Is research safe in their hands?BMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d284 (Published 19 January 2011) Cite this as: BMJ 2011;342:d284
All rapid responses
The "Wakefield saga" and other prominent examples of fraud in
scientific research on the background of rising numbers of retracted
articles  demand in-detail presentation of current and past figures of
publication retraction rates, as a necessity to any discussion on this
Scientific articles may be retracted "by their authors, academic or
institutional sponsor, editor or publisher, because of pervasive error or
unsubstantiated or irreproducible data" . A significant upward trend in
annual retraction rates was first noted several years ago .
On 31 December, 2010, the PubMed database (United States National
Library of Medicine, Bethesda, Maryland, USA) was searched for retracted
publications. A total of 1633 publications were found, of which 1492
(91.4%) were English-language publications on MEDLINE database only. Only
ten retracted publications were identified on MEDLINE before 1 January,
1978. Average time to retraction ranged previously between 21 and 28
Hence, the search was limited to a 30-year period (1/1/1978-
31/12/2008) and was also limited to publications in English, which appear
on MEDLINE database, since these constitute the main body of retracted
scientific literature on PubMed, as demonstrated above.
A total of 1368 retracted publications, including 93 reviews (6.8%),
were identified. Notably, six primary authors were responsible for 99
retracted publications (7.2% of the cohort). As of 2003 and later, a
marked increase in retracted publications rates, including reviews, was
observed (figure). Additionally, only one clinical-oriented journal (New
England Journal of Medicine) was found in the top ten retracting journals
(by absolute number of retracted publications during 1978-2008), whereas
the rest were high-impact journals of the basic sciences.
The inclining trend in retracted publications may be correlated to
the continuous technologic development and rapid expansion of broad-band
World Wide Web access to the MEDLINE database. Such an access was made
available to a growing number of organisations and individuals in the past
decade, thus, increasing the exposure of publications to scientific
criticism by colleagues.
Inability to reproduce published results in peers' laboratories
serves as a strong signal for publication retraction; apparently, no real
equivalent exists in the clinical field. This may explain the imbalance in
retraction numbers between clinical and basic sciences journals,
especially when considering that meticulous pre-publication peer-review
selection exists in high-impact journals of both types.
The age of information may have already brought retraction rates to
higher peaks than ever before. This should be regarded as a favourable
evolution to the fundamental of science - Trust.
1. Marcovitch H. Is research safe in their hands?. BMJ 2011;342:d284.
2. Errata, Retraction, Duplicate Publication, Comment, Update and
Patient Summary Policy for MEDLINE? [fact sheet]. Bethesda, Maryland:
United States National Library of Medicine, 2010.
3. Cokol M, Ozbay F, Rodriguez-Esteban R. Retraction rates are on the
rise. EMBO Rep 2008;9:2.
4. Budd JM, Sievert M, Schultz TR. Phenomena of retraction: reasons
for retraction and citations to the publications. JAMA 1998;280:296-7.
5. Redman BK, Yarandi HN, Merz JF. Empirical developments in
retraction. J Med Ethics 2008;34:807-9.
Figure. Annual retraction rates of MEDLINE publications in English (n = 1368) including reviews (n = 93) during 1978-2008 (as at 31 December, 2010).
Competing interests: No competing interests
The UK Research Integrity Office (UKRIO) welcomes the BMJ's recent
articles and editorials on research misconduct as important contributions
to raising awareness of this subject.
However, I would like to make it clear that UKRIO has not ceased its
operations and continues to fulfil its remit to provide independent and
expert support on issues of research conduct to researchers, organisations
and the public.
A full response from our Chair can be found at
http://www.bmj.com/content/342/bmj.d378.full/reply#bmj_el_248567 and I
would direct readers to it.
To summarise, in 2010 alone we addressed more than 60 cases, showing
that there is a clear need for our services. The level of our case work
and the uptake of our other services demonstrates that whistleblowers and
organisations alike, which might be expected to be hesitant about sharing
problems with a non-regulatory body, are, in fact, willing to come forward
and seek guidance on difficult issues and value UKRIO's confidential
For those who might understandably feel frustrated at the state of
research integrity in the UK, a large proportion of UKRIO's staff are
experts who give their time to the project pro bono. Individuals and
organisations with experience in addressing fraud and misconduct in health
and biomedical research are welcome to collaborate with UKRIO.
James Parry, Acting Head, UK Research Integrity Office.
Competing interests: I am Acting Head of the UK Research Integrity Office