How to Dissect a UKMLA SBA: Technique Masterclass

Most candidates prepare for the UKMLA by revising content. Few prepare by revising how to answer the question. That is the lever that separates a 60th-percentile score from a 90th-percentile score when the two candidates know roughly the same medicine. Single best answer (SBA) items reward technique as much as knowledge: the same facts, read by two different candidates, produce two very different mark totals.

This pillar is the strategy masterclass. Every heading below is a decision-making habit that, once internalised, adds percentage points to your mock scores within a week. Read it once. Apply it to 50 SBAs. Read it again. That is the routine that compounds.

---

1. Anatomy of a UKMLA SBA — stem, lead-in, options

Every SBA has three components:

1. Stem — the clinical vignette. History, examination, investigations.
2. Lead-in question — what the examiner is actually asking.
3. Options — typically five (A–E), one best answer.

The only part the examiner cares about is the lead-in. The stem contains the evidence you need; the options contain the choices. But the lead-in decides which part of the stem matters.

A stem describing chest pain, shortness of breath, elevated troponin, and ECG changes could be asked:

• "What is the most likely diagnosis?" → NSTEMI.
• "What is the next best investigation?" → coronary angiography.
• "What is the first-line definitive management?" → dual antiplatelet + fondaparinux + angiography within 72 h.
• "Which is the most appropriate discharge medication?" → ACEi + beta-blocker + high-intensity statin.

Same stem. Four different correct answers. The lead-in is the question.

Practical habit: read the lead-in before you read the stem. This is non-intuitive at first but becomes natural. It tells you what to look for.

---

2. Lead-in taxonomy: "most likely", "next best", "first-line", "definitive"

Eight recurring UKMLA lead-in types — learn the differences:

Traps:

• "Most appropriate next" ≠ "first-line". If the stem says "despite 6 months of ACEi, BP still 160/100" → next is add-on therapy, not "restart ACEi".
• "Immediate" vs "initial" — immediate means first 5 minutes (ABCDE, resuscitation); initial means first sensible step once stabilised.
• "Definitive" vs "first-line" — Graves' disease first-line = carbimazole; definitive = radioiodine or thyroidectomy.

Underline the lead-in physically on paper or mark it mentally every time.

---

3. Option analysis — elimination + pattern recognition

The UKMLA's five options usually break into three categories:

• Clearly wrong (2–3 options) — easy to eliminate.
• Plausible but suboptimal (1–2 options) — the distractors you must defeat.
• Correct (1 option) — usually aligned with UK guideline.

Elimination sequence:

1. Eliminate the clearly wrong on first pass. Don't read them carefully — just flag them out.
2. Compare the remaining 2–3 against the lead-in's exact wording.
3. Apply the UK-guideline filter (NICE, BNF, Royal College) to differentiate "plausible" from "correct".
4. Choose.

Pattern-recognition shortcuts (built by question practice):

• Any stem with "40-year-old man with chest pain radiating to arm + troponin rise" → ACS ladder.
• Any "meningococcal rash" stem → benzylpenicillin now.
• Any "Addisonian crisis biochemistry" stem → hydrocortisone before workup.

Pattern recognition is how experts read SBAs. It feels like guessing but is compressed reasoning. Build it by seeing thousands of questions — our UKMLA Question Bank Showdown walks through the major UK banks.

---

4. Red-herring detection — statistical vs clinical distractors

The UKMLA plants distractors in two flavours:

Statistical distractors — options that are true for most patients but wrong for this one.

• Example: "60-year-old man with chest pain, family history of MI" — a naïve candidate picks "ACS". But the stem mentions "sharp pain worse on inspiration, friction rub" → pericarditis.

Clinical distractors — options that are almost right.

• Example: asthma severity — "severe" and "life-threatening" are both present if SpO₂ <92% — but life-threatening needs "silent chest" or exhaustion. If only SpO₂ and PEF are given → severe. If silent chest mentioned → life-threatening.

Guarding against red herrings:

• Note demographics last — they bias you but rarely decide the answer alone.
• Look for one red flag in the stem that shifts diagnosis — "sudden", "worst", "post-op", "returned from travel", "pregnant".
• If two options seem equal, the one that better explains the entire stem wins.

