Investigation Report: Effects of Vitamin K1 in Preventing Neonatal Vitamin K Deficiency Bleeding (Zhou Jieru)

Academic fraud report · Geng Detector

Summary

This report presents a comprehensive analysis of a paper published in the Chinese Journal of Medical Research (DOI: 10.12417/2811-051X.26.04.094) claiming that oral vitamin K1 drops outperform injected vitamin K1 in preventing neonatal VKDB. The authors conclude this is a confirmed case of academic fraud. Key findings include fabricated chi-square statistics (a claimed χ²=3.042, P=0.041 is mathematically inconsistent with 0/73 vs 3/73 incidence rates, where the true Fisher exact P≈0.12), self-contradictory t-values in Table 2, and impossible pharmacokinetic claims that contradict established pediatric knowledge. The study is also suspected of being disguised commercial promotion for the 'Jianminxing' brand. Confidence in the verdict is high given the multiple independent lines of evidence, though final determination requires institutional investigation.

Verdict

🔴 Confirmed fraud – Multiple independent lines of evidence indicate deliberate data fabrication and statistical manipulation.

Key findings

Statistical fabrication (Finding 1): The reported χ²=3.042 (P=0.041) for VKDB incidence (0/73 vs 3/73, N=146) is mathematically inconsistent. With these small numbers, Fisher exact test is appropriate and yields P≈0.12 (not significant). The unadjusted χ²=3.06 matches the claimed value suspiciously, indicating reverse-engineered statistics to cross the 0.05 threshold.
Pharmacokinetic implausibility (Finding 2): Claiming that 0.03 ml/day oral vitamin K1 yields higher serum levels than 1 mg intramuscular injection in neonates contradicts established pediatric pharmacology. The neonatal gut has poor absorption of lipid-soluble vitamins; IM/IV routes are clinical standard precisely because of this.
Inconsistent t-values (Finding 3): For Day-7 PT (11.25±0.86 vs 12.35±0.92), the calculated independent-sample t should be approximately 7.48, not the reported 6.452. At least one component of these numbers (means, SDs, or t-value) is fabricated.
Copy-paste residue (Finding 4): The conclusion contains duplicated phrasing ("比常规维生素K1制剂比常规维生素K1制剂"), indicating careless template editing typical of low-effort, mass-produced manuscripts.
Fictitious multi-center design (Finding 5): The paper claims a multi-center study but lists only the author's private pediatric hospital (Guangzhou Tian'er Children's Hospital), which typically does not conduct routine deliveries.
Mismatched citation (Finding 6): Reference [2] concerns vitamin A and D deficiency, not vitamin K1 interventions, indicating indiscriminate citation padding.

Evidence highlights

DOI: 10.12417/2811-051X.26.04.094
VKDB incidence reported as 0.00% (0/73) observation vs 4.11% (3/73) control
Claimed χ²=3.042, P=0.041 vs computed Fisher exact P≈0.12
Day-7 PT: 11.25±0.86 vs 12.35±0.92, claimed t=6.452 vs computed t≈7.48
Intervention: 0.03 ml/day oral "Jianminxing" vitamin K1 vs 1 mg IM standard
Study period: 2023–2025, 146 neonates
Sole affiliation: Guangzhou Tian'er Children's Hospital

Notes

The investigation strongly suspects disguised commercial promotion for the "Jianminxing" (健敏星) oral vitamin K1 product.
Recommended actions include contacting the author for raw data and SPSS output, posting on PubPeer, reporting to the journal editorial office and the author's institutional ethics committee, and flagging potential violations of medical advertising regulations.
This report was AI-assisted; final determination of academic misconduct requires formal institutional investigation. The authors note the possibility of false positives and respect the author's right of reply.