स्तम्भ Stambh · Swanīti

Ask How Many Times

One question separates the surgeon you want from the one you were given, and it works on lawyers, agencies and founders too. The rule I use and the number that made me use it.

Nikhil Sharma · स्वनीति Swanīti · 2 September 2026 · 4 min read

I run AnandBodh, a wellness venture. This column gives no medical advice beyond a question to ask.

Before anyone cuts you open, ask how many times they did it last year.

Not how many years they have practised. Not where they trained. Not whether the hospital has a marble lobby. How many of this exact operation, in the last twelve months, by this hand. It is one question. Almost nobody asks it, and the reason nobody asks it is that it feels rude, and the reason it feels rude is that we have been taught to treat expertise as a title rather than a count.

Here is the number that made me start asking. In 2002 John Birkmeyer and his colleagues looked at 2.5 million operations on Medicare patients in the United States across fourteen procedures between 1994 and 1999. For pancreatic resection the death rate was 16.3 per cent at the hospitals that did the fewest and 3.8 per cent at the hospitals that did the most. Same operation. Same country. Same insurer. One patient in six against one in twenty-six, and the only variable was how often the building did it. For heart bypass the gap was under two points, because almost everyone does enough of those. For the rare and difficult operations the gap was the difference between a funeral and a follow-up.

Charaka put the quality the number measures in three words two thousand years ago. Having seen the work done many times. The modern paper and the ancient text agree, and the patient is still too polite to ask.

Expertise is not a title. It is a count, and the count is the only thing you are entitled to ask for.

How to ask it

Plainly, and early, and without apology. Doctor, how many of these did you do last year. A good surgeon answers in a number and is pleased you asked, because the number is the thing she is proudest of and nobody ever asks. A surgeon who answers with years, or with the name of the hospital, or with a small lecture on how every case is different, has told you the answer. It is low.

Then ask the second question, which is the one that actually matters. And what happened to them. Not every surgeon has the figure. The good ones know it roughly and will say so, and will say roughly, and the honesty of the roughly is itself a signal. A surgeon who has never counted has never looked, and a surgeon who has never looked has no way of knowing whether she is getting better or worse.

If the number is low, do not argue. Thank her and ask who in the city does the most. Surgeons know. They refer to each other all the time and will tell you if you give them a way to say it without insulting themselves. The one who sends you to a colleague has done you the second-largest service in medicine, which is knowing the limit of the first.

It is not only surgeons

I learnt the question in a hospital and I use it everywhere. How many times have you set up a company in this state. How many campaigns in this category in the last year. How many founders have you taken from the first cheque to the second. The answer is always more revealing than the pitch, and the people with the highest counts are always the least offended by the question, because the count is their whole argument and they are relieved somebody finally asked for it.

The question works on me too. Anyone who sits across from me is entitled to ask how many companies I have actually sat with through the gap between deciding and doing, rather than advised from a distance. The honest number is always smaller than the manner suggests. Give it anyway. A founder who gets a number instead of a story knows what he is buying, and the ones who sign on a number stay.

The rule for the person asking is one sentence. Anybody who will put a knife, a contract or a year of your life in their hands owes you a count, and anybody who will not give you one has given you one.

Written in Gurugram on 2 September 2026, a week before the long essay this column walks beside. The essay says India’s surgeons have done the operation two thousand times and the country sells the discount instead. This column is the same argument at the scale of one patient in one chair.

Sources: Birkmeyer and others, Hospital Volume and Surgical Mortality in the United States, New England Journal of Medicine, 11 April 2002, for 2.5 million procedures over fourteen operations from 1994 to 1999, mortality of 16.3 against 3.8 per cent for pancreatic resection and a gap under two points for coronary bypass. Charaka Saṃhitā, Sūtrasthāna 9.6, for bahuśo dṛṣṭakarmatā.