We read with interest your comments facing the future of medical journals.1 Many of the lessons from patient safety are applicable to mistakes in the medical literature. Unintentional errors should be used as learning opportunities and editors (and authors) should welcome their detection. Retractions and corrections are rare2 and (as in clinical medicine) probably massively under-reported due to concerns about embarrassment and discomfort by authors.
Errors may vary in degree and type from minor typographical, system error (study design and execution) to the intentional acts of scientific misconduct (plagiarism, research fraud, duplication of publication). One study found that the highest number of retractions were in the following journals: Science; Proceedings of the National Academy of Sciences; and Nature.2
Perhaps differential corrections rates reflect the quality of postpublication scrutiny and the rigour of its readership. It has been suggested that corrections should be prominent, informative and citable, and mentioned within the table of contents.3
We conducted a study to assess the number of corrections in the following popular medical journals published in 2008: New England Journal of Medicine (NEJM); Journal of the Royal Society of Medicine (JRSM); Journal of the American Geriatrics Society (JAGS); British Medical Journal including Careers Focus (BMJ); Lancet; and Age & Ageing. We also looked at how each journal dealt with corrections, as summarized in Table 1.
Table 1.
Corrections
| Journal | Total number of periodicals in print in 2008 | Total number of corrections articles in print in 2008 | Listed in table of contents | Font size* | Layout | Caption |
|---|---|---|---|---|---|---|
| NEJM | 52 | 46 | Y | Smaller | Main text | Corrections |
| BMJ | 51 | 22 | Y | Same | In box | Corrections/Clarification |
| Lancet | 51 | 59 | Y | Smaller | In blue box | Department of Error |
| JRSM | 12 | 5 | Y | Same | Main text | Erratum/Errata |
| JAGS | 12 | 9 | Y | Same | Separate page | Erratum |
| Age & Ageing | 6 | 2 | Y | Same | Separate page | Corrigendum |
Font size of corrections article when compared to font size within rest of journal
We found that the journal with the most corrections was also one of the most regularly issued. The vast majority of corrections were typographical errors of minor significance.
It was interesting to compare how different editors treat corrections. JAGS appears to invite reader participation by starting its ‘Erratum’ page with ‘Dear Readers’, while Age & Ageing’s title of ‘Corrigendum’ highlights authority, severity and perhaps unintentionally hides its meaning to those unversed in Latin.
The publishing of corrections allows us to see what is perhaps only the tip of the iceberg of mistakes. We should anticipate more corrections as journal editors strive to demonstrate transparency and probity while at the same time readers strive to ensure that the scientific record is free from error.
Footnotes
Competing interests None declared
References
- 1.Abbasi K. A new world order for medical journals. J R Soc Med 2009;102:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Nath SB, Marcus SC, Druss BG. Retractions in the medical literature: misconduct or mistakes? Med J Aust 2006;185:152–4 [DOI] [PubMed] [Google Scholar]
- 3.Snodgrass GL, Pfeifer MP. The characteristics of medical retraction notices. Bull Med Libr Assoc 1992;80:328–34 [PMC free article] [PubMed] [Google Scholar]