How long should an MRCPCH revision session be?

There is no evidence-based universal length for an MRCPCH revision session. Set the session by a complete learning output: attempt a bounded set without notes, review important errors and uncertain successes, record one next action, and schedule delayed retrieval. Use shorter blocks when time or alertness is limited, longer blocks for integrated review, and separate full-length sessions for realistic exam rehearsal. Stop or take a break when review quality or reasoning deteriorates, not merely when a fashionable timer rings.

A study session is long enough when it closes a useful learning loop. It is too short when the candidate attempts material but leaves the feedback unresolved. It is too long when extra minutes create superficial explanations, repeated guessing or a growing queue of work that will never be revisited.

This guide offers an evidence-informed way to choose session length for Foundation of Practice (FOP), Theory and Science (TAS) and Applied Knowledge in Practice (AKP). The suggested time bands are practical starting points, not RCPCH rules or experimentally validated prescriptions.

The direct answer: choose an output before a duration

RCPCH publishes the structure and duration of the real examinations, but it does not prescribe how long a daily revision session should be. Learning research supports retrieval and distributed practice across many contexts, yet it does not establish one ideal MRCPCH study-block length.

The sensible unit is therefore not "minutes at a desk". It is a complete loop:

1. define one observable output;
2. retrieve, discriminate or apply knowledge without reopening the source;
3. compare the response with reliable feedback;
4. repair the important error or uncertainty;
5. decide what will be retrieved again and when; and
6. leave a clear restart point.

Session length is a capacity decision. Choose the shortest block that can complete the intended learning loop at acceptable quality.

This prevents two common errors: stopping when the timer ends but before review, and continuing because time remains after the useful task is complete.

What the evidence can and cannot tell us

Distributed practice supports returning across time

A systematic review in health-professions education included 56 studies and 63 experiments examining distributed practice and retrieval practice. Most experiments reported benefits over comparison conditions, although designs, content, delays and assessments were heterogeneous. The review also noted that time on task was often poorly reported.

A newer meta-analysis of applied classroom research likewise found an overall advantage for distributed over massed practice, but the studies varied in subject, learner level, spacing schedule and retention interval.

The useful conclusion is modest: dividing learning across revisits is generally preferable to treating all available hours as one uninterrupted exposure. Neither review identifies a universal 25-, 45- or 90-minute session for postgraduate medical examinations.

Retrieval quality matters more than continuous exposure

Retrieval practice means attempting to produce or select an answer before restudying it. A 2025 systematic and meta-analytic review found a small overall advantage for retrieval practice compared with elaborative encoding conditions, with effects varying by retrieval format and other study features.

This does not mean that every minute should be spent answering questions. It means a revision block should contain an observable attempt, reliable correction and later opportunity to retrieve again. Passive exposure can feel sustainable for longer precisely because it demands less diagnostic effort.

Short breaks can help fatigue without guaranteeing performance

A systematic review and meta-analysis of micro-breaks lasting ten minutes or less found significant pooled effects for greater vigour and reduced fatigue. The pooled effect on overall performance was not statistically significant, and the cognitive-task subgroup did not show a significant performance benefit. Longer breaks within the micro-break range were associated with greater performance improvement, but the evidence was heterogeneous.

Therefore, a short break is a reasonable recovery tool, not a guaranteed score intervention. It can protect the next learning loop when concentration or review quality is falling. It should not be presented as proof that one branded timer cycle is optimal.

Use the CLOSE session framework

CLOSE is a PaedVault editorial framework, not an RCPCH requirement:

1. C - Choose one output
2. L - Limit the scope
3. O - Operate closed-book first
4. S - Study the feedback selectively
5. E - Exit with a return plan

The framework makes the block self-contained enough to measure and connected enough to support later retention.

C - Choose one observable output

"Revise cardiology" is not a session output. It gives no finish line and encourages open-ended reading. Better outputs include:

• attempt one first-seen question set and review every error;
• retrieve ten previously corrected distinctions after a delay;
• compare two commonly confused mechanisms from memory, then verify them;
• complete one timed section and diagnose the pacing losses; or
• repair one recurring error category using new transfer questions.

The output determines the duration. A delayed-retrieval check may be brief. A first-seen mixed block with careful feedback requires longer. A full paper rehearsal must match the official conditions rather than an ordinary study timer.

L - Limit the scope before starting

Write three boundaries:

Boundary	Example
Content	One syllabus area or one mixed set
Quantity	A fixed number of questions or prompts
Review	Errors, correct guesses and low-confidence answers

Without boundaries, the session expands whenever a link, guideline or unfamiliar term appears. Put genuine side questions in a parking list and continue the planned loop unless the missing information prevents accurate review.

A scope can be reduced without making the session worthless. Ten properly reviewed first-seen questions can produce more usable evidence than forty rushed questions with explanations left unread.

The daily MRCPCH question-target guide explains how to size question volume around complete-loop capacity rather than an arbitrary quota.

