AMC CAT MCQ common mistakes
AMC CAT MCQ · Understand the traps and how to avoid them.
Committing to the first plausible diagnosis in a clinical vignette and then selecting the option that matches it, without reading the remaining options to check for a better fit.
Under time pressure the brain pattern-matches the salient cue (for example 'sudden severe headache' triggering subarachnoid haemorrhage) and stops searching; the exam deliberately includes a near-miss option that fits the headline feature but not the discriminating details.
What to do instead
Adopt a forced routine: after forming a hypothesis, read every option and state to yourself one feature in the stem that each option fails to explain. Only then select. If two options survive, return to the stem for the single discriminating finding (time course, age, laterality, investigation result).
Answering management questions with the diagnosis rather than the next step, or jumping to definitive treatment when the stem asks for the immediate action.
Candidates conflate 'what is the most likely diagnosis' with 'what should be done next'; the stem's final interrogative is read quickly and the diagnosis option is chosen because it was just derived with confidence.
What to do instead
Underline the final sentence of each stem before reading options and classify it as diagnosis, next investigation, immediate management, or definitive management. For management items, apply the ABC/resuscitation-first hierarchy and reject any option that is definitive therapy before stabilisation or confirmation.
Misreading negative and absolute qualifiers such as 'EXCEPT', 'most appropriate', 'least likely', 'initial', and 'contraindicated', and answering the opposite question.
These words are short, often capitalised mid-stem, and are skipped during rapid scanning; the options frequently include the correct answer to the un-negated question as a trap.
What to do instead
Circle or mentally flag every qualifier on first read and restate the task in your own words before looking at options. On EXCEPT items, actively tick the three true statements and select the remaining one rather than hunting for a false statement directly.
Applying adult physiological thresholds and drug doses to neonates, infants, and children.
Most prior clinical exposure and revision is adult-weighted, and normal ranges for heart rate, respiratory rate, blood pressure, and haemoglobin differ markedly in children; neonatal jaundice and weight-based prescribing are frequent sources of error.
What to do instead
Before answering any paediatric item, explicitly recall age-specific norms (for example term neonatal bilirubin thresholds by hour of age, paediatric tachycardia cut-offs by age band) and confirm that any drug option is expressed per kilogram with the correct maximum. Treat every paediatric stem as a prompt to check age first.
Ignoring the discriminatory value of time course and symptom chronology when differentiating obstetric and gynaecological emergencies.
Ectopic pregnancy, miscarriage, molar pregnancy, placental abruption, and uterine rupture share pain and bleeding; candidates select on the presence of bleeding alone and overlook gestation, pain character, and haemodynamic trajectory.
What to do instead
For every obstetric/gynaecological bleeding or pain stem, build a four-column mental table (gestation, pain type, bleeding pattern, haemodynamic status) and eliminate options that conflict with any column. Always check whether a pregnancy test or ultrasound finding is given and treat it as decisive.
Selecting an investigation that is less appropriate for the acuity of the situation, such as ordering an outpatient test in an unstable patient or a radiation-heavy test as first line in a young person.
Candidates recall the 'best' diagnostic test in ideal conditions rather than the best next test for this patient; they also forget that unstable patients need bedside or rapid tests and that pregnancy must be excluded before irradiating.
What to do instead
Ask two questions before choosing any investigation: is the patient stable, and does the result change immediate management? Unstable patients get bedside tests (ECG, bedside ultrasound, blood gas, pregnancy test). Stable patients get the test with the highest sensitivity/specificity for the leading diagnosis.
Mixing up which antibiotic, anticoagulant, or fluid is first line because the options are all familiar drug classes, and choosing on name recognition rather than on indication.
Pharmacology revision is often by drug list rather than by clinical scenario, so candidates recognise all five options and pick the one most recently seen; spectrum, route, and contraindication are not systematically checked.
What to do instead
For each drug option, silently state the indication, the key contraindication, and the monitoring parameter. Eliminate any option contraindicated by a stem feature (renal impairment, pregnancy, allergy, breastfeeding) before comparing the remainder on spectrum or guideline first-line status.
Overlooking a single abnormal observation buried in a set of normal vital signs or laboratory values, and treating the case as routine.
Normal values create a false sense of reassurance and the eye skips repeated numbers; a lone tachypnoea, rising creatinine, or low platelet count is often the intended key to the answer.
What to do instead
Scan the observation and results block line by line against normal ranges before reading the question, and note any value outside range. If the stem seems bland, assume you have missed an abnormal value and re-read the numbers.
Choosing an answer based on a remembered guideline from a different country or an outdated version, particularly for screening intervals, vaccination schedules, and hypertensive management in pregnancy.
Guidelines differ between bodies (for example NICE versus ACOG versus WHO) and candidates recall the version encountered in a previous rotation or resource without noticing the exam's framing.
What to do instead
Learn the guideline source used by the exam and anchor on principles that are stable across guidelines (for example magnesium sulfate for eclampsia prophylaxis, delivery as definitive treatment for pre-eclampsia). Where guidelines genuinely diverge, prefer the answer reflecting physiological safety rather than a specific numeric threshold.
Running out of time because long stems are read in full before the question is identified, leaving later items answered in the final seconds.
Clinical vignettes are deliberately verbose and candidates read them linearly, so cognitive effort is spent on irrelevant history; the last items then receive rushed pattern-matching.
What to do instead
Read the final question sentence first, then scan the stem for the data that question requires, and allocate a fixed time per item with a hard cut-off. Flag and move on from any item exceeding twice the average, returning only if time remains, and never leave blanks since there is no negative marking.