• Aneurysm

Unruptured intracranial aneurysms (UIAs) are increasingly discovered and preventive interventions increasingly performed; from 2000 to 2010, rates of coiling have exploded (14-fold from 0.3 to 4.3 per 100 000 American Medicare beneficiaries).1 The problem is that no one knows whether preventive interventions do more harm than good. During the same period, rates of subarachnoid hemorrhage (SAH) also increased from 20 to 25 per 100 000.1 Questioning the merit of preventive interventions is long overdue, but once questions are raised, how to properly address them?

On the surface, one reasonable answer is to try to identify patients for whom treatment would be indicated and separate them from patients for whom treatment would be futile or harmful. The idea makes sense, provided we proceed with the right methods. The core problem starts when a ‘natural history’ approach is employed. This method had noble beginnings, with the observation of animals and plants by Aristotle and Pliny the Elder,2 ,3 but it has now disappeared from all other domains in medicine, except ours. The most recent product of this approach is the PHASES scoring system, which proposes ‘a risk prediction chart to guide clinical decisions’.4 ,5

A fundamental question is whether this approach is scientific or pseudoscientific. Karl Popper attempted to demarcate science from pseudoscience: ‘A theory which is not refutable by any conceivable event is nonscientific. Irrefutability is not a virtue of a theory (as people often think) but a vice’.6 Let's see whether PHASES was designed to be refutable.

Natural history studies have long looked for risk factors for rupture, such as aneurysm …