If it doesn't work, stop it

Editor—Assessing effectiveness and benefit in epidemiological studies entails several degrees of imprecision and uncertainty at the individual level while its estimator is defined for a population. A single disease entity may have varied manifestations in different patients and divergent outcomes.¹

Figure 1.

As Anderson and Groves emphasise,² Archie Cochrane posed three key questions to ask about a healthcare intervention: “Can it work?” “Does it work in practice?” and “Is it worth it?” We usually use rules to decide between “yes,” “not sure,” and “no.” By doing this we assume Aristotelian logic and the classic current definition that “health and disease are opposites and that they are dual and contradictory attributes.”¹

Why do doctors use treatments that do not work?³ People usually do not require precise numerical information, and yet they are capable of making decisions. They accept noisy and imprecise input⁴; so do doctors.

Having a huge number of input variables (patient background, expectations, behaviour, and beliefs, disease manifestations, laboratory results, etc), doctors use “fuzzy logic” algorithms (grade of evidence, personal knowledge, cost, ritual, mystique, etc) to decide treatments. Evidence is one of the most important pieces of the complex system, but not the only one. Fuzzy logic has been developed to deal with the concept of partial truth values between completely true and completely false. It mimics human control logic and may be applied to improve knowledge in epidemiological and medical problems.¹,⁴