Canada’s 25-Year Medical Fiction: Pediatric Journal Corrects 138 Fabricated Case Reports
Paediatrics & Child Health adds disclaimers to surveillance reports published as real patient data since 1999, exposing deep flaws in academic peer review and clinical guideline integrity.
A Canadian medical journal has issued corrections on 138 case reports published over 25 years to clarify what readers had no way of knowing: the patients described never existed. The cases, presented in the format of authentic clinical observations and indexed as genuine patient data, were teaching vignettes created by the Canadian Paediatric Surveillance Program but published without disclosure of their fictional status until February 2026.
According to Retraction Watch, Paediatrics & Child Health added correction notices to all 138 publications following a New Yorker investigation into a separate controversial infant death case. The journal’s editor-in-chief Joan Robinson confirmed the decision was made to “clarify that the cases are fictional,” though she maintained the articles covered “well recognized conditions” unlikely to generate citations or clinical interest.
The Canadian Paediatric Surveillance Program surveillance highlights were designed as educational tools for pediatric healthcare providers. When the journal first began publishing this article type, editors decided cases should be fictional to protect patient confidentiality. However, no disclosure appeared in the articles themselves, and readers encountered them as authentic case reports in the scientific record.
The Scale of Undisclosed Fiction
The 138 corrected articles span from 1999 to 2024 and cover conditions including congenital syphilis, fetal alcohol syndrome, serious ATV trauma, and hepatitis C infection. While not indexed in Scopus or Web of Science—limiting their integration into systematic reviews—the articles appeared in a peer-reviewed journal published by Oxford University Press on behalf of the Canadian Paediatric Society.
David Juurlink, professor of medicine and pediatrics at the University of Toronto, told Retraction Watch that regardless of author guideline statements, “the fact the cases are fictional should have been conveyed to the readers.” He characterized the practice as “functionally indistinguishable from fabrication in the scientific record” when fictional narratives are published as genuine case reports without disclosure.
The Catalyst: A Codeine Case Under Fire
The corrections emerged after scrutiny of an unrelated 2006 case report in The Lancet by pharmacologist Gideon Koren claiming an infant died from morphine poisoning through breast milk. That case, which influenced global codeine prescribing guidelines for nearly two decades, received an expression of concern in January 2026 following a year-long investigation by CBC News.
Juurlink and colleague Jonathan Zipursky demonstrated in a 2020 review that the toxicology results were “impossible consequences” of breastfeeding, noting the infant had codeine concentrations 100 times higher than breast milk could deliver. Two Canadian journals subsequently retracted related articles in 2020, citing “clear evidence that the findings are unreliable.” The Lancet case influenced Health Canada warnings, medication labeling changes, and clinical guidelines affecting millions of postpartum women globally.
The Peer Review Paradox
The incident exposes fundamental vulnerabilities in scientific publishing infrastructure. Research on fabricated data suggests that between 1.5% and 2% of papers published in 2022 showed textual similarities to known fake papers, according to a 2024 review in BMC Research Integrity. A separate analysis found problematic images—many intentionally fabricated—in 3.8% of articles published between 1995 and 2014.
According to Science, neuropsychologist Bernhard Sabel estimated up to 34% of neuroscience papers and 24% of medicine papers published in 2020 were likely fabricated or plagiarized—figures far exceeding the 2% baseline from earlier estimates. “Paper mills have made a fortune by basically attacking a system that has had no idea how to cope with this stuff,” Dorothy Bishop of the University of Oxford told the publication.
- Peer reviewers lack access to raw data and cannot verify patient existence
- Fictional cases covering “well recognized conditions” generate minimal citation scrutiny
- Non-indexed articles evade systematic review quality checks
- Correction notices depend on readers actively checking journal websites
Downstream Contamination
The integrity of clinical guidelines depends on the reliability of underlying evidence. Research published in BMJ in 2025 found that retracted trials substantially impact the evidence ecosystem, with meta-analyses changing direction or statistical significance when contaminated studies were removed. A separate study identified 127 systematic reviews and clinical guidelines citing already-retracted papers without caution, with less than one in 20 issuing corrections.
The Canadian case raises questions about how many fictional vignettes entered clinical decision-making pathways. While the journal maintains these surveillance highlights were not indexed in major databases and covered recognized conditions, the articles appeared in a publication distributed to 15,500 pediatricians and family physicians across Canada, according to historical records.
“A narrative that is fictional but published in the format of a genuine case report, without disclosure at the time of publication, is functionally indistinguishable from fabrication in the scientific record.”
— David Juurlink, Professor of Medicine and Pediatrics, University of Toronto
What to Watch
The Canadian Paediatric Society has updated its website to include disclaimers that clinical vignettes “describe fictional cases, created as teaching tools.” The journal committed to explicitly stating fictional status within case report bodies going forward. But the 25-year publishing practice highlights systemic issues extending beyond one journal’s editorial policy.
Major publishers including Springer Nature, Elsevier, and Wiley are developing detection tools through the International Association of Scientific, Technical, and Medical Publishers’ Integrity Hub, with 10 publishers expected to deploy paper mill detectors in 2026. However, the Canadian case involved deliberate editorial policy rather than external fraud—a distinction that existing detection systems may not address.
The incident also raises questions about journal retraction versus correction practices. While the journal added correction notices, the articles remain in the literature as published case reports with addendums rather than retractions—a decision that may influence how citing publications treat the underlying evidence.
For clinical practitioners and researchers, the episode underscores the necessity of verifying whether case reports in the literature represent actual patients or teaching constructs, particularly when such reports inform prescribing decisions affecting millions. The correction notices appeared only after external investigation forced the issue, suggesting passive reliance on journal integrity statements may be insufficient.