Hormesis

Traditionally, US regulatory agencies have used a linear no-threshold (LNT) model for assessing risks from carcinogens and other toxins. This article offers the argument that other models may actually be more pertinent.

The LNT model assumes dose response follows a straight line passing through zero. Thus, any dose above zero will, theoretically, produce some response in the population.

The threshold model posits that the response to a dose is zero until a certain minimum level is reached, and then rises above zero. That means any dose below the threshold is safe – that is, produces no ill effects at all.

The hormesis model posits that the response curve is less than zero below a certain threshold. That is, the dose actually benefits the organism until a threshold is reached. Thus, a small dose of some compound improves health, and only a larger dose will actually cause harm.

Hormesis is suspected, not only in the case of various chemicals, but also in the case of radiation. And just to make matters worse, it's possible that dose response curves could be non-linear, with the first few micrograms causing more of an effect than the last few.

We may even learn, as in Woody Allen's movie, "Sleeper", that all the stuff we think is killing us is the best stuff for us.