Fake PubMed Citation Tactics

A pattern catalogue. None of these are misprints — they are systematic ways of leaning on PubMed’s credibility without earning it.

Educational research-literacy content only. Not medical advice, not dosing guidance, not sourcing advice, and not a protocol for human or animal use. See our responsible information policy.

Pattern 1 — misquoting

A claim attributes a finding to a real PubMed-indexed study, but the study does not actually contain the finding. Either no figure supports it, or the study’s actual finding was the opposite or null. Always check the paper, not the claim.

Pattern 2 — citing unrelated peptides

A citation describes a paper on peptide X, but the page is making claims about peptide Y. The implicit logic is that “peptides do this,” which is far weaker than what the page reads as.

Pattern 3 — animal as human

A rodent study is cited, but the surrounding text presents the finding as if it were a clinical fact. Reading the abstract — even just the title — usually exposes this pattern. See: animal vs human.

Pattern 4 — old data presented as clinical proof

Decades-old pharmacology studies, or studies in obscure non-English journals, are presented as recent clinical confirmation. The age, venue, and methods are not flagged. Sometimes the paper is no longer findable.

Pattern 5 — vendor-funded claims

A study cited as independent evidence turns out, on inspection, to be funded by — or authored by — the vendor selling the product. Funding does not automatically invalidate findings, but it should be disclosed and weighed.

Pattern 6 — hallucinated citations

AI-generated content can produce citations that look real (correct author surnames, plausible journal names) but do not exist. PubMed lookup is the easiest check. See: AI use disclosure.

What this site does

Every citation on PeptideStacks should be verifiable — typically with a PubMed PMID or DOI. If you find a citation that does not check out, report it via our corrections policy.