COMT common mistakes
COMT · Understand the traps and how to avoid them.
Reciting the corneal layers in the wrong order or placing a pathology in the wrong layer - e.g., putting Descemet's membrane between the stroma and Bowman's layer, or labelling a corneal ulcer as 'endothelial' when the defect is epithelial.
Candidates memorize the layer names as a five-word list without anchoring each layer to a function or a clinical sign (epithelial defect stains with fluorescein; stromal oedema shows haze; endothelial failure shows microcystic oedema), so the list degrades under exam pressure.
What to do instead
Build a layered diagram from memory twice daily and attach one clinical sign and one disease to each layer (epithelium–abrasion, Bowman's–salzmann/scarring, stroma–keratoconus thinning, Descemet's–keratoconus rupture/striae, endothelium–Fuchs' guttae). Answer anatomy items by recalling the sign first, then the layer.
Tracing the aqueous outflow pathway incorrectly - stating aqueous is produced by the trabecular meshwork, or routing it from the posterior chamber directly into Schlemm's canal without passing the anterior chamber angle.
Aqueous production and outflow are taught in the same lecture, and the terms 'ciliary body' and 'trabecular meshwork' both appear in glaucoma context, so they get swapped.
What to do instead
Memorize the pathway as a single sentence chain with a producer and a drain: ciliary epithelium → posterior chamber → pupil → anterior chamber angle → trabecular meshwork → Schlemm's canal → collector channels → episcleral veins. Practise labelling a gonioscopy diagram so the meshwork is always tied to the angle, not to production.
Transposing a spectacle prescription but keeping the cylinder sign the same - e.g., converting +2.00 −1.00 × 180 to −1.00 +2.00 × 180 instead of −1.00 +1.00 × 090.
The arithmetic for the new sphere (add sphere and cylinder) is remembered, but the rule that the cylinder sign flips and the axis rotates 90° is applied only half the time.
What to do instead
Use the fixed three-step template every time - new sphere = old sphere + old cylinder; new cylinder = same magnitude, opposite sign; new axis = old axis ± 90° (subtract if over 90, add if 90 or under) - and self-check by neutralizing the transposed pair mentally to confirm it produces the same power in both principal meridians.
Converting a spectacle refraction to a contact lens power without compensating for vertex distance, so a −9.00 D spectacle Rx is ordered as a −9.00 D lens.
Vertex distance only matters above roughly ±4.00 D, so candidates who have done mostly low-power examples stop applying the formula and treat it as optional.
What to do instead
Adopt a threshold habit: any spherical equivalent beyond ±4.00 D triggers a vertex calculation. For minus lenses the contact lens power is less minus; for plus lenses it is more plus. Work two or three high-plus and high-minus examples until the direction of the correction is automatic, and always state the assumed vertex distance (12–13 mm) in your written answer.
Reading the Goldmann applanation tonometer force dial directly as mmHg - recording a dial reading of 2.0 as 2 mmHg instead of 20 mmHg.
The dial is graduated in grams of force and the conversion to mmHg is not printed prominently on the instrument, so the number on the dial is accepted at face value.
What to do instead
Fix the relationship in memory: 1 g of force = 10 mmHg, so multiply the dial reading by 10 (and account for the 0.1 g subdivisions). When reviewing a practice pressure, sanity-check that the result sits in the physiologic 10–21 mmHg range and re-derive if it does not.
Interpreting a visual field printout using the left side of the page as the left eye, or reporting a defect in the wrong eye on a Humphrey/OCT report.
Printouts are laid out in a patient-facing (mirror) convention, and the OD/OS identifier is small and in a corner, so candidates read the page spatially rather than by label.
What to do instead
Make the first action on every field or imaging item locating and stating the eye label before looking at any data. Then confirm the blind spot position: it lies temporal to fixation, so its side on the printout verifies the laterality. Never write an interpretation until the eye has been named.
Confusing reversal and side-effect profiles of common ophthalmic drugs - naming dapiprazole as the antidote to atropine, or attributing bronchospasm to pilocarpine rather than to a beta-blocker such as timolol.
Drug lists are memorized by class name and indication, while the reversal agents and systemic warnings get attached to 'dilation drugs' or 'glaucoma drugs' as a group rather than to specific molecules.
What to do instead
Build a two-column drug card per agent: intended ocular effect and the one systemic contraindication or reversal fact that is unique to it (timolol–bronchospasm/bradycardia; atropine–reversed only by cholinergic agents, not dapiprazole; dapiprazole–reverses alpha-agonist mydriasis such as phenylephrine; pilocarpine–induced miosis/brow ache). Test yourself by drug name, not by category.
Breaking or contaminating the sterile field during a minor procedure - reaching across the instrument tray, turning away from the field to retrieve an item, or passing an instrument by its working end.
Candidates focus on naming the correct instrument and forget that the examiner is also scoring field discipline; the hands move before the mind checks the sterile boundary.
What to do instead
Rehearse the sequence out loud: hands enter the field only from the front, items already on the tray are assumed contaminated if their sterility is uncertain, and instruments are passed handle-first with the tip guided into the surgeon's hand. Practise set-up and takedown of a standard cataract tray until the instrument names and their order are automatic.
Recording visual acuity in the wrong order of degradation - writing 'hand motion' for a patient who can still count fingers at one foot, or 'light perception' where 'no light perception' is correct - and omitting whether the acuity is corrected or pinhole.
The sub-chart categories (CF, HM, LP, NLP) are learned as a list without a fixed rank order, and the abbreviation habit of writing only the fraction carries over from the 20/x range.
What to do instead
Memorize the ladder once, descending: 20/20 … 20/400, then counting fingers, hand motion, light perception, no light perception. On every acuity entry write the condition tested (sc, cc, PH, PHNI) next to the value, and if the patient cannot see the largest line, force yourself down the ladder in order rather than jumping to a category.
Documenting a visit or assigning a code without capturing the elements that justify it - omitting the specific tests performed, the physician's interpretation, or the reason the test was ordered.
Candidates treat documentation as clerical transcription of what was done, so 'tests done' is recorded while the linkage to the diagnosis and the interpretation is left out, and the record no longer supports the code.
What to do instead
Before closing any record, run a three-question check: what was ordered and why, what was done, and what did the physician conclude from it. Ensure each diagnostic test has both a result and an interpretation attached, and that the diagnosis listed matches the tests billed. Practise on two or three sample encounters until the check becomes habitual.