Preparing for USMLE, PLAB and typical finals

• USMLE, PLAB/UKMLA and medical school finals are gatekeepers, but they test different things: the USMLE tests readiness for supervised then independent US practice; PLAB and the UKMLA set a common threshold for entry to UK foundation training; finals test what your own school has taught.
• Every exam rewards applying knowledge to a vignette, not recalling isolated facts. Preparation that measures itself against timed questions, then reviews errors, beats rereading.
• The best-supported study methods are spaced (distributed) practice and retrieval practice (testing yourself). Both show consistent advantages over cramming and over passive re-reading in cognitive and health-professions research.^[9][11]
• Sleep is a study aid: adults generally need 7 to 9 hours a night, and cutting sleep to gain hours harms recall and judgement.^[8]
• Treat preparation as a project with a calendar, not a feeling. Work backwards from the exam date, protect a daily routine, and use released, official blueprints to decide what to study.^[4]

By the end of this page you should be able to:

1. Explain what USMLE Steps 1, 2 CK and 3, PLAB 1 and 2, the UKMLA and the MCCEE are designed to measure, and how each is structured.
2. Find the official eligibility, registration and rule documents for each exam and know what each one contains.
3. Build a week-by-week preparation plan that uses spaced and retrieval-based learning and protects sleep.
4. Choose and use a question bank honestly and effectively.
5. Prepare for written finals and for an OSCE without last-minute cramming.
6. Recognise burnout, isolation and exam-integrity traps, and know where to get help.

A licensing examination is not a school test with a higher pass mark. It is a public-safety decision made from a short sample of behaviour: whether a doctor can be trusted to take responsibility for patients. The USMLE describes Step 3 as "the final examination in the USMLE sequence leading to a license to practice medicine without supervision".^[3] The UKMLA exists to set a common threshold for doctors entering UK foundation training.^[7] Finals and OSCEs exist so that a school, which takes responsibility for you, can satisfy itself and its regulator that you can practise safely on day one.

That difference changes how to prepare. School study often asks "did the learner understand this lecture?"; licensing study asks "can the clinician act safely and correctly with this patient, under time, from memory?". The preparation that follows therefore puts timed, mixed, applied practice at the centre and treats reading as support, not as the main activity.

These exams also differ in consequence and timing. A final examination usually sits inside an academic year; failure can be remediated within the course. A failed licensing examination usually costs a full cycle or a retake fee and may delay registration. Plan accordingly and build margin into your calendar.^[4]

What it measures. Step 1 "assesses whether you understand and can apply important concepts of the sciences basic to the practice of medicine, with special emphasis on principles and mechanisms underlying health, disease, and modes of therapy".^[1] It is therefore a mechanism and application exam, not a fact bank. Questions ask you to use a physiological or pathological fact to predict an outcome, a lab value, a drug effect or a diagnosis.

Structure. Step 1 is a one-day examination administered at Prometric centres. There are two current forms of the day: before 14 May 2026 it is divided into seven 60-minute blocks with at most 40 questions per block and at most 280 items overall, with at least 45 minutes of break time; from 14 May 2026 it is divided into fourteen 30-minute blocks with at most 20 questions per block, in one 8-hour session, with at least 55 minutes of break time and a 5-minute optional tutorial.^[1] The content outline and specifications for systems, physician tasks and disciplines are published on the Step 1 pages.^[1]

Scoring and eligibility. Eligibility requirements, time and attempt limits, retakes and policy on changed eligibility are listed in the USMLE Bulletin of Information and on the Step 1 pages.^[1][4] Check the Bulletin for the scoring rules that apply in your year, since passing standards and reporting formats are set by the sponsoring boards and change over time.^[4]

When people take it. Step 1 is ordinarily sat once the basic-science years and their integrated clinical exposure are complete, so that the clinical examples in questions have some grounding. Schools set their own policies and some have moved Step 1 after early clerkships. Your dean's office and the USMLE pages are the sources for the rules that apply to you.^[1]

What it measures. Step 2 Clinical Knowledge "assesses an examinee's ability to apply medical knowledge, skills, and understanding of clinical science essential for the provision of patient care under supervision", with emphasis on health promotion and disease prevention.^[2] Exams in this style reward integrated clinical reasoning: reading a vignette, generating a differential, choosing an investigation and deciding management.

Structure. Step 2 CK is a one-day examination at a Prometric centre. Before 7 May 2026 it is divided into eight 60-minute blocks (at most 40 questions a block, at most 318 items, at least 45 minutes of break time); from 7 May 2026 it is divided into sixteen 30-minute blocks (at most 20 a block, at least 55 minutes of break time plus a 5-minute tutorial) in one 9-hour session.^[2]

Preparation emphasis. Unlike Step 1, Step 2 CK assumes mastery of normal function and of basic mechanisms, and instead tests that you can do something with them. The most efficient preparation is therefore a broad review of clinical medicine organised by presenting problem and body system, plus practice at the 'next best step' question type.

