Credits: retractionwatch.com
Trusting peer-reviewed medical studies is something we take for granted. When we read one of these articles, we assume that the patients described are real, I mean it’s a simple, obvious assumption, right? It could be a rare case, complicated, incomplete but still real. After all, it’s on these articles that protocols, guidelines, and important clinical decisions are based.
And well, now, imagine that for 25 years, over 130 articles published by a pediatric journal described children and diseases... that never existed. This is what happened with Paediatrics & Child Health, the official journal of the Canadian Paediatric Society, so we’re not talking about some small-time newsletter. From 2000 to 2025, the journal published 138 “case reports” presented as real clinical cases. Only they were fake stories, completely invented, intended as teaching examples... the problem was that no one knew, and it wasn’t even stated.
How could people be tricked by those article is quite simple: the format was identical to normal peer-reviewed case reports, with symptoms, diagnoses, and detailed clinical discussions, how could it raise doubts?
All of this came to light only thanks to an investigation by the New Yorker, which exposed the 2010 case nicknamed “Baby boy blue”. The article described a newborn with alleged opioid toxicity through breast milk. The problem is that the case was cited for years by other researchers and used to guide clinical decisions, that's an absurdity! Only after the journalistic investigation it was discovered that the case had been completely invented by one of the authors. But the problem is not just this single case: dozens of these articles were cited in other studies, thus entering the scientific literature and influencing discussions, protocols, and decisions, spreading errors and wrong assessments far and wide. For years, doctors and researchers took as real something that was not, because readers interpreted them as normal clinical observations published in a peer-reviewed journal.
And as if that wasn’t enough, to cause even more damage, after the scandal when the journal decided to add the postscript “invented case”, some of these retroactive corrections even marked as “fictional” cases that were actually real, creating further confusion and harming scientific credibility. Professor David Juurlink of the University of Toronto commented:
A fictitious story presented as a real case is practically indistinguishable from fabrication in the scientific literature.
This post is not meant to show that science doesn’t work, however it shows that even peer-reviewed literature is human, and therefore subject to editorial errors and questionable decisions. Seeing this case, which went on for years, where clinical cases were presented as real without being so, it is legitimate to ask: how many other scientific “certainties” are taken for granted without ever being truly verified? How many therapies have been decided based on these invented articles?