Reassurance is a mighty tricky skill. Get it right, and your patient will be happy, and unnecessary investigations and prescriptions will be avoided. Get it wrong, and the consequences can be profound.

I have long been deeply irritated by healthcare pundits who look at a typical GP surgery and announce that most of the cases could have been dealt with by someone with far less clinical training. That’s an easy conclusion to reach at the end of a surgery or clinic.

A retrospective analysis of any list of patients might conclude that 70% of cases would have got better anyway just with reassurance. The skill comes from determining which 70% have a minor problem from the start, not the end, of the consultation.

Did that headache, that sore throat, that cough really matter? Was it the first sign of something much more serious? Or was it harmless? Making that decision can be complex. The most subtle of clues can take you in a very different direction. When applied erroneously, reassurance can have dire consequences.

A common complaint on any healthcare related radio phone in programme tends to be: ‘My GP kept reassuring me it was nothing.’ The caller then goes on to describe how he eventually had to make an emergency trip to hospital where the diagnosis of some serious condition was finally made. I often wonder what the doctor’s description of the encounters might be. After all, the events may not be quite as straightforward as the caller describes. However, I mostly feel sadness at the outcome.

Over my career I developed a very simple rule. It’s fine to reassure a patient once. Be very cautious about giving reassurance twice. Never, ever, give reassurance three times. After all, you need to keep remembering that you may be wrong, and reassurance is a clinical tool that gets massively diluted every time you use it.

Over the past couple of years, I have had personal experience of reassurance from the perspective of a patient. Presenting to a GP with a unilateral sore throat, I was extensively examined, and reassured. I have absolutely no complaints about that. It is exactly what I would have done.

However, when I returned with the same persistent symptom only a few days later, my symptom was taken very seriously. I was urgently referred and although the consultant’s examination was also reassuringly normal, a precautionary scan was carried out and a cancer of the tonsil eventually diagnosed. Initial reassurance was appropriate. Follow-up reassurance would have been disastrous.

I am now in remission, but the impact of the Covid-19 lockdown on the NHS has taught me something else about reassurance. My hospital follow-up had been by phone during this period, and one particular call came from one of the consultants looking after me. I mentioned a symptom that I had developed that was worrying me.

‘No need to worry,’ he said. ‘We know you’re fine. All the last tests were clear.’

What I immediately thought, and should have said, was: ‘If you know for certain that I’m fine, then why on earth are you phoning me?’ It reminded me forcibly of all the times I must have said to patients: ‘Everything’s looking good, but…’ And the ‘but’ might have been an arrangement for a follow-up, or advice as to what to do if problems flared up, or something else that Roger Neighbour has beautifully described as safety netting. However, the ‘but’ significantly dilutes the reassurance when you receive it as a patient.

The key, as ever, is honesty.

As a patient I crave certainty, but as a clinician I know certainty is a rarity. We do have to share our uncertainties, but phrase them in a reassuring way.