What it measures. Step 3 "assesses whether you can apply medical knowledge and understanding of biomedical and clinical science essential for the unsupervised practice of medicine, with emphasis on patient management in ambulatory settings".^[3] It is the final examination in the sequence leading to unsupervised licensure.^[3]

Structure. Step 3 is a licensing examination. It uses multiple-choice questions and computer-based case simulations, which test management over time rather than a single decision. The official Step 3 pages and the Bulletin of Information are the places for the current day structure, numbers of questions and cases, scheduling rules and eligibility.^[3][6]

Eligibility. Passing Step 1 and Step 2 CK and holding (or being close to holding) a medical degree are the usual prerequisites; foreign graduates need ECFMG certification in addition. Confirm the exact conditions, attempt limits and scheduling on the USMLE site, since these are the rules you are signing when you apply.^[3][4]

Who it is for. PLAB is the route for doctors whose primary medical qualification is outside the UK (or EEA/Switzerland) and who want to join the UK medical register without an exempting qualification. The PLAB guide on the General Medical Council (GMC) website is the definitive statement of who needs it and what passing allows.^[6]

Structure. PLAB part 1 is a written examination of 180 single-best-answer (SBA) questions.^[5][6] PLAB part 2 is an OSCE of 16 clinical scenarios (usually 18 stations in total, including 2 rest stations).^[6] Together they test whether a doctor trained elsewhere can meet the standards expected of a UK doctor starting foundation training. Part 2 in particular tests the communication, examination and safety behaviours that paper questions cannot reach: consent, chaperoning, handover, and shared decision-making.

Registration and rules. Registration, fees, venues, the number of permitted attempts, the time window after part 1 and the booking rules are published by the GMC on its PLAB pages.^[6] Rooms for part 2 are released in booking windows; planning means securing a slot before planning travel.

The UK Medical Licensing Assessment applies to UK medical students in their final year (from 2024 onwards) and to some international graduates seeking entry to foundation training. It is designed to set a common threshold across all UK medical schools.^[7]

Knowledge component. The applied knowledge test consists of 200 SBA questions, normally presented as two papers covering the MLA content map; a published attempt in which ChatGPT answered 191 of the 200 screened questions illustrates the format.^[7]

Skills component. The clinical and professional skills assessment is an OSCE-style examination of clinical skills, including communication and professional behaviours. UK schools run it as part of finals; the current structure and candidate numbers are published by the GMC.^[7]

What this means for preparation. UK students already sit the skills assessment as a final; international candidates need to check the GMC's UKMLA pages for the assessment offered in their route. Anyone preparing should build from the MLA content map rather than from a textbook index.

The Medical Council of Canada Evaluating Examination (MCCEE) is a screening examination that historically formed part of the pathway for international medical graduates, and in some areas for Canadian students, into the Canadian qualifying examinations.^[6] The route has changed over time and continues to change. A candidate should treat the MCC website, the MCC examinations catalogue and the MCCQE and NAC examination pages as the source of truth for which examinations are needed now, eligibility, deadlines and the fee schedule, rather than memory or forums.^[6]

Finals are set by individual schools and vary more in format than any licensing exam. Despite that variation, most have two parts:

• A written part, usually SBAs and short structured answers, sometimes with extended-matching questions. It tests knowledge and clinical reasoning in a form close to licensing exams, and practice questions are the natural preparation.
• A clinical part, usually an OSCE and sometimes a long case. An OSCE is a circuit of standardised stations assessed by trained examiners, each station testing one skill: history-taking, examination, a procedure, communication or professionalism. It was developed precisely so that clinical skills could be tested reliably rather than in an impressionistic "long case".^[6]

What preparation differs. Written finals reward breadth and a good error log. OSCEs reward rehearsed, observed performance: you must be able to examine a system methodically while speaking aloud, under timed pressure, with a stranger watching. Neither is learned from reading. Practise in front of a peer or teacher and get explicit feedback on structure ("tell me what you are ruling out", "that question was closed; reopen it").

Rules change and rumours do not. The table below lists where the official statement for each exam lives. Put those documents in your preparation folder and re-read them one month before you apply, because eligibility windows, attempt limits, permitted test dates and fee structures move.

Two practical points. First, the Bulletins and guides are the documents you certify you have read when you apply, and registrants have been failed for breaching rules they claimed to have read.^[4] Second, eligibility windows (for example, how long after a qualification you may sit) are enforced strictly, so check them early in the process, not in the application week.^[4]

There is no single correct timetable, but there are patterns that fail and patterns that work. Failing patterns: starting with the subject you like best; reading a single textbook end to end; switching to a new resource every two weeks because the last one feels "wrong"; and leaving all timed practice to the final month. Working patterns begin with the exam date and work backwards, and treat preparation as a course with weekly deliverables.

