When paradigms shift it’s always disconcerting. Thomas Samuel Kuhn published The Structure of Scientific Revolutions in 1962 and it’s become a decisive text on the nature of science. He used the term “paradigm” to describe the belief system that underpins puzzle solving in science. Far from discovering any absolute “truth,” normal science progresses routinely within the boundaries of the belief system as theories are refined, puzzling data are mostly explained, and measurements become increasingly precise. Aristotle’s Physics, Newton’s Principia, and Darwin’s Origin of Species are examples of works that defined paradigms of particular branches of science at certain times.

But as time passes the puzzling emerging data may become increasingly problematic, and the existing theories may gradually—sometimes suddenly—fail. Normal science’s relatively serene stability is punctuated by a crisis, which can only be resolved by revolution. Turbulence follows as revolutionary science takes over and some old paradigms give way to new ones. Eventually, what was once revolutionary becomes the new orthodoxy and the cycle begins again.

In 2014 a newly qualified doctor faces the daunting prospect of another 50 years of clinical practice, much of it looking after an ageing population—most of whom will have multiple long term conditions. The evidence base to support his or her decision making is constructed within the paradigm of truth searching, which goes back to at least Aristotle who favoured observation and measurement over Plato’s paradigm of reflection and reasoning. Alcahan in Persia went further and engaged in controlled experimentation to pursue the truth. Francis Bacon developed the principle of hypothesis testing . . . and so on, right up to Karl Popper and the impossibility of proving the absolute truth of an assertion.

The problems for our young doctor are many, not least that even if there is a well conducted and relevant experiment for the clinical problem of that moment, all of that science is based on population averages and largely in populations from which people with multiple conditions were excluded. Great for the purity of the science, better than expertise alone, but not always that helpful for an individual patient. To say nothing of the problem that it’s rare for one of our doctor’s interventions studied in a randomised controlled trial to help more than a few of the people who receive it; most people live or die, have an event or don’t have an event, irrespective of whether they get the experimental intervention or control.

Our young doctor might know what happens in this highly selected population, but cannot predict what will happen for this individual. Again, this is to say nothing of the volume of potentially useful information available, which is many orders of magnitude greater than our doctor’s cognitive capacity.

Time to say it then. This data is disturbing to the stability of clinical practice largely dependent on a scientific paradigm. It’s threatening, and turbulent, and disconcerting. Especially to those of us who have been seeking truth through the evidence base for 20 years. But we can’t ignore it now. We find ourselves in a situation which is far from normal. For our patients with long term conditions, risks and benefits of treatments are uncertain, values and preferences differ and are often tacit rather than explicit, stakes can be high, some decisions are urgent, complexity is the norm, and the risks of polypharmacy may be running out of control.

Funtowicz and Ravetz have extended Kuhn’s revolutionary science concept into post normal science (PNS). In medicine the old paradigm, where empirical data led to “true” conclusions and scientific reasoning led to “correct” clinical policies for individuals, always had an element of an illusion about it and is now no longer plausible. PNS helps, I think, for it changes the basic criterion from truth to quality and the dimension from technical to ethical. In PNS the ideal is not the attainment of some perfection in knowledge or practice, but the improvement of awareness of oneself and of one’s partners in a dialogue.

Isn’t that the business we’re actually in? Truth has always seemed out of reach, but I can absolutely empathise with quality as a goal. After about 115 years of mostly excellent health (fingers crossed), when I’m finally failing I will need a dialogue about my values and preferences with my family, physicians, and carers. That dialogue should help all of us in managing the uncertainties; conflict will not be removed but reconciliation based on understanding becomes possible. The evidence base informs such decisions, but it’s only a part of the decision making.

As a patient living 24/7 with my long term conditions, I need a bit of knowledge for sure, but mostly Olympic gold medal winning consultation skills from my physician. Increasingly, people will be able to suss out at least some of the evidence informed treatment options for themselves. This is good for us too. As a medic in his 40th postgrad year, for people facing a 50 year medical career it’s the unselfish effort on behalf of others that gets you through, not trying to know everything. It’s about trying desperately every day, despite the workload, to see and respect people as people with their baggage of past experiences—for it is these which shape their unique values, preferences, and lives. It’s people that matter in the end, and not what the latest science says. Kindness and caring is usually reciprocated in a therapeutic relationship, and keeps physicians as well as patients going.

Sometimes in paradigm shifts it’s difficult to be sure if you’re in one or not. Most Manchester United fans seem to accept that their 25 year belief system has shifted. In medicine, if there is a new paradigm, it carries forward large chunks of the treasured caring of people linked with technical competence, which goes back through the greats of medicine to Hippocrates. Perhaps what characterises it as at least a new era, is that while we must first construct and then know the evidence for the population, we can and should legitimately reject its absolute truth when applying it to decision making with or for individuals. We need to accept the requirement to always give explicit exposition of the uncertainties, and always seek the avoidance of conflict through dialogue with patients about values and preferences.

First do no harm, always try and do good, justice and equity, and patient autonomy are still the guiding stars of our decision making. But the greatest of these is actually autonomy: for without it, what we do is assault patients. It’s hard to argue that talking more with patients about what they think and want is a bad thing.

Neal Maskrey’s early career was as a GP before spending seven years as a medical manager and part time GP. After 12 years as a director of the National Prescribing Centre and programme director at NICE, he is now honorary professor of evidence informed decision making at Keele University, and consultant clinical adviser in the Medicines and Prescribing Centre, NICE.

Competing interests: I declare that I have read and understood the BMJ policy on declaration of interests and I hereby declare the following interest: Employed part time by the National Institute for Health and Care Excellence.