Toxic exposure and medical-harm cases are won or lost on data most readers will never read directly — epidemiological studies, dose-response curves, regulatory toxicology reports. The public story has to convey what that data means without becoming a distortion of it.
The failure mode on one side is oversimplification that a defense expert can credibly call misleading — a single correlation presented as certain causation, for instance. The failure mode on the other side is technical accuracy so dense that no reporter runs it and no reader absorbs it. Both failures cost the case something real.
Every simplified claim in a public account is checked against the underlying study or filing before it goes out, with the same scientific caveats intact even in plain language — "associated with an increased risk of" doesn't become "causes" just because the plain version is punchier. Where the science supports a specific comparison a reader can picture — concentration levels next to a regulatory safety threshold, for instance — that comparison does more work than a raw statistic. Simplified doesn't mean less true. It means true in a sentence a reader will actually finish.