So in history class yesterday, our prof was talking about the development of the medieaval legal institutions and to illustrate a point, he asked if medicine was a science or an art. We and he together agreed that it was both. The distinction being that medicine had to deal broad truths that hold in general, ie. over large populations, but when applied to individuals can break down and therefore can require a more "creative" approach. This got me to thinking about why this is. I realized that it has to do with math.
In hard science, the hardest of which is arguably pure mathematics, variables, that is the things in an experiment that we want to change or control, are few. The unknowns, that is the thing we are investigating, are even fewer and ideally in a scientific experiment is one. For example, in finding the acceleration of an object, we can use the equation F (force)= M (mass) x A (acceleration). If we know, or control any two of these variables, we can find the third value. Simple, straightforward, very scientific and highly reproducible.
In medical science we use another phenomenon, that is statistics, to control and reduce our variables when carrying out studies. We do this through the power of large numbers. Humans are wonderful in our averageness. Simply stated, on average, we are average. So when you give a medication to a large number of people, you will generally see "a" response. This is the average response, but due to the large number of participants, one can predict that on average, this would be the response. The large number of participants effectively reduces the huge number of variables that exist in the interaction of one person with one medication, to a much smaller number of variables. This is why medical experiments are often double blind and the data is then examined for correlations between effect and age, kidney function, prior disease states, etc, etc. to ensure that the average remains the average and none of these other factors effect the results. If one of these other factors does effect the results, it becomes another variable. The averaging of a large number of people also averages the unknowns and effectively limits them to the question of the effectiveness of the given therapy.
So where is the art part. Well, when doctors apply their knowledge of science to an individual, on average they will see either the disease in a "classic" (average) presentation or a medication will work as predicted, but because the individual is just one and has no average, all their variables and unknowns are in play. Doctors must apply their experience, their knowledge and their intuition to a problem in order to achieve the results they desire. Hence the art.
The fewer the variables and unknowns the more scientific the pursuit, the more variables and unknowns the closer to art something becomes. In the end, the difference is between what we know and can control, and that which we do not yet fully understand enough and are unable to control. It's a spectrum of mathematics with science at one end and art at the other.