O - Operate closed-book first

Begin with an attempt that makes current knowledge visible. Depending on the goal, this can be:

• an SBA answer with confidence recorded;
• a free-recall explanation;
• a blank-page reconstruction;
• a timed decision checkpoint; or
• a prediction before opening the source.

Opening notes first often changes the task from retrieval to recognition. Recognition is not useless, but it provides weaker evidence that the principle will be available without the source.

The attempt need not occupy most of the session. It creates the diagnostic question that the review should answer: what was missing, confused, misread or inaccessible?

S - Study feedback selectively

Not every question deserves the same review time. Prioritise:

• incorrect answers;
• correct answers reached by flawed reasoning;
• low-confidence correct answers;
• guesses between near alternatives;
• repeated error causes; and
• facts that are time-sensitive or require authoritative verification.

For a secure correct answer with sound reasoning, a brief confirmation may be enough. For a recurring discrimination error, the review may need a source check, a short contrast and a new question.

The session is too large if important feedback is repeatedly deferred. Unreviewed questions are not free progress; they are review debt.

E - Exit with a return plan

Use the final minutes to create continuity:

• one corrected principle;
• one recurring error cause;
• one item or distinction for delayed retrieval;
• one date or session for that return; and
• one restart note.

The restart note can be one sentence: "Next time, retrieve the difference between A and B before opening the explanation, then attempt five new mixed questions."

This small handoff makes a short session part of a distributed system rather than an isolated burst.

Practical session lengths to test

The following bands are editorial starting points, not evidence-based thresholds. Adjust them using output quality and delayed retention.

Twenty to thirty minutes: a minimum viable loop

Use a short block when time is fragmented or alertness is limited. A possible structure is:

1. two minutes to state the output;
2. ten to fifteen minutes of closed-book retrieval or a small question set;
3. eight to ten minutes of selective feedback; and
4. two minutes for the return plan.

This works best for delayed retrieval, a small error set or one narrow distinction. It is usually too short for a large first-seen block with detailed review.

Forty-five to sixty minutes: a standard focused session

This can support a moderate question block, selected explanation work and an exit plan. Keep the attempt and review inside the same session wherever practical.

If review regularly consumes all the available time, reduce the number of new questions rather than converting the next session into permanent clean-up.

Seventy-five to ninety minutes: an integrated block

A longer block may suit mixed first-seen practice, a timed section plus review, or a difficult error cluster. Plan a midpoint check or brief recovery pause if reasoning quality drops.

Longer is not automatically better. If the second half becomes rereading, answer recognition or shallow explanation, split the work across two sessions and insert delayed retrieval between them.

Full-length rehearsal: a separate session type

The current RCPCH structure lists FOP and TAS as two-hour examinations. AKP consists of two papers, each lasting two hours and thirty minutes. Realistic full-paper practice therefore has a different purpose from everyday learning blocks: it tests pacing, sustained attention, interface use and recovery under the official duration.

Do not use full-paper timing as the default daily session length. A rehearsal also creates substantial review work that must be scheduled separately.

The MRCPCH exam time-management guide provides pacing checkpoints for these longer conditions.

Choose length from the task, not the clock

Session purpose	Useful evidence at the end	Likely shape
Delayed retrieval	What remained accessible after a gap	Short, closed-book check
New question learning	First-seen decisions plus completed feedback	Attempt-and-review block
Error repair	A corrected cause tested in a changed item	Narrow, deeper block
Mixed transfer	Discrimination across several topics	Moderate mixed block
Timed rehearsal	Pacing, completion and process stability	Official or progressive duration
Weekly planning	Next actions based on evidence	Brief audit, not content study

Two candidates can spend the same 60 minutes on different valid sessions. The key comparison is not duration alone but whether the intended output was completed and remained retrievable later.

How to know when the session is too short

A session may be under-sized when:

• questions are attempted but important feedback is always postponed;
• the candidate cannot record why an answer was wrong;
• no delayed return is scheduled;
• every block ends in the middle of the same reasoning step; or
• setup and switching consume most of the available time.

The answer is not always "find more time". Reduce scope, prepare the materials in advance or use a smaller complete loop.

How to know when the session is too long

Look for process deterioration rather than a fixed minute:

• explanations are skimmed without identifying the decisive principle;
• several answers are changed without a named reason;
• confidence ratings stop distinguishing certainty levels;
• the same sentence is reread without a new question;
• errors are labelled only as "didn't know"; or
• the session creates more unresolved items than it repairs.

These are operational signals, not diagnoses. Persistent exhaustion, sleep difficulty, low mood or impaired functioning should not be treated as a productivity problem; the MRCPCH revision-fatigue guide explains that boundary.

When to take a break

Use a break at a natural boundary or when observable quality declines:

• after completing the attempt phase and before a demanding review;
• between two different task types;
• after a preplanned group of questions;
• when physical discomfort is distracting; or
• when two successive items receive superficial review.

A break should interrupt the task, not replace it with another high-friction attention stream. Stand, move, obtain water or look away from the screen if appropriate. Opening messages, social media or a new resource can turn a short recovery pause into uncontrolled task switching.

