The Final Push: Mocks, Weak Areas, and the Last Fortnight

In a 2024 randomised controlled trial of 153 medical students preparing for an OSCE, the group that practised under structured rehearsal conditions scored significantly higher on the exam itself (p < 0.001) [5]. The effect was not small. The intervention was not expensive or proprietary. It was the structure of the final preparation that did the work.

The fortnight before MRCS Part B is where that structure either gets built or doesn’t. This article covers what the evidence says about that fortnight:

• Where the hours go — and how to divide them between weak areas, mocks, and maintenance
• Why mocks are the single highest-yield activity, and how to run one without a course or a faculty
• How to fit serious preparation around an on-call rota
• What to do in the last 72 hours, and why those days specifically reward rehearsal over reading

Where the Hours Go

The final two weeks are not for maintaining strong areas. They are for moving ambers towards green and reds towards amber. A candidate who spends the last week revising anatomy because the anatomy revision feels productive will not change the marksheet. The marks live in the stations that are currently failing or borderline.

This sometimes means starting new content. If communication-with-relatives is the weakest station and there is a specific framework not yet learned, that framework is what to work on this week. New content addressing a specific gap is exactly the work the fortnight is for. What is less productive is broad new reading on topics that are already in the amber-to-green range.

Interleaving the work across station types is worth noting. Schorn and Knowlton's 2021 work on contextual interference showed that learners who interleaved different task types during practice consistently outperformed those who blocked them, with the effect holding for both retention and transfer to novel tasks [1]. The 2023 systematic review of distributed and retrieval practice in radiology education reached the same conclusion across the medical education literature [2]. In practical terms: a Tuesday evening that rotates anatomy → applied science → communication outperforms three hours of anatomy alone, even with the same total study time. The mind has to retrieve what kind of question is this before retrieving the content, and that retrieval is what produces durable learning.

Mocks Are the Differentiator

Phillips and colleagues' 2024 systematic review of distributed and retrieval practice in health professions education found that 43 of 63 experiments showed significant benefit over passive review [3]. The de Lima 2026 RCT puts numbers on the OSCE-specific effect: 67 students randomised either to a peer-led mock OSCE or to standard preparation; the mock group showed significantly lower state anxiety (Cohen's d = 0.69, p = 0.009) and a directional trend towards higher OSCE scores (p = 0.055) [4]. The structure of the rehearsal, not the formality of the setting, produced the effect.

The anxiety reduction is worth dwelling on, because it is the mechanism most candidates underestimate. Gulpinar and colleagues' 2024 RCT (n=153) of a structured stress and anxiety coping programme delivered alongside OSCE preparation found significantly lower post-intervention State-Trait Anxiety Inventory scores and significantly higher OSCE performance (both p<0.001) [5]. The two effects are linked: candidates whose anxiety is well-managed perform better. Mock practice is one of the cleanest ways to manage anxiety, because the unfamiliarity of the OSCE format is what most candidates are responding to. Rehearse the format and the format stops being threatening.

The review matters as much as the mock. After each one, the question is not “what did I score” but “which specific marks did I miss, and why”. Write next to each missed mark either “didn't know it” or “knew it but didn't say it”. The first is a content gap. The second is a delivery gap. They close differently — the first with reading, the second with repeated verbal rehearsal.

The Working Week

The realistic picture for most surgical trainees is that the fortnight before MRCS Part B contains on-calls, theatre lists, ward responsibilities, and a working week. The advice to “protect your sleep” is not useful in isolation when a candidate has a long day on Tuesday and a night shift on Thursday. The operational moves that change this, where flexibility exists, are these:

The surgical-residency sleep literature is more mixed than popular consensus suggests. Some studies found no decrement after 24-hour shifts; the 2024 anaesthesia study (Krueger et al.) found clear reduction in clinical reasoning after 24-hour shifts independent of subjective fatigue [6]. The picture is not that one bad night guarantees a poor exam — it is that chronic restriction across the fortnight erodes the gains of preparation. A single bad night is recoverable. Two weeks on five hours is not.

Peer rehearsal is the lever most trainees under-use. Bevan and colleagues' 2019 evaluation of peer-led practice OSCEs ran 15 structured peer-led sessions over five months and demonstrated that the format was feasible, sustainable, and rated as significantly improving exam preparedness by participating students [7]. The advantage is operational as much as pedagogical: a peer-led mock can be arranged for a Saturday morning by three trainees with a printer and a stopwatch, without faculty time, without cost, and without depending on an institutional course running at the right week. If your deanery is not running anything in your final fortnight, two evenings of peer rehearsal with a colleague who has sat the exam will close the gap.

The Last 72 Hours

The taper in the last three days is real, but it is a taper of volume rather than of focus. A workable shape:

Young and colleagues' 2014 review of cognitive load theory in medical education makes the case for why these last three days specifically reward structured rehearsal over additional content [8]. Working memory is a limited resource; the candidate who arrives with their working memory pre-loaded by chunked, well-practised opening sentences for each station type performs better than the candidate with the same factual knowledge but without the pre-formed chunks. The mechanism is not motivation. It is that the rehearsed opening sentence costs the working memory almost nothing to deploy, leaving more capacity for the substantive thinking the station actually requires. The three days before the exam are when those chunks are consolidated.

One observation about the night before. Sleep onset is often slower than usual, and most manage four to six hours rather than seven to eight. This is normal; the exam is still passable. Resting warm in bed is restorative even when sleep is intermittent.

The Day Itself

The corridor protocol between stations — the slow breaths, the explicit close of the previous station, the two reads of the next station's instructions — is covered in the wellbeing piece in this series. Rehearse it in the fortnight before, not on the morning itself. The candidates who pass do not arrive feeling nothing; they arrive having rehearsed enough that the day plays out as a sequence of familiar nine-minute encounters. The aim of the fortnight is to make the day feel that way.