Abstract
BackgroundThe annual number of retracted publications in the scientific literature is rapidly increasing. The objective of this study was to determine the frequency and reason for retraction of cancer publications and to determine how journals in the cancer field handle retracted articles.MethodsWe searched three online databases (MEDLINE, Embase, The Cochrane Library) from database inception until 2015 for retracted journal publications related to cancer research. For each article, the reason for retraction was categorized as plagiarism, duplicate publication, fraud, error, authorship issues, or ethical issues. Accessibility of the retracted article was defined as intact, removed, or available but with a watermark over each page. Descriptive data was collected on each retracted article including number of citations, journal name and impact factor, study design, and time between publication and retraction. The publications were screened in duplicated and two reviewers extracted and categorized data.ResultsFollowing database search and article screening, we identified 571 retracted cancer publications. The majority (76.4%) of cancer retractions were issued in the most recent decade, with 16.6 and 6.7% of the retractions in the prior two decades respectively. Retractions were issued by journals with impact factors ranging from 0 (discontinued) to 55.8. The average impact factor was 5.4 (median 3.54, IQR 1.8–5.5). On average, a retracted article was cited 45 times (median 18, IQR 6–51), with a range of 0–742. Reasons for retraction include plagiarism (14.4%), fraud (28.4%), duplicate publication (18.2%), error (24.2%), authorship issues (3.9%), and ethical issues (2.1%). The reason for retraction was not stated in 9.8% of cases. Twenty-nine percent of retracted articles remain available online in their original form.ConclusionsRetractions in cancer research are increasing in frequency at a similar rate to all biomedical research retractions. Cancer retractions are largely due to academic misconduct. Consequences to cancer patients, the public at large, and the research community can be substantial and should be addressed with future research. Despite the implications of this important issue, some cancer journals currently fall short of the current guidelines for clearly stating the reason for retraction and identifying the publication as retracted.
Highlights
The annual number of retracted publications in the scientific literature is rapidly increasing
The reason for retraction was not stated in 9.8% of cases
Reasons for retraction not directly related to academic misconduct such as authorship issues and novel research based on retracted work, together account for less than 5% of all retracted articles in the cancer literature
Summary
The annual number of retracted publications in the scientific literature is rapidly increasing. Several types of research misconduct warrant the retraction of a scientific paper. These include plagiarism, duplicate publication, fraud, authorship issues, ethical issues, and error [1,2,3]. It is well documented that the proportion of published studies that are being retracted from the scientific literature is rapidly increasing [4]. A study by Grieneisen et al found that the number of annual retractions, adjusted for number of publications, increased by a factor of 11.06 over this 10-year period [6]. Retractions are a worldwide phenomenon as authors from multiple countries of origin have been found to be involved in research misconduct [7].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
