A prospective new author recently asked me, "why 5 years?"
The author had submitted an article for peer review and had been informed that her references were not up-to-date. To be clear, this is not about an article submitted to the Journal of Neuroscience Nursing. The author is a colleague who was somewhat angry, and I was fair game because she knew that I was an editor. In talking to me, she not so calmly voiced that she had worked very hard and it took her several years to complete her study. The articles she referenced were important works, including some from people who helped her complete her study. She did not feel that these important works should only be discarded simply because they were 6 or 7 years old.
Deep breath. There is a bit of confusion on the author's behalf, and it is important to understand 3 points: (1) 5 years is not an exact number, (2) there are times when it is okay (sometimes preferred) to cite old (classic) literature, and (3) recent literature works are in everyone's best interest.
As a general rule, most journals, indeed most scientific writing, stress the need for recent references and citations. Many authors and reviewers translate "recent" to mean "about 5 years." I searched PubMed and CINAHL, and to the best of my knowledge, there is no scientific evidence supporting that 5 is the right number. I found no evidence that 5 years is better or worse than 4.5 years, 6 years, or 5 years, 2 months, 3 days. In other words, the 5-year mark is essentially an arbitrary benchmark. The scientific community sort of came to agreement on 5 years. Perhaps, this relates to the 5-year impact factor; perhaps, this relates to the practice of publishing a new book edition about every 5 years; or perhaps, this caught on after 1 professor in 1 university once declared, "FIVE is the magic number."
Certainly, there are times when the author(s) should reference older literature. This can be the case when there is a need to reference the primary source. For example, the following sentence requires both a recent and significantly older reference: "The story of Peter Pan, which first appeared as a chapter in a somewhat darker novel, has been hypothesized to represent death."1,2 In the aforementioned example, I include a reference from more than 100 years ago because I need to draw the reader to the original material. I also include a reference to an article that summarizes (in a fascinating read) the possibility that Peter and the Lost Boys did not grow up because they had died.
Despite it being the case that 5 years is arbitrary, there is a need to reference the most recent literature because that is how science works. Imagine the confusion if an author were to reference care of the patient with stroke using references from 30 years ago. As neuroscience nurses, we recognize that new therapies, medications, and evidence emerge every week. Medical knowledge is continually expanding, and our profession benefits from up-to-date literature.
The author was partially justified in being upset because 5 years is not an exact benchmark. However, the reviewer who knows that more recent literature exists is equally justified in asking that references be updated-the important point being that the more recent literature is more likely to reflect the current sum of scientific thought on a particular issue.
References