[Evidence-based medicine: reality and illusions. Extension of epistemological reflexions].
Evidence-based medicine (EBM) is a cultural and methodological approach to clinical practice helping to make decisions based on clinical expertise and an intimate knowledge of the individual patient's situations, beliefs, and priorities useful for the analysis of clinical research. As such, it can be considered the scientifically grounded art of medicine, as it appears to be an emerging paradigm of scientifically based clinical care. It de-emphasizes intuition and unsystematic clinical experience as grounds for medical decision-making and stresses the rigorous and formal analysis of evidence from clinical research. EBM converts the abstract exercise of reading and appraising the literature into the pragmatic process of using the literature to benefit individual patients, while simultaneously expanding the clinician's knowledge base. On EBM grounds, clinical, practice guidelines, pathways and algorithms or instructions can be developed with the aim of solving a problem or accomplishing a task. Nonetheless in these processes the theory of EBM shows internal and external bias. Among internal bias, economic-based interest may influence the development and diffusion of research and its results. In addition "systemic review" may be incorrectly guided, the quality filters of the literature can be inappropriately applied, the choice criteria can be only based on the positive results of evidence, but according to modern epistemology, it will be helpful for clinicians to know when their uncertainty stems from gaps between positive and negative evidence. Another bias is the difficulty to convert EBM into clinical practice recommendations. EBM set movement has shown that it is nearly impossible to make recommendations that are appropriate in every situation. Epistemological approach identifies external "bias" of EBM. It is consistent with the theory of "fact" as human construction. Every human fact can historically fade and then be restored according to new paradigms. EBM is a "fact" and its theory can be changed or removed every time by relevant new or emerging evidence approaching the development of up-to-date decision-making aids and so on, irrespective of how much previous processing of the evidence has taken place. Then EBM cannot be evaluated as the scientific "totem" of the third millennium, neither as the clinical digest of medical literature. Searching for clinical evidence, in fact, requires a great awareness of both the advantages and limitations of increasing bias. Clinicians are looking for new strategies to apply to diagnostic and therapeutic pathways and for the steps where EBM could be addressed when showing the full validity.