Clinical‑trial literature may contain a non‑trivial share of fabricated or irreproducible trials, so routine forensic audits (random raw‑data checks, statistical integrity screening and mandatory provenance deposits) should be implemented as a condition of publication and regulatory acceptance. Such audits would combine statistical forensics with mandatory access to trial records to catch fabricated datasets and prevent sham trials from informing care.
— If adopted, forensic auditing would shift where trust is placed—from reputation and peer review to verifiable data provenance—and could materially change drug approvals, clinical guidelines and patient safety.
2023.07.18
100% relevant
Nature cites John Carlisle’s estimate and multiple investigations reporting that up to ~25% of trials in some fields may be problematic, and reports researchers urging stronger scrutiny and checks by journals and regulators.
← Back to All Ideas