---

5. Time strategy — 90 seconds per question benchmark

The UKMLA AKT contains 150 items in 2 sessions (approx 75 items × 2 × 2 hours each). That is roughly 96 seconds per item.

The rule: never spend more than 90 seconds on first read.

Pacing heuristic:

• Easy (30–45 sec): classic pattern, clear answer. Move fast.
• Medium (60–90 sec): careful reading, elimination. Commit an answer.
• Hard (>90 sec): flag and move on. Return at the end.

Why this works: the hardest questions are often worth one mark each, same as easy ones. Wasting five minutes on a hard question costs you three certain marks on easy questions you didn't reach.

---

6. Flag-and-return vs guess-and-move

Two defensible strategies for hard questions:

Flag-and-return:

• Pick a best-guess answer.
• Flag the question.
• Return at the end with fresh eyes.

Guess-and-move:

• Spend no longer than 20 seconds.
• Commit to a calibrated guess.
• Move on and do not revisit.

Evidence from testing psychology (Bar-Hillel, Budescu, etc.) suggests first instincts are right more often than candidates believe. The risk of changing an answer on review is high if the candidate has fatigued cognitive capacity. Use review time to catch obvious misclicks, not to re-reason under fatigue.

Rule of thumb: change an answer only if you find a new fact you missed on first read. Never change because of "it doesn't feel right".

---

7. NICE-prioritisation heuristic — when first-line ≠ textbook answer

UK guidelines often differ from North American or international textbook first-line. The UKMLA examiners align with NICE, BNF, BTS/SIGN, Royal Colleges — not international consensus.

Examples where the UK answer diverges:

• T2DM add-on — NICE NG28 says SGLT2 inhibitor for any patient with established CVD, heart failure, or QRISK ≥10% — regardless of HbA1c. Many textbooks default to sulfonylurea or DPP-4i.
• AF anticoagulation — NICE prefers DOACs over warfarin for non-valvular AF. Older sources default warfarin.
• Hypertension — NICE NG136 uses ACE/ARB → CCB ladder by age/ethnicity. North American guidelines order differently.
• Asthma — NICE NG245 (2024) embeds MART/AIR as first-line; older guidance uses SABA monotherapy.
• Chronic pain — NICE NG193 does NOT recommend long-term opioids for chronic primary pain.

Heuristic: when two options both "work", choose the one that a UK NICE guideline explicitly endorses — that is the UKMLA's answer.

For the comprehensive NICE mapping to UKMLA conditions, see our NICE Guidelines pillar.

---

8. "First-line" vs "most appropriate next" traps

This is the single most frequently lost-mark trap.

"First-line" = the drug/investigation recommended to start with in an untreated patient.

"Most appropriate next" = the logical next step given what has already happened in the stem.

Worked example — T2DM:

• "Newly diagnosed T2DM, HbA1c 62 mmol/mol, BMI 30. First-line drug?" → metformin.
• "T2DM on metformin, HbA1c 58 despite 6 months, QRISK 15%. Most appropriate next drug?" → SGLT2 inhibitor.

Two stems, same disease, two different answers. If you missed that the second patient is already on metformin, you'd pick metformin again and lose the mark.

Rule: always read the stem for what the patient is already on, has already had, or has already failed. The answer is the next step after that.

---

9. Multi-step reasoning — when the question hides two steps

Some UKMLA SBAs test two consecutive decisions disguised as one question.

Example:

"A 24-year-old woman presents with a 3-week history of cough, haemoptysis, weight loss and night sweats. She is a recent arrival from Somalia. CXR shows right upper lobe cavitation. What is the next best step?"

Two steps hidden:

1. Recognise TB as the diagnosis.
2. Decide the correct next step in UK TB workup — sputum AFB smear + culture + NAAT, not "start RIPE regimen".

A candidate who jumps to treatment skips the confirmation step. The lead-in "next best" usually means the investigation step, not the treatment step.

Practical rule: for infectious disease or cancer stems, the "next best" step is almost always confirmatory investigation before empirical treatment (except when empirical treatment precedes investigation — meningitis, encephalitis, neutropenic sepsis).

