LNT - A Hypothesis, Model Or Theory?

The "Linear No Threshold" is often followed by one of these three words "Hypothesis", "Model", "Theory" when the discussion is regarding radiation cancer risk.  In an informal sense it doesn't really matter which word is used as long as the people engaged in the discussion are sharing a common understanding.

But what about in a more formal sense?  Or what if someone is intentionally using one of those words to demean our scientific understanding?

So let's dig in.



A hypothesis can be thought of as an educated guess or a proposition to be tested.  That there is a LNT dose-response between exposure and cancer risk is a hypothesis.  Then one can perform an experiment, let's say with mice, and produce results which support the hypothesis.  BEIR VII's conclusion is in response to that simple hypothesis:

"CONCLUSION

The committee concludes that current scientific evidence is consistent with the hypothesis that there is a linear, no-threshold dose-response relationship between exposure to ionizing radiation and the development of cancer in humans."

Now BEIR VII isn't just about one mice experiment which supports LNT.  BEIR VII is a meta-analysis which examined a broad range of scientific evidence from the atomic scale to the atomic bomb survivor scale (including animal experiments).

A theory is “a well-substantiated explanation of some aspect of the natural world, based on a body of facts that have been repeatedly confirmed through observation and experiment.”

Now that describes BEIR VII's text very nicely.

So LNT is a theory in the context of our understanding of the breadth of scientific evidence.  It is much more than a simple hypothesis in reality.

If you scroll down within the "theory" link above you will find a discussion of theories as models.  Models tend to be more descriptive while theories tend to be both descriptive and explanatory.  In accord with those thoughts I think use of "model" makes sense when doing a calculation, ie. "we estimate the excess cancer risk to this population to be on the order of 0.0001% determined by employing the LNT Model of BEIR VII".  

I conclude and propose that use of the terms "hypothesis" and "model" should be narrowly used when appropriate.

In the most general sense, LNT is a scientific theory.

I'd be troubled if someone referred to LNT as "just a hypothesis". (Or as "just a theory" when the person confuses "theory" for  "hypothesis").  Likewise "just a model" is wrong, because we have much more than a descriptive model.