Medicine is not science

Abstract

If scientific knowledge cannot be distinguished from other knowledge we cannot prohibit use of unscientific and non-scientific knowledge in medicine. Science addresses narrow closely defined questions (hypotheses) in closely controlled conditions by carefully designed experiments to elicit a narrow range of evidence of high reliability, adopting the highest standards of proof. Science proves cause and effect by intervention demonstrating regularity of outcome.  Universal generalisation by theory enables accurate and reliable prediction in existing previously observed and unobserved cases and in wholly new cases.

Neither the practice of medicine nor medical research is science. A fundamental limit to universal generalised theorisation is medical research’s inability to test by rigorous scientific experiment. This is an inevitable consequence of treating individuals in heterogeneous populations of complex biological organisms.

Randomised controlled trials (RCTs) are not science. RCTs do not tell us: when we can predict X will cause Y in a particular patient; that X is the only cause or factor in the cause of Y; when X does not cause Y; why or when Y will appear when X is not present. RCTs cannot therefore verify hypotheses about when or why X does or does not cause Y. RCTs also provide no information enabling us to predict new outcomes in addition to Y. RCTs provide no mechanism for prediction beyond a probability of how often we might expect X to cause Y. RCTs also provide no knowledge to  predict the outcome of a treatment involving 2 or more drugs of unknown interactivity. Meta-analyses of RCTs, also, cannot therefore be science and are subject to other limitations.

Observational statistical studies are not science. They tell us only that X is seen in association with Y. There is no intervention, so they provide no information about whether X is a cause of Y. The assessment of causality is not scientific, but a matter of judgement requiring evaluation of clinical data including detailed medical histories of cases. 

Medicine is concerned with living individuals, not predicting outcomes of natural phenomena occurring throughout the universe. Universal theoretical generalisation may be impossible, inappropriate or unnecessary. Medicine may not need the same high standards of evidence and proof as science. Scientific method, the types of evidence it admits and that which it rejects and the standard of proof applied in science, are less relevant to, and in any event unattainable in, medicine, but improvements can be identified and should be implemented. 

The practice of medicine cannot be scientific.  Medicine requires the assessment of probabilistic information applied to broad and not narrow questions and so is reliant on professional judgement based on experience and expertise. Well documented case and clinical histories of individual cases can provide knowledge of potentially high reliability through systematic, methodical and rigorous research. The compilation of databases of such histories should be a priority. Medicine also uses non and unscientific information and knowledge which is not the result of medical research nor established by RCT. Like Complementary and Alternative Medicine (CAM), much modern medicine fails to meet that latter standard. Unlike medical research, CAM research is relatively recent and grossly underfunded. It is unscientific to abandon or dismiss medical fields which have not been the subject of formal research and publication. In their discussion paper, Miles and Mezzich talk of medicine, science, scientific medicine, scientific medicine and biomedical and technological research and its use within clinical practice. In this response to Miles and Mezzich, we examine all such notions in order to assist conceptual and terminological clarity.