BIOETHICS

The Bioethics Imperative XVI

Ethics and the Literature: Citations IV

“Mokita”: The truth we all know and agree not to talk about.

Scenario: Gettin’ Frustrated is a researcher entering a new area. She has wonderful data that she thinks belongs in Nature or Science, but when she does a background literature search, she finds either too many articles or none at all. Plowing through the articles she does retrieve, she finds only vague mention of her topic or search terms in Introduction or Discussion sections. She wastes a lot of time reading nonrelevant or tangential literature. Although her article is drafted, she delays finishing it because she cannot face the reference conundrum in which she finds herself, and she does not know how to resolve the problem. Her progress falters, her guilt mounts, and her collaborators become disgruntled.

Gettin’ Frustrated needs to get help from a librarian (see the seventh bullet below)!

In this issue of “The Bioethics Imperative,” we share our conversations about guidelines for ethical citations of the literature. Our guiding principle is that one doesn’t have to know everything but should know how to look for it. There is a method for proceeding with literature searches just as there are methods for designing experiments. In other words, we should think of our literature searches as experiments that require careful thought, detailed crafting, and perhaps multiple experimental approaches.

• Be as specific as possible: If you’re looking for information on seals, don’t look under “fur-bearing animals.”
• In a retrospective search: If a topic simply disappears before a certain date, find out if the topic was called something different before that date. An example is AIDS (see Bioethics Imperative XV).
• Before citing a reference that you find in an article: Verify it by locating the actual article and reading it. You must assess the relevance of that reference to your work and the accuracy of your statements about that reference before you cite it. Although it is tempting to cite opinions or conjecture from the Introduction and Discussion of an article, be aware that unless cited properly, such statements can take on the status of actual data in subsequent publications.
• Whenever your search turns up a relevant article: Keep the complete citation or you’ll spend time looking it up again later, if it’s needed as a reference.
• Read the relevant references cited in 25 of the most important papers you find: This is a gold mine of what others have found in their literature searches, so read these first! In an analogy to doing forward and reverse genetics, you can do forward and reverse literature searching in two steps. Use the bibliographies of papers cited in the most important papers to find other references going back in time. Use e-journal web sites to search forward in time (click the button labeled “find articles that cite this article”) to find more recent articles that cite the most important papers.
• For publication, cite your references exactly as specified by the journal in its “Instructions to Authors.” For example, in the widely used “Vancouver style,” journals such as Science must be cited with both volume and issue number. Some journals require the first and last pages of cited articles, and others require only the first page.
• Ask for help from the professionals! Your librarian has a master’s degree and is an expert in choosing appropriate sources and databases. Don’t expect yourself to be an expert at this: One cannot be faulted for admitting unfamiliarity and inexperience, but one can be faulted for trying to pretend or bluff one’s way through.
• Look for errata pertaining to the citations that you use. These were often hard or nearly impossible to track in the age of paper journals, but e-journals and web sites of authors may well make this task easier. You do not want to propagate mistakes made in the literature. So, just as you must file an erratum if you find a mistake in one of your own articles, you must check others’ articles for errata and cite these, too. Journals should find better ways to make links between a paper and its errata more accessible (MEDLINE does it for you).

The mokita here is that we all suffer from information overload and too little time. One solution is to make your literature searching as efficient as possible. Some people study this for a living! Librarians often keep handouts of online search guidelines; just ask. Below are some Internet resources for constructing searches adroitly:

General
http://newarkwww.rutgers.edu/guides/searching.htm
http://www.kcl.ac.uk/depsta/iss/schools/bdhmn/learning/searchstrategies.html

Web
http://www.learnwebskills.com/search/main.html

Thus far, we have discussed the responsibilities of authors. However, journals also have an ethical responsibility or a stake in the quality of the papers they publish. Many journals require a perjury statement confirming that data in an article have not been published elsewhere. More and more journals require that each author of an article submit a signed statement specifying any ties with commercial entities that might constitute a conflict of interest. Many newer journals have begun to require statements in the Acknowledgments that detail the efforts of each author (see the BioMedCentral journals) and/or demand structured abstracts that use specified content headings, e.g., Introduction or Objective, Design, Sample, Data Sources (databases and search terms used, and which years were searched in the databases as well as the date that those searches were done), Materials and Methods, Results or Main Outcome Measures, and Conclusions. Journals could also consider requiring all authors to sign a statement that at least one of them has read and verified each citation in their paper.

Next issues: Integrating bioethics scenarios into your teaching and research.

Tamara Turner
Librarian and editor, Seattle

Dina Mandoli
mandoli@u.washington.edu