NDEB AFK common mistakes
NDEB AFK · Understand the traps and how to avoid them.
Memorising isolated basic-science facts (a single enzyme, a single nerve, a single drug side effect) and answering items by recognising a keyword rather than reasoning from mechanism to consequence.
Most candidates prepare for the AFK the way they prepared for undergraduate recall exams; the AFK instead gives short clinical vignettes that require a two-step link (structure/mechanism → patient finding). Keyword recognition feels productive during study and gives a false sense of mastery.
What to do instead
Convert every high-yield fact into a written 'because' chain - e.g. 'loop of Henle ascending limb is impermeable to water BECAUSE it uses Na/K/2Cl to dilute tubular fluid, THEREFORE loop diuretics cause hypokalaemia and ototoxicity'. Cover the answer, say the chain aloud, and only count the fact as learned when the chain is complete without prompts.
Choosing the first answer option that is factually true instead of the single best answer among several true options.
Much of the exam is written so that three or four options are technically correct; candidates who were trained on single-best-fact MCQs stop evaluating once they find something they recognise as correct, and under time pressure the first true option becomes the answer.
What to do instead
Apply a two-pass rule to every item: pass one, strike out options that are false; pass two, rank the survivors against the exact wording of the stem ('most likely', 'initial', 'most appropriate', 'best'). If two options both survive, the discriminator is almost always safety, urgency, or the least invasive step - choose that one and note why the other is second-best.
Answering a 'what is the next step / initial management' item with the definitive treatment instead of the immediate next step.
Candidates hold the diagnosis in mind and jump to the treatment they associate with it; they do not separate the sequence of confirm diagnosis → stabilise → definitive care. The stem's qualifier ('next', 'initial', 'first') is read but not operationalised.
What to do instead
For each major clinical topic, write the management pathway as an ordered list with the diagnostic step before the therapeutic step (e.g. incisional biopsy before excision; radiograph before extraction; pulp testing before restoration). Before answering, explicitly state which rung of the ladder the stem is asking for.
Diagnosing from the first two lines of a vignette using a memorised 'classic' picture, without reading the detail that makes the presentation atypical.
Illness scripts built from textbooks are triggered by a single salient cue (age, one symptom, one location); the remaining text is skimmed. Exam writers deliberately include a discriminating detail late in the stem - a duration, a bilaterally, a negative finding, a medication.
What to do instead
Read the stem twice: first for the overall picture, second specifically hunting for the one detail that does not fit the obvious diagnosis. Before looking at options, state in one sentence what the patient has, what else it could be, and which single finding rules the alternative out.
Mishandling negatively worded stems ('EXCEPT', 'least likely', 'is NOT') and selecting the option that is true.
Positive stems dominate study material and real-world thinking, so the brain pattern-matches to 'find the correct statement' before the negative cue is registered; this error clusters in the last third of the exam when reading speed increases.
What to do instead
Physically mark the operative word the moment it appears, then silently restate the stem as a positive instruction before reading the options ('I am looking for the four that are true and the one that is false'). Re-check the marked word after selecting.
Treating anatomy, histology and embryology as lists of names rather than three-dimensional structures, so items describing a nerve's course, a fascial space, or root/canal morphology cannot be visualised.
These subjects are usually revised from tables and lecture notes because they are faster to re-read than to redraw; the AFK tests spatial relationships (spread of odontogenic infection, nerve injury site and resulting deficit, referred pain).
What to do instead
Redraw from memory: cross-sections of the neck and floor of mouth, the course of the trigeminal divisions with the branches at each landmark, and root/canal anatomy per tooth. Then trace a pathway (e.g. spread from a mandibular molar to the submandibular, buccal and lateral pharyngeal spaces) on the blank diagram and label the resulting clinical sign at each step.
Answering biostatistics and epidemiology items by recalling a formula from memory rather than constructing the data table, leading to mix-ups between incidence and prevalence, sensitivity and positive predictive value, and absolute and relative risk.
These items are a small fraction of the exam and are under-rehearsed; formulas are remembered as strings of letters without the underlying concept, and prevalence is mentally treated as a nuisance variable rather than the driver of predictive value.
What to do instead
Draw the 2×2 table for every screening/diagnostic question before doing arithmetic, and always label which margin is disease status and which is test result. Practise the same test characteristics at two different prevalences until the direction of the PPV change is automatic. Similarly, label every risk figure as absolute or relative in words before answering.
Answering ethics, consent, capacity, confidentiality and communication items from personal opinion or 'what I would do' rather than from the professional and legal standard.
Behavioural science is often assumed to be common sense and is therefore not studied as a discrete content area; candidates substitute their own values or their memory of a single clinical encounter for the regulatory framework.
What to do instead
Build a short written framework for each topic - elements of valid informed consent, the test for capacity, when confidentiality may be breached, mandatory reporting of child abuse, consent in minors, and the principle that a competent adult may refuse treatment. Answer each item by naming the governing principle first, then selecting the option that respects patient autonomy and the legal duty.
Integrating basic science and clinical dentistry poorly - answering local anaesthetic, antibiotic and drug-interaction items from memorised lists without connecting mechanism to contraindication and to the dental procedure at hand.
Pharmacology and pathology are revised in isolation as 'basic science', so when the same content appears inside a clinical scenario the candidate cannot retrieve it in the reverse direction (patient factor → mechanism → dental risk).
What to do instead
Build two-column tables for high-yield drugs: mechanism/route/half-life on one side, dental implication on the other (maximum dose by weight, vasoconstrictor interaction, bleeding risk with warfarin/DOACs, bisphosphonate-related osteonecrosis risk). Rehearse each row by starting from the patient factor and working forwards to the chairside decision.
Answering prevention, health-promotion and dental-public-health items from clinical intuition instead of the guideline-based or population-level concept being tested (levels of prevention, risk-based recall intervals, water fluoridation, screening criteria, study design).
This domain feels less 'dental' and is deprioritised during revision; candidates therefore answer at the individual-patient level when the item is asking about a population measure, or misclassify primary, secondary and tertiary prevention.
What to do instead
Write out and memorise the short canonical lists: levels of prevention with a dental example of each, criteria for a screening programme, strengths and weaknesses of cohort versus case-control versus RCT designs, and the guideline rationale for risk-based recall rather than fixed intervals. Answer these items by naming the category first, then choosing the option that matches it.