MECEE-PG Nepal common mistakes
MECEE-PG Nepal · Understand the traps and how to avoid them.
In physiology and pathology items, answering from a memorized 'textbook normal' value instead of reading the qualifier in the stem (e.g., 'in a patient with chronic renal failure' or 'at high altitude') and adjusting the expected value accordingly.
Candidates rehearse isolated numbers (GFR, PAO2, serum K+, anion gap) and pattern-match the number rather than the scenario, so the modifier clause is skipped under time pressure.
What to do instead
Before choosing, underline every qualifier in the stem (chronic/acute, arterial/venous, fasting/fed, treated/untreated) and re-derive the direction of change (up/down/normal) before matching to an option.
On biochemistry items involving enzyme deficiencies or metabolic blocks, naming the deficient enzyme correctly but selecting the wrong accumulated substrate or the wrong direction of the pathway (e.g., confusing the substrate that accumulates with the product that falls).
Candidates memorize disease names and enzyme names as a single fact but never draw the reaction arrow, so they cannot say which side of the block builds up.
What to do instead
For each inherited metabolic disease, write the one-line pathway with the block marked and label 'accumulates before block / deficient after block' during revision; reproduce that arrow mentally in the exam.
In pharmacology, choosing the drug class that is 'most effective' for a condition rather than the drug that matches the specific contraindication or patient factor stated (e.g., choosing a beta-blocker in an asthmatic, an NSAID in renal impairment, or a tetracycline in pregnancy).
Revision is organized by 'drug of choice for disease X', so safety and contraindication screening is under-trained relative to efficacy recall.
What to do instead
For high-yield drug classes, build a three-column table: indication, absolute contraindication, and monitoring parameter; in the exam, scan the stem for the contraindication keyword before scanning options.
In microbiology, answering organism-identification items from a single clue (e.g., 'gram-positive cocci in clusters') without integrating the second and third clues given (culture medium, oxidase/catalase, hemolysis, growth condition), leading to a plausible but wrong organism.
Candidates learn organisms as one-line associations and stop reading once a familiar keyword appears.
What to do instead
Treat each identification stem as a checklist: collect all clues into a grid (morphology, arrangement, oxygen requirement, key biochemical, classic clinical setting) and eliminate options that fail any one clue.
In internal medicine and surgery items, committing to a diagnosis before reading the final question line, which often asks for the next best step in management, the most likely complication, or the single most useful investigation rather than the diagnosis itself.
The clinical vignette triggers premature closure; candidates diagnose and then pick the option that restates their diagnosis.
What to do instead
Read the last sentence of the vignette first to identify the task (diagnosis vs next step vs investigation vs complication), then read the vignette with that task in mind.
In surgery items, selecting the definitive operative management when the stem describes an unstable patient or the question asks for initial/resuscitative management (or vice versa), and ignoring stated hemodynamic status.
Candidates memorize 'management of X is surgery Y' without the triage layer that determines sequencing.
What to do instead
For each acute surgical condition, rehearse a two-step ladder: immediate stabilization/resuscitation first, definitive procedure second; decide which rung the question is asking about from the wording and vital signs.
In obstetrics and gynaecology items, applying adult non-pregnant reference ranges or drug dosing to a pregnant patient, and failing to date the pregnancy from the stated gestational age before choosing management.
Pregnancy physiology (hemodilution, altered reference ranges, changed drug safety) is learned as a separate chapter and not integrated into general clinical reasoning.
What to do instead
Make 'gestational age and pregnancy status' the first data point extracted from every obstetric stem, then apply pregnancy-specific ranges and pregnancy-category drug rules before evaluating options.
In paediatrics, choosing management based on the named disease while ignoring the age/weight band that determines the answer (neonatal vs infant vs older child), and missing feeding, birth, and immunization history that the stem deliberately supplies.
Candidates treat paediatric items as adult items with a smaller patient and skim over the demographic and feeding details.
What to do instead
Extract age in weeks/months, weight, feeding type, and immunization status into a short list before reading the options; discard any option inappropriate to that age band.
In community medicine, computing an epidemiological measure (incidence, prevalence, sensitivity, specificity, positive predictive value, relative risk, attributable risk) with the wrong denominator, or confusing a measure of association with a measure of impact.
Formulas are memorized as symbols without a fixed habit of identifying numerator, denominator, and study design from the stem.
What to do instead
For every calculation item, write out the 2x2 table or the numerator and denominator in words before substituting numbers, and confirm the study design (cohort vs case-control vs cross-sectional) because it constrains which measure is valid.
Across all domains, spending disproportionate time on a few difficult recall items early and then rushing or leaving blank the later, easier items, with no marking of uncertain answers for review.
Exam-day anxiety plus an unplanned time budget leads to sunk-cost persistence on single items and abandoned marks elsewhere.
What to do instead
Adopt a fixed per-item time cap, answer every item (no blanks where negative marking is absent or known), flag uncertain items for a single review pass, and rehearse this pacing on full-length timed mock tests.