IARC have a five-point scale for classifying agents in their monographs:
Note that IARC say: "The terms probably carcinogenic and possibly carcinogenic have no quantitative significance and are used simply as descriptors of different levels of evidence of human carcinogenicity, with probably carcinogenic signifying a higher level of evidence than possibly carcinogenic."
IARC have quite complicated but precise rules for deriving the overall classification from the separate classifications of the animal and human evidence. We summarise them here. We colour each box according to the normal classification it produces. Where there are alternative classifications, we list these in brackets - these often have to do with mechanistic evidence which is not captured in either the human or animal evidence, and there are one or two further exceptions as well which don't fit neatly in this presentation.
If you want to see the complete set of classification rules, including all these special circumstances and exceptions, set out as a flowchart, open the drop-down box:
Flowchart of the complete rules
On this scheme, the following table shows where ELF magnetic were classified. The epidemiological evidence was "limited" for childhood leukaemia and "inadequate" for all other cancers, so if they were classified separately, other cancers would be one box lower, as "unclassifiable". That is the same box ELF electric fields were in.