MDCN Assessment common mistakes
MDCN Assessment · Understand the traps and how to avoid them.
When a clinical vignette ends with a lab value or vital sign that contradicts the initial impression, candidates lock onto their first diagnosis and reinterpret the discordant data as supporting it, rather than reopening the differential.
Vignettes are read in order and the first two sentences often suggest a familiar pattern; once a pattern is recognised, the candidate reads the rest of the stem confirmatorily. Under time pressure, revising a diagnosis feels like losing ground already gained.
What to do instead
Before selecting an answer, state the single finding that is least compatible with your leading diagnosis. If you cannot explain it in one sentence, the diagnosis is wrong or incomplete. Practise by covering the question and options, writing the leading diagnosis plus one discriminating negative, then uncovering the options.
Candidates choose the management option that is the definitive treatment rather than the immediate next step, answering 'what is the best treatment' when the stem asks 'what should be done now'.
Definitive treatments are more memorable and are over-represented in study material, so they surface first. The stem's wording ('next', 'initial', 'most appropriate immediate') is skimmed rather than parsed.
What to do instead
Underline the interrogative phrase in every stem before reading the options. Sort any management question into one of three buckets: resuscitate/stabilise, investigate, or treat definitively. Only the bucket the stem asks for is eligible, and options from other buckets are distractors by construction.
Candidates treat 'most likely' and 'most appropriate' as interchangeable, selecting the diagnosis they consider most serious rather than the one with the highest probability given the described frequency and typical presentation.
Clinical training rewards vigilance for dangerous diagnoses, so serious-but-rare conditions feel like safer answers. Candidates also confuse prevalence with importance when the stem gives no explicit epidemiological clue.
What to do instead
For 'most likely' questions, rank options by base rate and typicality, not severity. For 'most appropriate' questions, rank by guideline-concordant action. When two options are both plausible, ask which one the stem's specific details were written to support; the extra detail usually points to the answer.
In data-interpretation items, candidates read a single abnormal value in isolation and act on it without checking whether it is consistent with the rest of the panel or whether it is a known artefact.
Reference ranges are memorised as isolated thresholds, and a flagged value draws attention away from the pattern. Candidates also forget that haemolysis, dilution, or timing of sampling can produce spurious results.
What to do instead
For every abnormal value, apply a three-question check: does it fit the clinical picture, does it fit the other analytes on the panel, and could it be pre-analytical? Only act when all three are answered. Learn the common artefact patterns for the analytes that appear in your exam's data questions.
Candidates answer pharmacotherapy questions using the drug they know best rather than the agent indicated by the patient's comorbidities, age, renal function, or pregnancy status described in the stem.
Familiarity drives recall, and the commonest drug for a condition is often the one most practised. Contraindication cues are embedded mid-stem and are easy to read past when the diagnosis is already clear.
What to do instead
Before choosing any drug, extract four patient variables from the stem: age, renal and hepatic function, pregnancy or breastfeeding status, and key comorbidities. Eliminate any option contraindicated by those variables first, then choose among survivors. This converts a recall task into an elimination task.
When a radiograph or clinical photograph description is given in text, candidates diagnose from the history alone and treat the imaging description as confirmatory decoration.
Textual descriptions of images are read quickly and feel less informative than a narrative history. Candidates without confident image interpretation default to the history and then rationalise the imaging findings.
What to do instead
Read the imaging description first, form an independent radiological impression, then read the history and reconcile the two. Where they conflict, the conflict itself is usually the point of the question. Practise describing common findings in words so the vocabulary is familiar before the exam.
In oral medicine and pathology items, candidates classify a lesion by its appearance alone without using the site, duration, and behaviour (progressive, recurrent, resolving) given in the stem.
Lesion names are taught alongside classic appearances, so visual pattern-matching dominates. Duration and behaviour are short phrases in a long stem and are under-weighted relative to the descriptive appearance.
What to do instead
For every lesion question, build a four-line summary before answering: site, duration, behaviour over time, and associated systemic features. Diagnosis must be consistent with all four. Any option that requires ignoring one line is eliminated.
Candidates plan restorative or surgical treatment without first establishing periodontal status, occlusion, or the patient's ability to maintain oral hygiene, and select the option that is technically most elegant rather than the one the patient can sustain.
Treatment planning is taught procedure by procedure, so the presenting complaint dominates. Prognostic and patient-factor cues are perceived as background rather than as determinants of the correct option.
What to do instead
Adopt a fixed planning sequence for every case: disease control first, then reassess, then definitive care, then maintenance. Any option that places definitive restoration or prosthodontics before periodontal control is wrong unless the stem explicitly establishes stability.
Candidates answer local anaesthesia and emergency items by reciting maximum doses and drug names but ignore the patient's weight, the vasoconstrictor concentration, and the volume already administered, producing a dose calculation that is internally inconsistent.
Maximum doses are memorised as single numbers without units or weight basis, and calculations are done mentally under time pressure. Mixing mg/kg limits with cartridge volumes is a common source of silent arithmetic error.
What to do instead
Write the calculation in three explicit lines: dose limit in mg, concentration in mg per mL, and volume in mL. Convert to cartridges only in the final step. Always state units at each line. If the stem gives weight, the answer almost always depends on it.
Candidates spend disproportionate time on a small number of difficult items, leaving later straightforward items unanswered or rushed, and change correct answers during a final review without a specific reason.
Sunk-cost thinking keeps candidates on items they have already invested in, and anxiety about unfinished items drives rapid late guessing. Answer-changing is driven by the assumption that a second thought is more reliable than the first.
What to do instead
Set a per-item time cap and a hard rule: if the cap is reached, mark the best current answer and move on. Return only if time remains. On review, change an answer only when you can name the specific fact you misread or misremembered; otherwise keep the original.