A reasonable overall shape:

1. Weeks before formal study (foundation phase). Complete or consolidate the course material as you go, using short retrieval sessions (a few self-test questions after each lecture) rather than last-minute review. This is where spaced practice does the most to prevent the forgetting curve from taking hold.^[10] Spacing works best when retention is tested over months: in a meta-analysis of distributed practice, every study with a retention interval longer than a month showed a clear benefit of spaced study over massed study.^{[9] [16]}
2. A dedicated preparation block. Most high-stakes exams have a final block in which ordinary teaching stops. Reduce your commitments, fix a daily routine, and use it for consolidation and for gap-filling rather than for first exposure. Keep the load sustainable; volume without absorption just creates fatigue.
3. Final two weeks. Shift to full-length, timed practice, review error logs, and gentle tapering of volume for sleep. Do not introduce new resources.

How long should a dedicated block be? There is no evidence-based prescribed length, and figures you see online (for example, quoted hour totals or "six weeks") are opinions or industry averages, not standards. Decide from three practical tests: the number of unmastered topics identified by your first two mock exams, the date by which your school or job will release you, and the date you will be unable to concentrate on study. Plan backwards from the latest of those constraints and keep a week of slack for illness and for the inevitable slippage.

Memory decays rapidly after first learning (the forgetting curve) and spacing short reviews across time slows that decay.^[10] Distributed practice outperforms massed practice for long retention: in Cepeda and colleagues' meta-analysis, every study with a retention interval of more than a month showed a clear benefit of distributed over massed practice.^[9][15][16] Practical rules:

• Re-encounter a topic soon after first learning it, then after several days, then after weeks.
• Mixed blocks (interleaving cardiology, renal and pharmacology in one session) feel harder and are slower, but they produce better discrimination and retention than blocked practice.^[9]
• "Optimal lag" exists: sessions that are far too close together are just cramming, and sessions that are very far apart help less; the research is complex, but a sensible default is to let the gap grow as your mastery grows.^[9] A flashcard program that schedules reviews by difficulty automates this.

Being tested on material improves retention more than re-reading it, even when the test is difficult and initial recall is poor. This testing effect is among the most robust findings in cognitive psychology and in health-professions education; Dunlosky and colleagues' review of learning techniques rated practice testing (and distributed practice) among the most effective strategies.^[11][14][15] For exam preparation:

• Do mixed, timed questions early, even when you feel unprepared. Wrong answers drive the strongest subsequent learning if they are reviewed properly.^[10][11]
• Prefer generating an answer before looking at options where the item type permits; SBA questions still require a full review of why each wrong option is wrong.
• Use low-stakes quizzes, flashcards and closed-book "brain dumps" rather than only highlights.

Kerfoot's randomised trial in clinical education found that spaced online education improved retention of clinical knowledge by medical students compared with a single block.^[12] Digital spaced systems build on this, but the benefit comes from the spacing and feedback, not from the app: a paper card box used faithfully outperforms a sophisticated app used passively.

Internet-based learning in the health professions shows positive effects on knowledge when it includes practice, feedback and interaction; the medium matters less than the design.^[13] Therefore, replace a video course with a question bank if you are short of time; replace neither with passive notes.

Adults generally need 7 to 9 hours of sleep a night. Seven hours is the minimum associated with healthy adult functioning, and sustained sleep deprivation is associated with impaired performance and serious health problems.^[8] Sleep is also when memory is consolidated, so late-night studying at the expense of sleep is not a trade but a loss.

Preparation is a long, often isolating project. Regular physical activity supports mood and sleep during demanding study periods. Keep a small, regular commitment rather than heroic sessions.

Daily structure. Front-load the most demanding cognitive work into your best hours and protect a daily focused question session every day, even on lighter days. Fill gaps with review rather than with new content late at night. Take short breaks between study blocks; the breaks are part of the schedule, not concessions.

Weekly structure. A workable week has four elements: one to two new or consolidated topics, one mixed timed question block, one full review of the error log, and one genuine rest day. In the final month, convert the mixed block into a full or half exam under exam conditions.

The error log. Every missed or guessed question goes into a log with four fields: the topic, the specific fact or judgement missed, why you got it wrong (knowledge gap, misread vignette, algorithm error, fatigue), and the rule you will use next time. Review the log weekly; it is the highest-yield revision in the last month. Never copy or photograph live exam content; the log should be written in your own words.^[4]

Progress measures. Track questions completed, mixed-block percentage correct over time, and sleep, rather than hours sat at a desk. A single mock exam early in preparation is useful for building the gap list; late mocks track readiness. Do not chase a perfect score on a single bank; no bank is the exam.

