ABD Dermatology common mistakes
ABD Dermatology · Understand the traps and how to avoid them.
Committing to a single diagnosis and stopping the differential once one plausible entity fits the morphology, rather than explicitly ranking alternatives and naming the finding that discriminates them.
Under time pressure the first pattern match feels sufficient, and exam stems often describe a classic entity; candidates therefore anchor and skip the disconfirming step that the question is actually testing.
What to do instead
For each clinical vignette, write the top three differentials and one discriminating feature (distribution, time course, mucosal involvement, KOH result) before selecting an answer; choose the option that the discriminator supports, not the first one recognised.
Describing lesions with vague nouns (rash, spot, bump) instead of a structured morphological description using primary lesion type, colour, border, distribution, and arrangement.
Trainees read images or stems quickly and translate directly to a diagnosis, so the descriptive vocabulary is never rehearsed and remains imprecise.
What to do instead
Practise a fixed descriptive template (primary lesion, secondary change, colour, shape/border, arrangement, distribution, mucosal/nail/palmoplantar involvement) on every practice stem until it is automatic.
Ignoring the drug and travel/occupational history in the stem and attributing a morbilliform or annular eruption to an idiopathic dermatosis.
History-taking is deprioritised as 'soft' relative to morphology, and candidates do not habitually run a medication timeline against rash onset.
What to do instead
Always construct a medication timeline (start date versus rash onset, 1-6 weeks for most drug eruptions) and ask about recent travel, new pets, and occupational exposures before finalising the diagnosis.
Applying management steps without first establishing severity, extent, or the presence of systemic involvement, so a mild localised process is over-treated or a severe one is under-treated.
Candidates memorise first-line therapies by disease name and skip the staging step that determines whether topical, systemic, or emergency management is appropriate.
What to do instead
Before choosing therapy, classify severity explicitly (body surface area, mucosal involvement, systemic symptoms, functional impairment) and let that classification select the treatment tier.
Failing to recognise dermatologic emergencies (e.g., Stevens-Johnson syndrome/toxic epidermal necrolysis, erythroderma, angioedema with airway risk, necrotizing soft tissue infection) and treating them as routine outpatient problems.
Emergency presentations are rare in routine clinics, and early findings are subtle and easily filed under a benign descriptor.
What to do instead
Maintain a short emergency checklist (mucosal erosions, skin sloughing/nikolsky sign, fever with rash, rapid progression, facial or airway swelling) and screen every stem against it before answering.
Reading a histology description and naming a diagnosis from the overall pattern without first identifying the level of the pathology (epidermis, dermoepidermal junction, dermis) and the dominant inflammatory cell type.
Pattern recognition is trained on classic images, so candidates jump to a diagnosis and then back-fill the features instead of building the diagnosis from the observed findings.
What to do instead
Use a stepwise algorithm: identify the primary site of injury, the dominant cell type, and the presence or absence of epidermal change; then match to a pattern category before selecting the specific entity.
Treating a single biopsy as definitive and not accounting for sampling error, stage of evolution, or the effect of prior treatment on the histology.
Candidates assume the biopsy represents the whole lesion and the whole disease course, and do not consider that an evolving or partially treated lesion may show non-diagnostic changes.
What to do instead
When a histologic description is ambiguous or partially treated, state explicitly that clinicopathologic correlation and re-biopsy may be required, and look for the option that reflects that limitation.
Applying adult dosing, adult treatment algorithms, or adult morphological expectations directly to infants and children without adjusting for age-specific physiology and presentations.
Most training exposure is adult-focused, and paediatric presentations of common diseases (e.g., atopic dermatitis distribution, psoriasis variants, vascular anomalies) differ enough to mislead pattern matching.
What to do instead
For any paediatric stem, first anchor the age and developmental stage, then recall the age-specific differential and weight-based dosing before selecting an answer.
Missing a congenital or inherited condition because the stem presents a single isolated finding and the candidate does not ask about family history, consanguinity, developmental milestones, or associated systemic features.
Genodermatoses present with common individual findings, and the syndromic link is only apparent when the associated features are actively sought.
What to do instead
When a paediatric stem describes an unusual lesion or multiple findings, systematically check for family history, growth and developmental concerns, and known associated organ involvement before concluding it is isolated.
Failing to integrate the clinical history with the histopathology and instead answering from whichever of the two is more familiar, producing a diagnosis that satisfies one data source but contradicts the other.
Subspecialty questions deliberately present a clinical picture and a biopsy description that must be reconciled, and candidates default to their stronger area rather than resolving the conflict.
What to do instead
Explicitly list the clinical features and the histologic features side by side, identify any conflict, and choose the entity that explains both; if none does, select the option that recommends further clinicopathologic correlation.