The viva is a short conversation about a document you know better than anyone else in the room. That asymmetry is an advantage only if you have prepared from your own thesis rather than from a list of predicted questions.
What the external examiner is assessing
Under the National Medical Commission's Post Graduate Medical Education Regulations 2023[1], the MD and MS thesis is assessed within the practical and viva examination and is judged by an external examiner from outside the state. For DNB and DrNB, the thesis is assessed by assessors appointed by the National Board of Examinations in Medical Sciences[2] before the final examination, and the questioning at the table is about work that has already been read and marked. Either way, nobody is re-running your statistics in the room.
What is being tested is narrower and more human than most residents expect: whether the work is substantively yours, and whether you understand why it was done the way it was.
- Do you know your own data well enough to answer without looking it up?
- Can you justify the design, the setting, the period and the criteria you chose?
- Do you understand the tests that were run, at the level of what they assume about your data?
- Do you know your limitations before the examiner names them?
- Can you place your finding beside the published literature, including the studies that disagree?
The first two minutes
You will usually be asked to describe your work at the start, though the format varies by institution. Have five sentences ready and practise them aloud until they are not recited. One sentence on the clinical gap that made the question worth asking. One on design, setting, period and number of participants. One or two on the principal finding, with direction and magnitude. One on what it means for practice in your setting. One on the main limitation, offered by you rather than extracted from you. Offering the limitation yourself changes the tone of everything that follows, because it establishes that you are the person in the room who has thought about this the longest.
Do not open with a definition of the disease, and do not open with the burden of the disease in the world. Both signal that you are delivering a presentation rather than discussing your own work.
The questions that come up almost every time
- Why this topic? Give two reasons, one clinical and one practical: the specific gap you noticed in your own department, and the fact that your hospital could realistically answer it in the time available.
- Why this sample size? Know the formula, the assumed values, the study those values came from, the power and the alpha, and the attrition allowance. Know whether you reached the number, and if not, say so plainly and say what it cost you.
- Why this statistical test? Start from the variable, not from the test: what you measured, on what scale, how the distribution looked, what you did about it, and which test follows from that.
- Why this inclusion or exclusion criterion? Every criterion narrowed your population for a reason. Be able to give the reason for the one that narrowed it most.
- What is your most important finding? One sentence and one number. A candidate who needs three sentences here usually has not decided what the thesis is about.
- What would you do differently? Answer at the level of design, not effort. A larger multicentre sample, a randomised allocation, an objective endpoint, a longer follow-up. Not that you would work faster.
- Has anyone found the opposite? Know the two or three papers that disagree with you and the reason each one might differ.
- Who did the statistics and who wrote this? Answer honestly. You are expected to have had a guide, and in many departments a statistician. You are expected to understand and be able to defend every choice that was made.
Prepare from the thesis, not from a question bank
Read your own thesis with a pen and write a question in the margin at every point where you made a choice: why this definition, why this cut-off, why this duration, why this test, why this table. You will generate dozens of questions, and they will be better than any list because they are about your document. Then answer them out loud, because an answer that is clear in your head is often not yet clear in your mouth.
- Make one page you can revise in five minutes: objectives, design, setting, period, sample size with formula and source, the tests used, your three headline numbers, three limitations, three papers that agree and two that disagree.
- Be able to draw your study flow from memory: screened, eligible, enrolled, completed, analysed, with the losses at each step.
- Know your master chart, including how missing values were handled and what your total actually is in each table.
- Tab your own bound copy so you can reach any table within a few seconds.
- Ask your guide or a senior resident for twenty minutes of questions you have not seen, and do it at least a week before, not the night before.
Answering a question you cannot answer
Say so, briefly, without apologising, and then give the nearest thing you do know. That you do not have the incidence for that age group, but in your own sample the figure was this, and the paper you relied on reported that for adults. Do not invent a number. A guessed figure invites a second question on the same ground, and the second question is the one that does the damage. Do not argue with an examiner about a fact neither of you can check in the room.
When a question sounds hostile, it is usually testing whether you will defend the work or abandon it. Do neither reflexively. If the criticism is correct, concede it precisely and show where the thesis already accounts for it, because a conclusion stated for tertiary-care patients is not undone by the observation that it may not apply in primary care. If the criticism rests on a misreading, point at the page rather than contradicting the examiner.
The practical side
What you carry and whether you present at all varies by institution, so confirm the format with your department rather than assuming. Ask whether there is a presentation, how long it runs, and whether slides are expected. Keep together your bound copies, the declaration signed by you and your guide, the ethics committee approval letter with its date, the similarity report your institution requires, your e-logbook, and evidence of the research activity and presentations your regulations require. The regulations themselves are worth one read in the original before the viva, because questions about process are asked more often than residents expect and are the easiest of all to prepare for.