Complex systems, explanation and policy: implications of the crisis of replication for public health research

Abstract

Much public health research considers interventions that influence and are influenced by both individuals’ health and the society around them; this can be described as a complex system. We consider the role of explanation alongside statistical inference as crucial to obtaining credible and useful insights, particularly in light of concerns about a ‘crisis of replication’, and reflect on the difficulty in researching complex systems. In this paper, we make a connection with Lipton’s philosophy of inference to the best explanation, and offer some tentative practical recommendations, extending suggestions from different fields. An extended hypothetical example is given on introducing ‘sugar tax’ in England.

Keywords:

• Complexity
• crisis of replication
• inference to the best explanation
• methodology
• policy
• public health
• sugar tax