---

10. Common cognitive biases (anchoring, availability, confirmation)

The UKMLA exploits three cognitive biases that affect clinical reasoning:

Anchoring — locking onto the first piece of data. Stem opens with "chest pain" → you anchor on ACS. Miss the "pleuritic, worse on inspiration, post-long-haul flight" → PE is the actual answer.

Availability — over-weighting recent or vivid cases. You revised endocarditis this morning → you see endocarditis everywhere.

Confirmation — once you have a hypothesis, you cherry-pick evidence to support it. Half-read stems trigger this.

Countermeasures:

• Read the whole stem before generating a hypothesis.
• Note the last few lines of the stem — red flags often hide there (travel, pregnancy, immunosuppression, medication history).
• If two features conflict, re-read the whole stem. Don't defend your first hypothesis.
• In mock review, note which questions you got wrong due to bias, not knowledge. Track the pattern.

---

11. How high-scorers use process of elimination

High-scorers don't just pick the correct answer — they prove the others wrong in under 10 seconds each.

Elimination worked example:

"A 45-year-old man presents with RUQ pain, fever, and jaundice. Bilirubin 120, ALT 180, ALP 420. What is the next best investigation?" A. USS abdomen. B. CT abdomen. C. MRCP. D. ERCP. E. Liver biopsy.

• E (biopsy) — wrong in acute obstruction; risks bleeding, no role first-up.
• D (ERCP) — therapeutic, not first diagnostic. Post-confirmation.
• B (CT) — second-line after USS for biliary disease. Radiation dose in a 45-year-old when USS is non-invasive.
• C (MRCP) — excellent biliary imaging but not first if USS available.
• A (USS) — first-line for suspected biliary disease. Dilated ducts + stones ± gallbladder wall thickening. No radiation. Cheap.

Answer: A. Each elimination is a sentence of reasoning, not a gut feeling.

Discipline it: on every question, write one reason each option is wrong before committing. After 50 questions this compresses into seconds per option.

---

12. Using flags + review-end strategy

Flag types:

Flag for review: you committed an answer but want to sanity-check if time allows.

Flag as "unsure": you guessed and want to revisit with new cognitive energy.

Flag as "lookup": you know the answer is close but want to double-check (only useful if BNF/reference is available — in the actual AKT, there is none).

End-of-session strategy:

• Last 15 minutes — review flagged questions in order.
• Check each against the stem, not against your memory.
• Change answers only if you identify a missed fact.
• If you are uncertain between two, your first instinct usually wins.

Do not review unflagged questions. That is a time sink for negligible gain.

---

13. Ten worked SBAs — one per specialty, with reasoning

Cardiology:

"65-year-old, 2 hrs chest pain radiating to jaw, ECG: 2 mm ST elevation II, III, aVF. Next best?"

• Inferior STEMI → primary PCI activation + aspirin 300 mg + ticagrelor 180 mg. Right-sided ECG before GTN.

Respiratory:

"22-year-old asthmatic, SpO₂ 91% RA, PEF 35% predicted, unable to complete sentences. Severity?"

• Severe asthma. Silent chest or exhaustion would make it life-threatening.

Gastro:

"55-year-old cirrhotic, vomiting blood, BP 95/60. First-line next?"

• Terlipressin + ceftriaxone + urgent OGD within 24 h. Blood + FFP + platelets as needed.

Neurology:

"70-year-old, 2 hrs left-sided weakness. CT: no bleed. NIHSS 14. Next?"

• Thrombolysis (within 4.5 h window) + consider thrombectomy if LVO on CTA.

Endocrine:

"Type 1 DM, vomiting, glucose 28, ketones 4.2, pH 7.18, HCO₃ 8. First step?"

• 1 L 0.9% saline over 1 h + fixed-rate IV insulin 0.1 units/kg/h. JBDS protocol.

Renal:

"AKI + K⁺ 6.9, ECG peaked T waves. Immediate?"

• 10 mL 10% calcium gluconate IV — cardiac stabilisation. Then insulin-dextrose, salbutamol, Lokelma.

