OCANZ COE common mistakes
OCANZ COE · Understand the traps and how to avoid them.
In the written paper, answering management questions with a diagnosis only, e.g. writing 'bacterial conjunctivitis' when the question asks what you would do next, and omitting the specific drug, concentration, dosing frequency, duration and review interval.
Candidates revise from diagnostic pattern-recognition material and treat management as an afterthought, so they can name the condition but have never rehearsed writing a complete, prescriber-level plan under time pressure. Markers award discrete points for each management element, so a correct diagnosis with no plan scores poorly.
What to do instead
Build a one-page prescribing template for each condition class you may be asked about (bacterial conjunctivitis, blepharitis, allergic eye disease, dry eye, anterior uveitis, acute angle closure) listing first-line agent, concentration, frequency, taper or duration, review timing, and red flags requiring referral. Practise writing full plans to a strict time limit and self-mark against the template until every element appears without prompting.
Interpreting a visual field printout or OCT report by reading the summary indices and the machine's flag text first, then deciding the diagnosis, rather than first checking reliability indices and then tracing the actual defect pattern.
Printouts are designed to draw the eye to coloured probability symbols and global indices, and candidates under time pressure default to the most visually salient information. This produces answers that describe a 'field defect' or 'thinning' without localising it, and misses unreliable tests that should be discounted.
What to do instead
Adopt a fixed reading order for every printout: reliability (fixation losses, false positives, false negatives), then pattern deviation plot, then whether the defect respects the vertical or horizontal meridian, then the global indices last. Practise on at least twenty de-identified printouts, verbalising the sequence aloud, until the order is automatic.
Giving a single diagnosis for a case that is designed to have more than one concurrent problem, such as a patient with reduced acuity, glare and a cataract plus macular changes, and never listing the differential or the ranking.
Candidates treat cases as single-answer puzzles and stop searching once they have found one abnormality that explains the chief complaint. Exam cases are often deliberately constructed with a co-existing finding that changes management, and the question wording ('list your differential diagnosis in order of likelihood') is skimmed.
What to do instead
Before writing, underline every abnormal finding in the vignette and check each one is accounted for by your answer. Always structure the response as ranked differential with the evidence for and against each option, then a separate line for incidental findings and how they alter the plan. Practise on cases with two or more findings until listing a ranked differential becomes reflex.
In the clinical examination, performing a test but not recording the actual measurement, writing 'visual fields normal' or 'fundus OK' instead of recording the method, the result and the patient's response.
Candidates practise techniques in isolation and are never drilled on documentation, so under observation they concentrate on the procedure and treat the record as a formality. Examiners mark what is written and what is verbalised, so an unrecorded competent test earns no credit.
What to do instead
Rehearse every station with a spoken running commentary that ends in a stated result, for example naming the target, distance and the acuity achieved for each eye. Write the record immediately after each practice station and have a peer check it against a checklist of required data points. Treat the sentence 'I would now record...' as part of the technique itself.
Skipping or rushing the pre-test history and immediately moving to instrumentation, so the clinical station begins with a slit lamp or ophthalmoscope before any presenting complaint, symptoms, ocular and medical history, or medications have been established.
History-taking feels like talking rather than examining, and candidates worry about filling the station time with visible procedures. In reality the history is heavily weighted and it also supplies the reasoning that later stations are marked on, so omitting it costs marks twice.
What to do instead
Use a fixed mnemonic-driven history framework and practise delivering it in under three minutes with a simulated patient. Open every station by stating that you will take a history first, then cover presenting complaint with onset, duration, severity and associated symptoms, past ocular history, general health and medications, family history and occupational visual demands. Time yourself until the framework is complete and fluent.
Performing a procedure with poor patient communication and no consent, explanation or warning, for example instilling drops or bringing the slit lamp towards the patient without telling them what is happening or what they will feel.
Candidates focus entirely on the technical steps they are being marked on and treat communication as an optional extra. Under examination stress, attention narrows to hand technique, and the professional-conduct marks for consent, explanation and comfort are lost without the candidate noticing.
What to do instead
Script a standard preamble for every station: introduce yourself, confirm identity, explain the test and why it is needed, warn about any sensation or light, obtain verbal consent, and check comfort throughout. Practise with a partner instructed to note every occasion you touch the patient or the equipment without a preceding explanation, and repeat until the preamble is automatic.
When a clinical finding is abnormal, continuing the routine sequence as if nothing had happened instead of pursuing the finding with the appropriate supplementary tests and stating the resulting management.
Candidates have rehearsed a linear routine and fear that deviating will lose their place or run out of time. This produces a technically complete examination that never addresses the abnormality, which is precisely what the station was designed to test.
What to do instead
Before each practice station, write down the branch points: for each test, what you would do if the result were abnormal. Drill the follow-on sequences, such as additional tests, dilation, referral urgency and safety-netting advice, for the common abnormal findings in each station type, and practise deliberately triggering the branch by having a partner report an abnormal result mid-routine.
In written clinical-reasoning questions, describing what a test measures instead of interpreting the specific values given, for example explaining the principle of tonometry when the question supplies pressures and asks for an interpretation and plan.
Candidates over-learn theory and under-practise applying it to numbers, and textbook-style revision rewards definitions. Exam questions frequently supply data and mark the interpretation, so a definitional answer scores little even though it is factually correct.
What to do instead
Convert every theory topic into data-interpretation practice: take a measurement type, generate five sets of plausible values spanning normal, borderline and abnormal, and write the interpretation, the differential and the next step for each. Answer past-style questions by first underlining the numbers and the actual question verb in the prompt, then writing only what those numbers imply.
Managing time badly in the written paper by over-writing early long-answer questions until the later, often higher-yield short questions are left blank or answered in note form.
Candidates are anxious about the first question and keep adding detail to feel safe, without any sense of marks per minute. Long-answer questions reward breadth of key points, not length, so extra prose earns nothing while the unattempted questions earn nothing at all.
What to do instead
Before the exam, calculate marks per minute from the paper's structure and rehearse full papers under timed conditions with a visible clock. Adopt a rule of allocating time in proportion to marks and moving on when the allocation expires, leaving space to return. Practise writing bullet-point key phrases rather than continuous prose, since markers look for discrete points.
Answering a clinical station's viva questions with vague hedging such as 'it could be a few things' or 'I would probably refer', without committing to a ranked diagnosis, an urgency category or a named destination for referral.
Candidates fear being wrong and believe hedging protects them, and they have often never been taught the specific referral urgency categories or pathways. Examiners mark the decision and its justification, so an uncommitted answer cannot be credited even if the underlying knowledge is present.
What to do instead
Learn and rehearse the explicit urgency categories used in your jurisdiction (emergency same day, urgent within days, routine) and name the actual destination for each common condition. In practice vivas, force yourself to state a single most likely diagnosis with a confidence level, then the alternatives, then the urgency and destination, and have a partner penalise any answer containing 'probably' or 'maybe' without a stated plan.