On the observed specific and non-specific effects of complex therapeutic interventions: truly separate or complementary?

Abstract

Specific and non-specific effects observed in randomised controlled trials (RCTs) are generally treated implicitly as ontologically separate and purely additive. Building on the notions of Heisenberg uncertainty and complementarity from the discourse of quantum theory, and using a simple arithmetic argument, it is demonstrated how this separation enables results of RCTs (particularly of complex interventions) to be treated in a convenient but ultimately incorrect manner. Conclusions drawn from RCTs (that justify--and are justified by--a reductionist approach to therapeutic efficacy) should therefore be open to question.