• Start from the blueprint, not from the bank's coverage. Work through the official outline (USMLE Step content outline, the MLA content map, your school's assessment map) and let it set the order.^[1][7]
• Mix subjects early. Interleaved practice feels worse but retains better.^[9]
• Time from the start. Untimed success creates false confidence. Once a topic has been met once, move to timed sets.
• Review aggressively. Spend at least as long reviewing an answer as you spend generating it. The learning occurs in the review, not in the click.^[11]
• Write your own reminders. Turn each miss into one sentence in your own words; never store or share live exam items, which is both an integrity breach and grounds for score cancellation.^[4]
• Do not question-hunt. Reading "high-yield" lists without self-testing produces recognition, not recall.^[11]

An OSCE rewards routine. You will not have time to improvise a whole examination technique, so build one and overlearn it:

• Learn one method per system and say it out loud while you do it, from the first day of practice. An examiner can only credit what is audible.
• Give a running commentary: "looking for X, which would suggest Y". State what you are excluding, not only what you find.
• Time management: most stations are short. Rehearse until the full sequence fits with time to spare, and learn your personal shortcut phrases.
• Professionalism stations: practise introductions, consent, confidentiality, and breaking bad news with a peer and a real checklist, not in your head.
• Examiner sheets: ask your school for the station marking domain sheets (history, examination, communication, diagnosis, management) and practise against them.
• On the day: read the door note carefully, note the task, introduce yourself, wash hands, and finish with a summary and a safe plan. A calm, structured candidate beats a rushed, brilliant one.

• No paid bank is mandatory. A well-managed paper question set from your school, a syllabus and a systematic review schedule is enough if used with retrieval practice.^[11]
• Shared resources: past questions released by the school or society are legitimate; do not copy items recalled from live exams or paid private banks without permission.^[4]
• Internet is uneven: download the official blueprints and guidelines when connected, and work offline from them.
• International graduates (PLAB, USMLE, Canadian route): plan extra time for unfamiliar systems (consent styles, prescribing conventions, documentation), and use supervised clinical observerships if possible before the clinical exam.^[6]
• Language and culture affect communication stations; practise in the language the exam will use, and ask for accent-neutral feedback.
• Sleep, movement and nutrition cost nothing and change everything.^[8]

• Isolation and despair: long projects produce them. Study with a small group, keep one non-study contact per week, and speak early to student or pastoral services if mood, sleep or substance use is slipping.
• Burnout hides as "just need a harder plan". If another two hours does not help, the answer is usually recovery and a better plan, not more hours.
• No mock progress over six weeks despite honest effort: get an external diagnosis of your study method from a teacher or learning adviser; many students are using a perfect system on the wrong content.
• Integrity traps: copying, proxy testing, sharing recalled items, or buying "exam feedback" all put registration and career at risk, and registrants have been cancelled for them.^[4]
• Health problems: do not sit a high-stakes exam significantly unwell or while on sedating medication without an accommodation plan; official exam bodies run accommodation and deferral processes that must be used in advance.^[4]

• Licensing and final examinations reward applied, timed clinical reasoning, not recognition of facts.
• Step 1 is a one-day exam of at most 280 items arranged in 60-minute blocks before 14 May 2026, or fourteen 30-minute blocks from that date; Step 2 CK is a one-day exam of up to 318 items; Step 3 is the final licensing exam.^[1][2][3]
• PLAB part 1 tests 180 SBA questions; part 2 tests 16 OSCE scenarios (18 stations including rest stations); the UKMLA tests 200 SBA questions across two papers plus a clinical skills assessment.^[5][6][7]
• The official Bulletins, guides and school handbooks are the only sources of eligibility, attempt limits and registration rules.^[4]
• Spacing produces clear benefits over massed study once retention is tested beyond a month.^[9][15]
• Retrieval practice (testing yourself) beats re-reading for long-term retention, even when initial performance is poor.^{[11][15] [10]}
• Sleep is a study resource: adults generally need 7 to 9 hours a night.^[8]
• Interleaving subjects feels harder but separates topics better than blocked study.^[9]
• Spaced education in clinical settings improved knowledge retention in a randomised trial.^[12]
• Progress should be measured by questions, mocks and sleep, not by hours at a desk.
• OSCEs require rehearsed, audible, timed performance with feedback, not reading.
• Exam integrity is absolute: never create or share live exam content.^[4]
• Plan backwards from the exam date with a week of slack, and protect a rest day each week.

• Medical professionalism and ethics (plain text; see MedGrasp)
• Biostatistics and screening for medical students (plain text; see MedGrasp)
• Stress, burnout and mental health in students and doctors (plain text; see MedGrasp)
• Clinical communication and breaking bad news (plain text; see MedGrasp)
• Safe prescribing and practical procedures (plain text; see MedGrasp)