I keep getting the very scary sensation that absolutelyeverything that I learnt at medical school will one day be proved to be wrong. Well, maybe I'm exaggerating. It's probably unlikely that much will change in the world of gross anatomy. But, when it comes to areas such as therapeutics and management, then few drugs or theories ever survive the onslaught of new research.

The last straw seemed to be the recent reasonable questioning of whether lowering the temperature of feverish children is a good thing or prolongs their illness. I may have to add that to the list of activities that took up countless hours of my time in the early years of my career and for which it now appears there is no evidence that I was doing anything other than wasting time.

On countless occasions when I was on call in the years before out-of-hours co-ops, I would get out of bed in the middle of the night to visit a crying child with earache and to deliver the antibiotics as an emergency. Parents were hugely grateful, the pain seemed to ease rapidly, my time out of bed seemed justified, and then the published evidence made it clear that my antibiotics were no better than placebo. Was I wasting my time?

And then there were all those tubes of chloramphenicol eye ointment that we prescribed for conjunctivitis - another apparent waste of time. Going further back, myocardial infarction was treated with three weeks of strict bed rest - with no attempt to minimise the risk of venous thromboembolism - which means that it was a miracle that anyone actually survived. It is little wonder that before the invention of thrombolysis, recovery from heart attacks was just as good following home management as it was in a coronary care unit.

Indeed, I still have a letter sent me by a very grateful patient when I was a house officer expressing intense gratitude for our wonderful care - care that we now know was positively dangerous.

It certainly taught me that gratitude and recovery can never be used as evidence of effectiveness - something all too many people fail to recognise.

When I first worked in general practice we typically didn't admit patients with a stroke as there seemed ‘little point'. The list of changing therapies goes on and on.

Which is exactly how it should be. After all, much of what we do is scientific, and a quest for evidence and truth is essential. To deny the importance of science would leave us no better than many of the other therapists who peddle their wares in the health arena.

But I am fascinated by the way that much of the human side of our actions has not changed in the slightest. Most patients are frightened, anxious, or at least concerned - kindness and understanding count for a lot.

Those parents who were grateful for my arrival at 3 a.m. with amoxicillin for their child's ear were probably just as grateful for the apparent compassion, caring, and reassurance that such an act demonstrated as they were for the magic mixture itself. I always felt relieved that they couldn't hear the complaining and slamming of doors  every time I got out of a warm bed.

The art of medicine may not be as glamorous or as well researched as the science, but that certainly doesn't mean that it isn't important. And when the science demonstrates that yet another therapy is valueless, let us not throw too many babies out with the bathwater. Especially if the bathwater is cooling the babies even more effectively than the paracetamol.