Retractions in the medical literature: how many patients are put at risk by flawed research?

Abstract

BACKGROUND: Clinical papers so flawed that they are eventually retracted may put patients at risk. Patient risk could arise in a retracted primary study or in any secondary study that draws ideas or inspiration from a primary study. METHODS: To determine how many patients were put at risk, we evaluated 788 retracted English-language papers published from 2000 to 2010, describing new research with humans or freshly derived human material. These primary papers-together with all secondary studies citing them-were evaluated using ISI Web of Knowledge. Excluded from study were 468 basic science papers not studying fresh human material; 88 reviews presenting older data; 22 case reports; 7 papers retracted for journal error and 23 papers unavailable on Web of Knowledge. Overall, 180 retracted primary papers (22.8%) met the inclusion criteria. Subjects enrolled and patients treated in 180 primary studies and 851 secondary studies were combined. RESULTS: Retracted papers were cited over 5000 times, with 93% of citations being research related, suggesting that ideas promulgated in retracted papers can influence subsequent research. Over 28 000 subjects were enrolled-and 9189 patients were treated-in 180 retracted primary studies. Over 400 000 subjects were enrolled-and 70 501 patients were treated-in 851 secondary studies which cited a retracted paper. Papers retracted for fraud (n=70) treated more patients per study (p<0.01) than papers retracted for error (n=110). CONCLUSIONS: Many patients are put at risk by retracted studies. These are conservative estimates, as only patients enrolled in published clinical studies were tallied.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 156 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.