings show that the useful gap depends partly on how long the material needs to be retained. They do not establish one universal sequence of review days that every revision application should follow.
In experiments examining transfer, Butler found benefits of repeated testing over repeated study on later questions, including questions requiring inferences beyond the original wording. This supports examining whether learning can be used in a new problem, while leaving open how large the benefit will be for a particular clinical topic or individual candidate.
Choose a concept that can be tested clearly
Start with a question that exposed an important misunderstanding. Avoid creating a review item for every sentence in an explanation. A useful prompt should make you retrieve a decision, relationship or explanation that you previously could not use reliably.
For example, a candidate preparing for MRCGP AKT, UKMLA or another examination may confuse relative and absolute risk reduction. A vague review note such as "revise statistics" provides little direction. A better prompt asks how to calculate each measure and why the same relative reduction can correspond to different absolute benefits.
The target is then specific enough to test without rereading an entire chapter. It can also be varied, making it harder to answer through memory for one familiar set of numbers.
First attempt: explain the original example
Consider a hypothetical trial with 1,000 participants in each group, followed for two years. The event occurs in 40 people in the control group and 20 in the treatment group. These are invented teaching data, not results from a clinical study or evidence for a treatment.
The control risk is 4%, and the treatment risk is 2%. The absolute risk reduction is 2 percentage points, or 0.02. The relative risk is 0.5, giving a relative risk reduction of 50%. The number needed to treat is 1 divided by 0.02, giving 50 over the specified two-year period.
Before reading those answers again, explain the difference in words. Relative reduction describes the change in proportion to the control risk. Absolute reduction describes the difference in event risk between groups. The time period matters when interpreting the treatment benefit.
A few days later: retrieve before looking
An illustrative first review might occur after two or three days. Close the explanation and reconstruct the method from the prompt. If you immediately open the notes, you are checking whether the explanation feels familiar rather than whether you can produce it.
Record the point at which you become uncertain. Perhaps the arithmetic is straightforward but you confuse percentage points with a relative percentage change. Perhaps you calculate the absolute reduction correctly and then forget to convert it from a percentage to a proportion before taking the reciprocal.
Correct the specific error, then make a brief independent attempt. The review record should identify the misconception rather than merely mark the item as completed. A confidently incorrect explanation needs attention even if the final numerical answer happened to be right.
One week later: change the numbers and interpretation
Now consider another hypothetical two-year comparison. There are 4 events per 1,000 people in the control group and 2 per 1,000 in the treatment group. The relative risk reduction remains 50%, but the absolute reduction is 0.2 percentage points, or 0.002. The number needed to treat is therefore 500.
The useful question is why the relative figure stayed the same while the absolute benefit became smaller. If you can explain that without returning to the first example, you are demonstrating more than memory for its answer.
Then ask what additional information would matter before using these figures in a real clinical discussion. Relevant considerations include the population, outcome, follow-up period, uncertainty and harms. The numerical exercise alone does not establish that a treatment is appropriate for an individual patient.
Three weeks later: recognise the concept without its label
Place the principle inside a mixed practice session. The question should not announce "absolute risk reduction" in the title. It might ask how many additional people avoid an outcome, how to communicate benefit or why two apparently similar relative effects have different practical implications.
This checks whether you can identify the relevant concept when attention is divided between different subjects. If the topic label is necessary before the method becomes accessible, the next review should focus on recognising the kind of question being asked.
A missed answer at this stage does not make the previous work worthless. It supplies information about what remains fragile. Reconstruct the reasoning, correct the failure and schedule another meaningful encounter rather than repeatedly pressing through the identical item in the same session.
Adapt the schedule to what happens
| What happens at review | Useful adjustment |
|---|---|
| Correct answer with a clear explanation | Increase the interval or vary the context |
| Correct answer with uncertain reasoning | Examine the uncertainty before extending the gap |
| Wrong answer caused by a specific misconception | Correct it and use a fresh example sooner |
| Familiar item is easy but a new version is difficult | Add variation and compare the underlying structure |
| Clinical guidance has changed | Update the source and the review prompt |
The table describes a practical decision process. It is not a claim that a fixed algorithm can know the exact moment at which you will forget a concept. Exam timing and the importance of the material also influence when it should be revisited.
Use the tools to support the learning sequence
The iatroX question banks bring adaptive sequencing, spaced review and question-linked Socratic tutoring into the same study environment. That combination is useful when each attempted question leads to a clear next action: correction, explanation, delayed retrieval or a contrasting example.
A Tutor discussion can help expose why you confused relative and absolute benefit, but the discussion should end with a task you can complete independently. A later question then checks whether the explanation remains usable. Reading a persuasive answer is one stage of learning, not the final demonstration of it.
For guidance-sensitive clinical material, keep the source and jurisdiction attached to the concept. Spacing an outdated recommendation makes it easier to remember the wrong thing. The desired outcome is knowledge that remains both accessible and appropriate when a new question requires it.
Frequently asked questions
Is reviewing after 1, 3, 7 and 21 days the best schedule?
It can be a practical starting pattern, but there is no single optimal sequence for every topic and learner. Adjust the interval according to retrieval difficulty, the examination horizon and whether you can use the concept in unfamiliar examples.
Should I repeat the same question or choose a different one?
Both can have a role. Repeating a question can check recall, while a different example helps test whether you understand the underlying principle. If the original item becomes easy but new variants remain difficult, add explanation and variation rather than more identical repetitions.
Does spaced repetition guarantee that I will remember everything?
No. It organises further opportunities to retrieve and correct knowledge, but learning quality, feedback, topic difficulty and subsequent use still matter. The useful test is whether you can explain and apply the concept independently after a meaningful delay.
Connect question practice, guided explanation and later review with the iatroX exam question banks.