In "Checkpoints" in the February 9 issue of the New Yorker, John McPhee writes about the magazine's fact checkers. He describes one who attempted to verify his statements about the highly secretive Manhattan Project during World War II. She called the police at the last moment before publication to ask them to track down the only person who could confirm a detail in his article.
As I read about the lengths the New Yorker goes to in ensuring the accuracy of its articles, I asked myself why readers of AJN or any other nursing journal should expect any less. After all, errors in the New Yorker won't often lead to someone being maimed or killed. Errors in nursing journals can.
So how accurate are the nursing and medical journals you're reading? After 10 years at AJN, I've concluded that nursing's journalism is not what it ought to be, for several reasons:
Peer reviewers can't pick up all inaccuracies. We accept papers after blinded peer review and evaluation by a staff editor. But once we get into editing, we do what very few (if any) nursing journals can do: we obtain the source materials and check the author's facts against their sources. The number of errors and misinterpretations we detect is stunning. Perhaps it's the "publish or perish" syndrome that stresses some writers, but that's no excuse for sloppy writing. Even submissions of original research often have numbers that don't add up, requiring that authors rerun their data during editing. It seems that some authors throw in references to appear "scholarly." But overreferencing isn't the solution.
Many authors don't know the difference between primary and secondary sources. This is a huge problem for us. Several years ago, a senior nurse scholar submitted an article on a legal and ethical matter, using a secondary source to cite a well-known Supreme Court decision. Then our copyeditor read the original Supreme Court decision. He informed the author that her interpretation was inaccurate and sent her the original text from the court's decision. She was dumbfounded (though grateful): the literature was replete, she said, with the same misinterpretation of the court's decision.
In his New Yorker article, McPhee quotes a fact checker who said that a printed error "will live on and on in libraries carefully catalogued, scrupulously indexed [horizontal ellipsis] deceiving researcher after researcher down through the ages, all of whom will make new errors on the strength of the original errors and so on and so on into an exponential explosion of errata."
Students and nurses engaged in evidence-based practice projects are incorrectly being told to search the literature for the last three to five years. One paper we received took such an approach and missed a significant body of writing on the topic published 20 years ago. We rejected the paper. Limiting literature reviews to an arbitrary number of years sometimes wrongly suggests that the best work on the issue occurred recently.
Nonprofit and for-profit publishers of biomedical journals have an ethical responsibility to ensure the accuracy of what they publish. Of course, this means investing resources at a time when some publishers are struggling to survive-and fact checking isn't free. I don't have the answers to this challenge, but I do believe that there's no need for journal publishing if it can't be differentiated in quality from other writings, such as a lot of what is posted on the Web.
To help authors (and faculty members, who should be teaching nursing students about good writing), we've developed a statement about primary and secondary sources that's in this issue and online at http://ajn.edmgr.com, our Web site for manuscript submission. Please use it and share it with colleagues and students. I hope authors will do their part. And at AJN we'll continue to chase the New Yorker's standards for accuracy and readability.