The distraction-resistant MRCPCH study guide explains how to protect the return point.

Why a Pomodoro timer is optional

The Pomodoro technique can provide a clear start, a visible boundary and a reminder to pause. Those functions may be useful. The familiar 25-minute interval is not, however, an evidence-based universal optimum for MRCPCH learning.

Use a timer as infrastructure:

• set it to fit the task;
• do not stop midway through recording a critical correction;
• do not continue poor-quality work merely because time remains;
• preserve the planned break; and
• compare the result with delayed retrieval, not just completed cycles.

If 25 minutes repeatedly fragments question review, lengthen the block. If 60 minutes repeatedly produces shallow work, shorten it. The timer serves the learning loop.

Adapt sessions to shifts and variable capacity

A fixed duration can fail when rota days differ substantially. Instead, define three modes:

Mode	Suitable output
Minimum	Small delayed-retrieval set plus one correction
Standard	First-seen block with completed selective review
Rehearsal	Progressive timed section or full paper with scheduled review

Choose the mode before starting. A minimum session is not a failed standard session; it is a deliberately smaller complete loop.

The rota-proof MRCPCH study plan shows how to place different session types into a changing week without creating revision debt.

Run a two-week session audit

Test session length rather than adopting it by identity. For two weeks, record:

Measure	Question
Planned output	What was supposed to be completed?
Loop completion	Were attempt, feedback and next action finished?
Review debt	What important work was left unresolved?
Reasoning quality	Could you name the cause and corrected principle?
Delayed retrieval	Was the correction available later?
Friction	What consumed time without contributing to learning?

Compare similar session types. A full-paper rehearsal should not be judged against a 25-minute retrieval set.

Keep the shortest duration that reliably completes the intended loop. If two durations produce comparable delayed retention, the shorter one may be easier to sustain. If the shorter block repeatedly abandons feedback, it is not efficient.

A worked planning example

Suppose a candidate has one hour after work and wants to practise a weak area. A poor plan is "study for 60 minutes". A stronger plan is:

1. attempt twelve first-seen questions without notes;
2. record confidence before feedback;
3. review every error, guessed answer and low-confidence success;
4. write one-sentence contrasts for no more than three recurring problems;
5. schedule those three for retrieval in two later sessions; and
6. stop when the loop is complete, even if a few minutes remain.

If twelve questions repeatedly leave feedback unfinished, the next experiment is eight or ten questions, not faster skimming. If the full loop finishes with stable quality, the candidate can cautiously increase scope while keeping the same session length.

Final answer

There is no single best MRCPCH revision-session length. Start with a practical band that fits the task, then judge it by loop completion, review quality and delayed retrieval. Twenty to thirty minutes can support a minimum viable retrieval loop, forty-five to sixty minutes can support ordinary question-and-review work, and longer blocks may suit integrated practice. These are starting points, not rules.

Reserve official exam-length sessions for realistic rehearsal. During ordinary revision, finish the learning loop before chasing more minutes.

Frequently asked questions

Is 25 minutes long enough for an MRCPCH revision session?

It can be enough for a small delayed-retrieval set, one narrow error repair or a minimum viable study loop. It is often too short for a substantial first-seen question block plus careful feedback. Judge the block by whether attempt, review and a return plan are completed, not by whether one Pomodoro cycle was logged.

Is two hours of MRCPCH revision too long?

Not necessarily. Two hours may be appropriate for an FOP or TAS rehearsal because that matches the current official examination duration, or for a planned integrated block with a break. It is too long when the later work becomes superficial, unresolved review debt grows or reasoning quality deteriorates. Long duration alone is not evidence of useful practice.

Should I take a break every 25 minutes while revising for MRCPCH?

No universal break interval has been established for MRCPCH preparation. Micro-break research supports reduced fatigue and greater vigour, but does not show a guaranteed overall performance benefit for cognitive work. Place breaks at useful boundaries or when quality declines, and choose an interval that allows the intended learning step to finish cleanly.

How many MRCPCH study sessions should I do each day?

There is no universal number. Count complete learning loops rather than calendar blocks. One session that includes first-seen retrieval, selective feedback and a delayed return can be more useful than several incomplete sessions. Choose a sustainable daily capacity that still permits sleep, recovery and broad syllabus coverage across the preparation period.

When should I stop an MRCPCH revision session?

Stop when the planned output and important feedback are complete, the next delayed retrieval is scheduled and a restart note exists. Stop earlier or take a break if explanations become superficial, the same text is repeatedly reread, reasoning cannot be stated or new questions are creating unresolved review faster than it can be repaired.

Sources

1. RCPCH - Theory exams: structure and syllabi
2. RCPCH - Theory exam sample papers
3. Systematic review of distributed practice and retrieval practice in health professions education
4. The distributed practice effect on classroom learning: a meta-analytic review of applied research
5. Retrieval practice versus elaborative encoding: a systematic and meta-analytic review
6. "Give me a break!" A systematic review and meta-analysis on the efficacy of micro-breaks