This is extra explanatory material for the series. It need not be read to understand the series.
The topic is causal criteria.
What are some of the criteria that epidemiology considers when assessing a causal connection (that A causes B):
1. Strength - the range of doses with which we see cancer induction (as opposed to another effect).
2. Consistency - cancer induction at low radiation doses is consistent to what is observed at higher doses.
3. Specificity - Is there any other effect which is observed?
4. Temporality - the cause most precede the effect. This is difficult to apply to radiocarcinogenesis.
5. Biologic gradient - this is the essence of the dose response curve, easily visible for radiocarcingenesis.
6. Plausibility - how do alternative explanations rank in plausability, relative to what is known?
7. Coherence -which alternative explanations yield the most coherence with what else is known?
8. Experimental evidence -what have experiments shown us as we've continued to resolve an effect under a lowering dose situation? In other words, to the extent our measurements improve, does LNT continue to be valid or do we see some other phenomenon?
9. Analogy - what other model analogies exist? How does radiocarcinogenesis compare to other similar toxics?
None of these things guarantee a 100% accurate understanding of a causal connection. One can still be wrong.
Some would claim that BEIR VII is just an "opinion" as if they just whimsically chose what to eat for breakfast that morning.
No, they applied criteria such as that outlined above in order to reach their conclusions.
Epidemiologists do this routinely with many different types of agents.
There is no "establishment" or "priesthood" out to intentionally keep radiation risks unnecessarily high.
Source
No comments:
Post a Comment