Correction, please! Study finds frequent errors in medical publications

"This is a clarion call to authors to make certain that the information they are publishing is as accurate and error-free as possible," said Paul Hauptman, M.D., professor of internal medicine at Saint Louis University and principal investigator of the study.

Often times, the errors in medical journals are minor such as misspelling of an author's name or mislabeling of a table. But sometimes articles have major mistakes in the text, figures or table that can alter the meaning of the information or findings. Hauptman and his team argue that authors and publications need to follow a uniform, more transparent system that reports these major mistakes and not bury them under the list of minor errors.

"We are also concerned about minor errors as they can be a distraction. When a major error comes along, people might not be focused on it because they are numbed by minor errors," Hauptman said. "Critical errors in the medical literature can often result in misappropriation of study findings by readers, especially if they don't subsequently review the published corrections."

Hauptman adds that at this point, it is not possible to measure how frequently a journal reader incorporates errata into clinical care or the extent to which patient outcomes may be affected.

For the study, which was published in The American Journal of Medicine, Hauptman and his team reviewed 20 prominent English language journals in the fields of general medicine and cardiology for an 18-month period and found 577 error reports. More than 24 percent reported an error that the research team rated as major. They also found that 51 percent of errors were not corrected in the original text or the errata report did not specify whether a correction was made.

"As the volume of research publications continues to rise, the scientific community needs to examine how it manages its mistakes," said Eric Armbrecht, Ph.D., associate professor at the Saint Louis University Center for Outcomes Research (SLUCOR). "Transparency, consistency and clarity are essential. Our study found that these are not common among some of the top medical journals."

One of the examples of a major error report in the study includes a paper about depression. In the original article, the incidence of new depression cases was misquoted by a factor of 10 (15.8% of women with new depression rather than the real figure, later corrected, of 1.58%).

The International Committee of Medical Journal Editors (ICMJE) has proposed that corrections should appear on a numbered page, be listed in the table of contents, include the complete original citation, and link to the original article and vice versa if online. However, this is infrequently done.

Armbrecht also points out that since many publications today have an electronic version of the article, it is not clear how often the online paper is corrected.

A number of possible reasons exist that might explain the presence of errors in publications but Hauptman and Armbrecht said the study could not definitively evaluate why errors occur. One possibility is that most authors do not read and edit carefully the final pre-publication version of their manuscripts.

"It's noteworthy that although final approval of an article may fall to the first or corresponding author, the criteria put forth by ICMJE specifies that each author must provide final approval of the version to be published," said Hauptman.