Concerns Arising over Increased Retractions in Scientific Articles

Dusit/shutterstock.com

The number of scientific articles retracted increased 10-fold between 1988 and 2008 (from 0.002% in the early 1980s to 0.02% in 2005–2009), according to a paper published in the Journal of Medical Ethics.¹ The authors did note that the number of articles listed in Medline each year has also increased over time, from approximately 300,000 per year in the 1980s to more than 800,000 per year in 2009.

Concerns over the increasing number of retractions have also entered research in rheumatic disease. An article in the Journal of the American Medical Association on the effects of nitroglycerin ointment on bone density and strength in women originally published in 2011 was recently retracted after a committee at the lead author’s institution “found evidence of falsified and/or fabricated data.”²

“By and large, those contributing to the medical literature advance our understanding and do phenomenal work making the world a better place for our patients,” says Marian T. Hannan, DSc, MPH, professor of medicine at Harvard Medical School and editor in chief of Arthritis Care & Research (AC&R). “With any process, things can slip through, but with the two American College of Rheumatology [ACR] publications, we have our eyes open and minimize these concerns to the extent we can.”

Systems Closely Scrutinize Articles

Both AC&R and Arthritis & Rheumatology (A&R) have systems in place to closely scrutinize incoming articles for content, interest, significance and validity. This starts with the authors themselves.

“Most articles have several authors due to the complexity of medical data,” says Richard Bucala, MD, PhD, professor of medicine/rheumatology at the Yale University School of Medicine and editor in chief of A&R. “All authors sign a statement acknowledging they are qualified as authors, have reviewed all of the data and personally stand by the findings in the paper.”

First Look by Editors & Editorial Boards

The first cut is usually with the editor in chief and the editorial boards of the journals. They evaluate the scientific merit, the innovation and the significance of the findings. Only 20–50% of submissions make it past this initial evaluation, an indication of how stringent the requirements are.

Articles are sent to an editorial assistant to make sure they are formatted properly. Then, they are run through a software program called iThenticate, which searches the medical literature for plagiarism. The next step is the most important—peer review.

Peer Review an Important Step

“Like most upper-level medical journals, both of the ACR publications are peer reviewed,” says Dr. Bucala. “At least two, and sometimes more, experts look at the paper, the facts and the data, and give the editor assigned to that study their comments and critiques. When there is a need for statistical or other more specialized review, we have reviewers available as needed.”

Both AC&R & A&R have systems in place to closely scrutinize incoming articles for content, interest, significance & validity. This starts with the authors themselves.

All peer reviewers undergo an extensive vetting process of their own. Instead of picking the next person on the list, reviewers are assigned based on their expertise, both self-stated and published. Each is evaluated for the timeliness and thoroughness of their work after each assignment.