PMC Residency Entrance Exam common mistakes
PMC Residency Entrance Exam · Understand the traps and how to avoid them.
Choosing the single best answer by pattern-matching a buzzword (e.g., seeing 'cherry-red spot' and jumping to a diagnosis) without reading the rest of the stem, so the option chosen is the classic answer rather than the one that fits the specific age, sex, and time course given.
Under timed conditions with roughly 100+ single-best-answer items, candidates default to recall of the most familiar association and stop reading after the first recognisable cue; the exam deliberately builds stems where the buzzword is a distractor and a less famous diagnosis fits the full vignette.
What to do instead
Adopt a fixed reading order: read the final question line first, then the last sentence of the stem (the discriminator), then the whole stem, then the options. For every option you reject, state in one silent sentence why it fails a specific stem fact. Practise on 50-item blocks and log each buzzword trap you fall for.
Answering basic-science items from isolated first-year memory (e.g., naming a nerve or enzyme from a diagram-style description) without connecting it to the clinical consequence the stem is actually testing, such as the deficit pattern produced by a lesion at that level.
Anatomy, physiology, and biochemistry are revised as standalone lists during undergraduate years, and candidates do not rehearse them in a clinical frame, so a question that asks 'which structure is at risk and what happens' is answered with a name rather than a functional prediction.
What to do instead
Re-study each region or pathway as a three-column table: structure/enzyme, function, and the clinical deficit when it fails. Then answer at least 20 integrated items per system and force yourself to write the deficit before looking at the options.
Managing obstetrics and gynecology vignettes by choosing the intervention for the named condition while ignoring the gestational age, hemodynamic status, or fetal condition stated in the stem, so a correct drug or procedure is selected at the wrong point in the pathway.
Labour ward and emergency protocols are memorised as condition-to-treatment pairs, but the exam tests sequencing: the same diagnosis has different first steps at 8 weeks, 28 weeks, and 36 weeks, and in a stable versus unstable patient.
What to do instead
Before answering any obstetric item, write the gestational age and maternal stability on your scratch paper, then ask 'what is the immediate next step, not the definitive treatment'. Drill ectopic, pre-eclampsia, postpartum haemorrhage, and preterm labour as decision trees with gestational-age branches.
Treating paediatric questions with adult dosing, adult fluid calculations, or adult vital-sign thresholds, and forgetting that the same disease has age-specific presentations and different first-line drugs in neonates, infants, and older children.
Most clinical exposure and most revision resources are adult-oriented, so weight-based dosing, maintenance fluid rates, and normal ranges for age are the least rehearsed numbers, and candidates approximate them from adult figures.
What to do instead
Build a one-page paediatric card: weight-based drug doses for the ten most tested drugs (paracetamol, gentamicin, ceftriaxone, adrenaline, dexamethasone), 4-2-1 maintenance fluids, and age-specific normal respiratory and heart rates. Re-answer every paediatric item against that card until the numbers are automatic.
Answering community medicine items with clinical intuition instead of the epidemiological definition being tested, for example confusing incidence with prevalence, sensitivity with positive predictive value, or a cohort with a case-control design.
These terms look familiar and candidates assume they know them, but the exam tests precise operational definitions and the effect of prevalence on predictive value; clinical reasoning pulls toward 'who has the disease' rather than 'what the measure describes'.
What to do instead
Write out the 2x2 table for every screening and study-design question before choosing, and derive sensitivity, specificity, PPV, and NPV from it. Memorise one worked example each of cohort, case-control, and cross-sectional design, and check whether the stem describes exposure-to-outcome or outcome-to-exposure direction.
Handling psychiatry and behavioral science items by choosing the most humane-sounding option or the most familiar diagnosis, without applying the stated duration criteria, functional impairment, and exclusion of substance or medical causes that the classification requires.
Candidates rely on prototype impressions of mental illness and are uncomfortable with criteria-based reasoning; duration thresholds (two weeks for depression, six months for schizophrenia spectrum, one month for brief psychotic disorder) and the medical-cause exclusion are the discriminating facts and are easily skipped.
What to do instead
For each major category, memorise the duration criterion, the required symptom count, and the mandatory exclusion in a three-line summary. On every item, explicitly check the timeline given in the stem against that criterion before selecting, and eliminate any option that ignores a stated medical or substance cause.
Misanswering forensic medicine and ethics items by giving the morally preferred answer rather than the legally correct one, for example confusing culpable homicide with murder, or consent in a minor or an unconscious patient with implied consent in an emergency.
Ethics and law are often revised last and superficially, and candidates substitute personal values for jurisdiction-specific legal definitions; the wording of options is deliberately close and hinges on a single legal element such as intention or age of consent.
What to do instead
Tabulate the legal definitions side by side (murder, culpable homicide, causing death by negligence, dowry death) with the one element that separates each. For consent, learn the four valid elements and the exceptions for minors, emergencies, and mental incapacity, then answer 30 ethics items and justify each rejection with the missing legal element.
Approaching surgery items by naming the operation for a diagnosis while skipping the resuscitation, investigation, or timing step the question asks for, and missing when conservative management or a damage-control approach is the correct answer.
Surgical revision centres on procedures and complications, so the preoperative phase and the decision not to operate are under-rehearsed; exam stems often ask for the next step in a patient who is not yet fit for theatre.
What to do instead
For every major surgical presentation (acute abdomen, obstruction, trauma, hernia, thyroid, breast lump), write the pathway as resuscitation, investigation, decision, operation, complication. Practise items that end in 'most appropriate next step' and check whether the patient is stable before choosing an operation.
In anesthesiology and radiology items, selecting the finding or technique named in the stem's first clause without checking the contraindication, the safety threshold, or the imaging modality that actually answers the clinical question, such as choosing contrast CT in a patient with a stated renal or allergy history.
These specialties are taught as lists of signs and drugs, and candidates do not rehearse the safety filters (airway grading, Mallampati, ASA class, contrast contraindications, radiation in pregnancy) that the exam attaches to otherwise straightforward facts.
What to do instead
Build a contraindication column next to every technique and drug you revise: suxamethonium and hyperkalaemia or burns, spinal anaesthesia and coagulopathy or raised intracranial pressure, iodinated contrast and renal impairment or prior reaction, MRI and pacemakers. On each item, scan the stem for the stated comorbidity before choosing the modality or agent.
Spending uniform time on every question and leaving no reserve, so the last 20 items are answered in a rush and multi-step calculation or image-description items are abandoned or guessed.
Candidates practise untimed or self-paced, never rehearse pacing, and cannot judge when an item is genuinely hard versus merely long; the resulting time debt hits the highest-yield multi-step questions hardest.
What to do instead
Simulate the real format at least four times: fixed item count, fixed clock, no pauses. Set a per-item ceiling (total time divided by item count), mark and move on when exceeded, and reserve the final ten minutes for marked items. Review your log to see which domains cause overruns and pre-decide to skip those item types on first pass.