“I don’t know.” Three truthful little words which we doctors don’t seem to be able to say to our patients. Why not? Have we joined our patients in believing that there is an answer to every one of their problems and concerns? That there is a diagnosis for every discomfort or unhappiness? Think DSM IV. That there is a magic bullet for every ailment? That randomised controlled clinical trials (RCTs) will tell us which medicine will cure which patient’s illness or alleviate their symptoms?

Why is it so hard for us to say those three simple words to our patients? Is it because, throughout our medical education, we were embarrassed to say them in front of our peers, or that we were mocked by bullying authoritarian teachers if we owned up to a lack of knowledge?

Perhaps I was lucky. I learnt very early on to say “I don’t know.” My father taught anatomy at our medical school. From my third year on, other teachers would turn to me for answers to anatomical questions. I soon learnt that there was no shame in answering, “I don’t know.” I would often cheekily add the rider, “knowledge is not inherited.”

I well remember colleagues who would stumble through a half-remembered, but clearly wrong, answer, or who would guess – with disastrous consequences.

Before joining the general practice in which I eventually became the senior partner, the old founding partner once found me looking up something in a book while a patient was with me. I had told the patient, “I don’t know.”

“You never say that to a patient,” he admonished me in private. “You tell them that their question is complicated, and that it would better for them to return another day when you will have time to explain properly – in the meantime, you look up the answer.”

Let’s be honest: there are questions to which we don’t know the answers. Not only do we, as individual doctors, not know the answers, but nor do our specialist colleagues to whom we might refer our patients, and nor does Google. There are RCTs galore, about almost every conceivable illness (and about many non-illnesses). But none of them can tell us what advice or treatment will benefit the individual patient who is with us at the moment.

I was told in 1989 by the late Victor Chang, after my coronary bypass surgery, that I had “a 70% chance of living for 10 years.” What did that mean? It really meant that I was in one or other category of survivor – those 30 who would die within 10 years or those 70 who would live longer than a decade.

The truth was that Victor did not know which group I was in – but he didn’t say so.

With our patients demanding answers from us, we are, like my late old partner, born in 1900, unwilling (perhaps unconditioned) to say “I don’t know.”

“Should I still be on hormone replacement therapy?” – “Should I have a PSA test?” – “Which statin should I take?’” – “Will I get side-effects from these tablets?” – “Can I safely take the oral contraceptive pill?” and so it goes … Our patients want clear-cut answers; the truth is that we cannot always tell them.

Is it not time to give up our antediluvian pretence at omniscience and learn to tell our patients the truth – “I don’t know?” Is it not time that we conditioned ourselves, and our patients, to learn that life is a lottery, that we are all compelled to take chances, that no human action is risk-free and that we are, quite simply and honestly, unable to truthfully answer many of their questions?

We just don’t know.

Peter Arnold is a retired Sydney suburban general practitioner.