O&G:

"34-week pregnancy, BP 165/110, headache, +++ protein. First drug?"

• Labetalol (or nifedipine). Magnesium sulphate if seizure or severe.

Psychiatry:

"Man detained under Section 136, refusing assessment. Section duration?"

• Section 136 lasts up to 24 hours (extendable to 36 in some areas) for assessment in a place of safety.

Paediatrics:

"6-week-old, fever 38.5°C, drowsy. Next?"

• Full septic screen + IV cefotaxime + amoxicillin (Listeria cover <1 month). Never "paracetamol and safety-net".

Emergency:

"Anaphylaxis, RR 30, stridor, BP 80/50. Adult adrenaline dose?"

• 500 μg (0.5 mL of 1:1,000) IM into anterolateral thigh. Can repeat every 5 minutes.

Each question above is a pattern. Internalise the pattern, and the answer comes reflexively.

---

14. What to do when you know nothing (calibrated guessing)

Even the best-prepared candidate will meet questions they cannot solve. Calibrated guessing preserves marks.

Heuristics for blind guessing:

1. Avoid extreme-sounding options. UKMLA rarely rewards "do nothing" or "immediate surgery" in ambiguous stems.
2. Prefer conservative management when ambiguous — UK practice is risk-averse.
3. Prefer guideline-consistent options — if one option sounds like something NICE would write, pick it.
4. If two options are opposites, one is usually right — the examiner offered the contrast deliberately.
5. Eliminate the two most extreme options first, then flip a mental coin between the remaining three.

Do not leave blanks. The UKMLA does not penalise wrong answers. Every option has a 20% baseline probability of being right; educated elimination raises that to 33–50%.

After-the-fact review: in mock marking, categorise every wrong answer as:

• Knowledge gap (study it).
• Misread (slow down).
• Bias (note pattern).
• Random unlucky guess (acceptable noise).

Only the first category rewards more revision. The others reward technique.

---

15. Building SBA-technique practice into a study plan

Most candidates under-practise technique. The fix is calendared integration.

Week 1–4 — content phase:

• 50 questions/day, un-timed, with detailed explanation review.
• Focus: knowledge acquisition.

Week 5–8 — speed phase:

• 50 questions/day, timed at 90 seconds each.
• Focus: pace, elimination, flag discipline.

Week 9–12 — mock phase:

• 2 full-length mocks per week.
• Post-mock review: knowledge gaps, biases, misreads logged separately.
• Tailor subsequent content review to knowledge gaps only.

Every day — explanation review:

• For every question you got wrong, write one sentence explaining why the wrong answer was wrong and why the right one was right.
• Re-read your explanation log weekly. Patterns emerge — these are your technique gaps.

The Ultimate 12-Week UKMLA Study Plan structures this. For spaced repetition of recurring patterns, see our Active Recall + Spaced Repetition for UKMLA pillar. For intensive mock cadence in the final month, see the Last-Minute UKMLA Prep guide.

---

Exam-day checklist for SBA technique

Before the exam:

• Sleep 7+ hours. Fatigue kills elimination discipline.
• Eat a protein breakfast. Avoid high-sugar crashes mid-exam.
• Arrive early — reduce stress-induced cognitive load.

During the exam:

• Read the lead-in first.
• Stem second.
• Options third.
• 90-second cap on first pass.
• Flag-and-commit, don't flag-and-leave blank.
• Review flagged only in last 15 minutes.

Psychologically:

• First instinct > second-guessed.
• Don't revisit a question after you've moved on unless flagged.
• Mid-exam slump is normal — push through; do not recalibrate.

---

Summary — five technique reflexes that add marks

1. Read the lead-in first. It decides what in the stem matters.
2. Eliminate before selecting. Prove each option wrong before you commit.
3. 90-second pace. Flag anything that takes longer.
4. NICE-consistent answer. When in doubt, choose the UK guideline option.
5. First instincts beat second-guesses. Change only when you find a new fact.

SBA technique is learnable, measurable, and cumulative. Candidates who drill it close the gap between knowledge and score. Review this pillar monthly during your 12-week plan, and your mock scores will climb independently of content mastery. That is the compounding advantage.