DNB Ophthalmology Previous Year Questions: What Repeats and How Often

Previous-year questions are useful for DNB Ophthalmology when you use them as evidence about concepts you must retrieve, not as a promise that a remembered stem will return unchanged. NBEMS does not publish a public repeat-frequency table for ophthalmology. A responsible answer to “what repeats?” is therefore: recurring principles, standard comparisons and familiar decision points may recur in different wording; a candidate recall is not an official paper or a prediction.

That distinction saves a great deal of late-night, low-yield revision. The official NBEMS DNB/DrNB Exit Examinations page says DNB Final has theory and practical stages, and that a candidate who qualifies theory may appear for practical. It does not tell you that retina is worth a fixed number of marks, that a particular instrument will be asked, or that a recalled question will repeat. Keep official examination rules, standard references and recall material in separate folders in your head and on your desk.

Direct answer: use previous-year questions to build a concept-and-error ledger. Count recurrence only inside a source-labelled collection you have audited yourself. Do not turn a coaching compilation into an NBEMS blueprint.

First, define what you actually possess

The phrase “previous-year question” covers material of very different reliability. Treating all of it as equivalent is the main reason residents overestimate how much has “repeated.”

Start by adding a provenance column to every item. Record the claimed sitting, where you obtained it, whether the stem is complete, and whether two independent recall sources agree on the concept. If none of those fields can be filled, label it UNVERIFIED RECALL. You may still use it as a prompt to open Kanski, BCSC, Ryan, Elkington or another chosen core reference. You should not build a timetable around it.

What “repeats” usually means in a useful study sense

There are three distinct kinds of recurrence.

1. Same domain, new clinical clothing

A question may return to the same domain—say corneal oedema, retinal detachment, visual fields, IOL calculation or glaucoma—but change the patient age, image, measurement, differential or question verb. Recognising the domain is useful; memorising an old answer is not enough.

2. Same discriminating comparison

Exams regularly reward distinctions that are central to ophthalmology: pseudophakia versus aphakia in an optical scenario; rhegmatogenous versus tractional detachment; infective versus sterile inflammation; open versus closed-angle mechanism; afferent versus efferent pupillary defect. The individual question can change while the discriminating feature remains stable. Build cards around that feature.

3. Same question remembered by several candidates

Several recalls of a similar stem may indicate that it was memorable, not necessarily that it was common across years. This is particularly important for image-heavy, unusual or difficult questions. A collective memory can create a false sense of frequency.

For that reason, do not publish or repeat a statement such as “topic X repeats every year” unless NBEMS has released a current, explicit blueprint that supports it. As checked on 18 August 2026, the official DNB page does not provide such a topic-by-topic repeat count.

Build a recall-audit sheet before doing more questions

A decent recall collection becomes useful only after it becomes searchable. Create one row per concept, not one row per screenshot. The following fields are enough.

The most important column is tested decision. “Retina question” is not retrievable. “Given a detachment pattern, identify the mechanism and expected associated finding” is. If the recall merely says “asked PVR classification,” write the classification from your reference, create a comparison card and answer fresh questions around it. Do not memorise an unverified option list.

A practical method for turning one recall into durable learning

Use this five-step conversion each time you meet a potentially useful old question.

1. Freeze the recall. Copy it exactly as received and label the source. Do not silently repair a missing detail.
2. State the concept in one sentence. For example: “This is testing the relationship between a visual-field defect and lesion location,” not “the answer is B.”
3. Verify from a standard source. Check the mechanism, definition, classification or calculation. If the recollection conflicts with the reference, the reference wins.
4. Create a neighbouring comparison. Ask what single feature would reverse the answer. That is where SBA/MCQ distractors usually live.
5. Re-test in a new form. Use a blank diagram, altered values, a different image or an oral explanation. If you only recognise the old wording, you have memorised a recall rather than learned a principle.

Here is how that looks in three common types of work.

This is slower than racing through a PDF on the first pass. It is also the point of past-paper work: a small number of verified, well-reviewed prompts should change how you handle many unseen stems.

Track recurrence without inventing weightage

You can calculate your own collection’s recurrence without making an NBEMS claim. Give every concept one of four labels:

• Single observation: appears once in one source.
• Corroborated observation: appears in independently sourced recalls, but wording differs.
• Concept family: several questions test the same decision through different cases.
• Core curriculum anchor: a concept that is central in your verified curriculum/reference whether or not recalls mention it.

