Writing Clinic: Nuts and Bolts of the Manuscript, Part 3

Michael Gerchufsky, ELS, is the managing editor of Consultant. E-mail him with thoughts on this post at

In previous Writing Clinic blog posts, part 1, part 2, and part 3 of “Choose Your Topic Carefully,” I recommended a few first steps to contributing an article to Consultant. In part 1, part 2, and part 3 of “Choosing an Approach,” I discussed the array of possible article formats in the journal and other tips for approaching the writing of an article. I also advised about contacting the editor and performing a literature search. Part 1 and part 2 of “Nuts and Bolts of the Manuscript discussed article lengths and article extras, respectively.

With this post, I’ll offer advice about constructing a proper annotated reference list.

Annotating and referencing an article often presents writers with a significant and unexpected challenge. In my experience, the challenge is part technological (that is, the reference tools or footnote tools in Microsoft Word and other word processing programs can be complicated and confusing to use), part organizational (as I mentioned in a previous post, keeping track of the results of the literature search as you go is critical to knowing which information came from which source), and part commonsense (if you wouldn’t trust an information source to be authoritative enough to help diagnose a patient’s illness in clinic, you shouldn’t trust it as a source of information in a clinical article).

The two parts of the reference process—the annotations in text, and the corresponding list of references at the end of the article—both present potential pitfalls.

It’s very easy to get almost hopelessly off-track with the annotations in the text, which direct the reader to sources in the reference list. While not all of my editor colleagues will agree with me, my most emphatic advice to you about this process is this:

Do NOT NOT NOT use the automatic footnote feature in Word!

Using the footnote feature essentially places reference information in a separate area that is not technically a part of the text itself and that can be hard to find and update. Moreover, Word’s footnote feature doesn’t cooperate well with the publishing software that most publications use. Instead, I recommend placing annotations in text using parentheses, like this.(1)

It may be easiest to use author names in place of numbers as you write and edit the article, then convert the names numbers as a last step. That way, you can be sure that a given bit of information is linked to the proper corresponding resource. Note that a number of sources on a given topic very often have the same first author, so it’s important to distinguish which article by a given author corresponds to which facts. In these cases, consider using the first author’s name along with the date, the journal name, or a portion of the article’s title to differentiate among the references.

Like many other medical publications, Consultant uses the AMA Manual of Style for reference list style. If you don’t have access to this manual, please provide as much information as possible about the sources you’ve used, especially for electronic references, as well as for other more less common sources of information (eg, databases, podcasts, social media posts, one-off monographs, etc).

(Pro tip: The National Library of Medicine's PubMed site recently added a citation feature that automatically formats a an article's information in copy-and-paste-ready text. When you find an article that you'd like to cite — I chose this one at random, — click the blue "Cite" box on the right of the page. In the bottom right corner of the pop-up box, select "Format: AMA," and then click "Copy" on the bottom left corner of the pop-up box. Now you can easily paste the preformatted AMA-style citation directly in your document.)

Before you get your sources squared away and annotated in the text, however, let’s discuss a few larger points about sources used as references. I don’t want to dwell on the obvious, so I’ll boil it down to a few bullets.

  • Be careful and meticulous: Garbled or incomplete references send readers (and editors) on wild goose chases.
  • WebMD, Medscape,, etc, are not acceptable primary references. They can be a good starting point, but these sites generally are compendiums of information from other sources; thus, it’s important to determine the source of information at these sites. Always trace statistics, epidemiologic data, and other information back to the original source. "Whisper down the lane" is real in the medical literature, and information can degrade when one author cites another author who in turn had cited the original author's data or observation.
  • Subscription-only services (UpToDate, ClinicalKey, VisualDx, etc) generally are not acceptable primary references, for the same reasons as WebMD and others, but also because they are not easily accessed for verification by editors, who likely do not have a subscription, and they may not be accessible to a large number of readers, too.
  • Links to the most recent medical literature discussing a given medical topic is a click away on PubMed at
  • Because medical knowledge changes so rapidly, textbook citations should be rare; if you do use a textbook as a source, be sure that you have consulted the most recent edition. 
  • Wikipedia is not an acceptable reference in an article submitted to a medical journal. Ever. (Yes, you’d be surprised.)


As always, when in doubt, ask the editor. I’d be happy to offer advice and information about references before you submit the article.

In the next few blog posts, I’ll cover what to expect after you’ve submitted your article to Consultant for consideration. Thanks for reading!