Causal interactions, fuzzy sets and cerebrovascular 'accident': the limits of evidence-based medicine and the advent of complexity-based medicine.

In evidence-based medicine, stroke subtype is diagnosed after a sequential search for etiology; the first positive test result of significant severity rounds off to one overwhelming cause. Degree of severity, interaction among variables, and concomitant variable conditions are not considered in defining the cause of stroke. Yet, thrombus formation, and possibly vascular rupture, is an interactive process involving the vascular wall, flow properties of the blood and blood constituents; this process occurs in homeostasis and pathology. Evidence-based medicine ignores this process and instead studies stroke using crisp 'all or none' classification where subtypes are distinct and interactively relate only to outcome. As a result, scientific inquiry is focused on prediction for the collective of patients. The statistical approach of evidence-based medicine is founded on probability theory, itself rooted in classical set theory where elementhood is all (1) or none (0), and opposites interact only to form the null set. Fuzzy set theory, where set membership is to degree [0, 1], encompasses classical set theory, allows for an interactive process between variables, and therefore becomes the measure of complexity. Fuzzy set theory can change the scientific method of evidence-based medicine.