Your weekly priority should not be “highest count first.” A narrow, weakly sourced set can dominate a small recall archive. Use a two-axis matrix: source confidence and curriculum importance.

This approach is especially valuable if you are preparing around calls, OPD and night duty. It prevents a loud, newly forwarded “repeat” list from repeatedly stealing time from optics, anatomy, pathology, imaging interpretation and other foundations you still cannot retrieve cleanly.

A six-week past-question cycle for a working DNB resident

This is a study template, not a pass plan. Fit the number of questions to your rota and your baseline.

Weeks 1–2: clean the archive

Collect material, remove duplicate screenshots and add provenance. Do one short mixed diagnostic block from verified material if available. For every recall, identify the underlying topic and create at most one durable card. Do not yet count “frequency.” The purpose is to reveal what you actually have.

Weeks 3–4: make concept families

Choose two to three major domains a week. Pair each recall-derived card with fresh questions from a source you use for practice. Review wrong answers and guessed-correct answers alike. If multiple recalls appear to test the same decision, write one comparison table instead of making five near-identical cards.

At the end of each week, identify the three error types that cost you most: missing knowledge, confusion between close options, process errors such as units/negatives, or timing. Those categories lead to different corrections.

Week 5: mix and time

Stop doing a retina-only or glaucoma-only recall set in isolation. Mix domains. Old papers are most useful when they expose whether you can switch from optics to uveitis to neuro-ophthalmology without a chapter heading telling you what is being tested. Flag questions; do not spend a large proportion of the block wrestling one uncertain item.

Week 6: consolidate the ledger

Review only your high-importance concepts, repeated error classes and formula/pathway sheets. Keep a separate “do not re-read” list for obscure, low-confidence recollections. It is a useful guardrail: every item you exclude creates time for a verified weak area.

For a wider structure that integrates theory and practical preparation, use the DNB Ophthalmology preparation hub. Do not assume that a theory recall is a forecast of a practical station; practical planning needs its own rehearsal method.

Errors that waste past papers

Counting files rather than concepts

Five PDFs may contain the same original recall copied forward. Without source labels, you can count one memory five times. Deduplicate before assigning importance.

Reading explanations passively

After reading an explanation, close it and say why the nearest alternative is wrong. If you cannot, the explanation has not become an exam decision.

Treating a candidate recall as an official question paper

Recall wording can omit a critical negative, image or value. Preserve uncertainty. It is safer and more useful to learn the concept from a reference than to “correct” an incomplete stem by guesswork.

Making the product the syllabus

Past-paper material can organise practice, but it cannot replace the current curriculum, standard references or candidate instructions. If you want a source-labelled resource to use within this audit method, the site’s Past 5-Year Question Papers DNB/DO page is an optional route to inspect. It is not an NBEMS publication and does not predict questions.

Ignoring practical preparation until after theory

NBEMS describes Final as theory followed by practical eligibility after theory qualification. That procedural sequence is not a reason to postpone all clinical presentation practice. Keep light, regular bedside and viva rehearsal alongside theory, then intensify it when the current practical instructions are available. The OSCE, Practical & Viva Voce Ready bundle is a separate preparation route; always let your current NBEMS candidate instruction control logistics and format.

The one-page checklist

Before you make a “high-yield repeats” list, be able to answer yes to each item:

• Can I identify the source and claimed sitting for every recall?
• Have I separated exact documents from partial memories?
• Have I checked the underlying concept in a standard reference?
• Does my note say what decision the question tested?
• Can I explain why the closest alternative is wrong?
• Have I avoided assigning official marks, frequency or probability without an official source?
• Have I kept the current NBEMS portal and my candidate instructions separate from third-party material?

If the answer is no, do that work before downloading another recall pack. It is the difference between accumulating questions and building a revision system that can handle new wording.

Sources

• NBEMS: DNB/DrNB Exit Examinations — current official Final examination process; checked 18 August 2026.
• NBEMS curriculum library — official curriculum documents, including Ophthalmology; checked 18 August 2026.
• NBEMS practical examination portal — session-specific practical schedule and candidate information; checked 18 August 2026.