In an excellent discussion, Marco Martuzzi describes the precautionary principle as meaning that scientific uncertainty must not be used as a reason to delay taking action.¹ Deeply rooted in the history of public and environmental health, the precautionary principle has been embraced by ministers of health and environment across Europe. Martuzzi asserts that precaution is especially needed in areas fraught with complexity and uncertainty, for threats that may be irreversible or felt across generations, or when technological or societal change outpaces the accumulation of data.

Martuzzi is correct that the role of science is central to the application of the precautionary principle, but “caution” means something different in science. The hypothesis testing at the core of the scientific method tends to avoid false positives and tolerate false negatives. To illustrate, epidemiologists make a strong distinction between the terms “association” and “causal association”. The latter is used only when there are studies of high quality and a rigorous examination shows that other factors are unlikely to explain an association.² There are also inherent limitations to what epidemiology is able …