Does the paper this page cites actually exist, and does it say this? Every research bullet on the site was a paraphrase. None carried a PMID, DOI or exact title, so none could be checked by a reader. Each was resolved against PubMed and either replaced with the real record or deleted.
How it was checked
Each research bullet was queried against PubMed by title words.
A match required the actual article title, not a topical resemblance.
Matches found only by species keywords were rejected wholesale: that route paired a claim about sage extract and cognitive function with a paper on antihyperlipidemic effects, and once one pairing is that wrong the whole class is untrustworthy.
Unresolved bullets were deleted rather than softened. Pages that lost material say so.