The discussion section: the biomedical researchers’ time to shine
This part of your manuscript should contain the appropriate content in the appropriate structure and also be convincing and easy to understand.
Question: I find writing the methods and results sections of manuscripts a lot more straightforward than the discussion section. When I write up my process or my findings, all I’m doing is reporting what I did or found; this work comes pretty easy for me. In the discussion section, though, I find that I struggle. What makes for a good discussion section? – Anonymous, Cardiology.
Dr. Editor’s answer:
One of my favourite, most empathetic descriptions of reading journal articles comes from a 2013 paper: “The reader passes through a test of boredom while reading paragraphs of the discussion section” (Şanlı et al., 2013). Poor reader!
My hope, dear letter-writer, is that your discussion section won’t serve as a test of boredom for your reader. After all, if you want your work to be influential, cited, or implemented, then you should strive to craft a discussion section that helps your reader to understand why your study is relevant to your reader and important in your field – and even, potentially, in fields beyond yours.
In short, the purpose of your discussion is to articulate “what the results mean rather than what they show” (Ghasemi et al., 2019, emphasis added). And so it makes sense that you’d struggle with this section – it requires a shift from observing to explaining, and doing so credibly.
There are lots of resources out there that advise how to write discussion sections. I like Vieria et al. (2019)’s guiding questions, PLOS’s Dos and Don’ts, and Höfler et al. (2018)’s seven specific and four general recommendations. These resources provide guidance on what content should go into your discussion and how you should organize that material.
But you asked about a “good” discussion section, dear letter-writer, and to my mind, a good discussion section contains the appropriate content in the appropriate structure and is also both convincing and easy to understand – so that the importance of your work sticks around in your reader’s memory.
To learn more about what makes for a good discussion section, I interviewed Emily Lam, an editor who specializes in supporting biomedical researchers. Emily is one of the three coaches I’ve hired to support participants in the May/June 2024 section of “Becoming a Better Editor of Your Own Work.” Here’s what Emily advises:
- Reduce wordiness
In the BMJ’s Author Submission Toolkit, deputy editor Trish Groves refers to the discussion section as the one in which “editors and reviewers describe [… as] usually too wordy and often contain[ing] non-critical information” (2010, p. 1976).
Some strategies for reducing wordiness include cutting the verb “is” and reducing the number of nominalizations and hollow verbs used in your work. Emily also recommends reviewing the length of your sentences: “If your sentences are not concise,” she told me, “your reader might have a hard time understanding your key messages (or worse, they may not understand the significance of your research).”
I developed the free tool writingwellishard.com to help academics identify the “to be” verb, nominalizations, and sentence length patterns in your writing; it doesn’t capture or keep any text popped into its text boxes, so consider experimenting with it as you revise your discussion section to reduce wordy expressions.
- Identify and clarify abstractions
Emily notes that clinical researchers frequently use abstract language – words that are hard for the reader to picture in their mind, like “evaluation” or “development.” The tendency to do so is understandable. Consider, for example, Steven Pinker’s description of a neuroscientist writing about their work:
The mental movie of a mouse cowering the corner of a cage that has another mouse in it gets chunked into ‘social avoidance.’ You can’t blame the neuroscientist for thinking this way. She’s seen the movie thousands of times; she doesn’t need to hit the PLAY button in her visual memory and watch the critters quivering every time she talks about whether her experiment worked. But we do need to watch them, at least the first time, to appreciate what actually happened
Dr. Pinker argues that academics like this neuroscientist tend to use abstract language in their writing because they “chunk” together discrete events – the many instances of mice cowering and quivering – into broader categories like “social avoidance.” Yet, as he also notes, many readers won’t have done this same “chunking,” or will have different conceptual “chunks” of knowledge, and so will need to “watch” the mouse avoid its peer at least once in a piece of writing in order to follow the writer’s chain of thought. To put it another way: language that is concrete – almost tangible – is much easier to follow, because it can be pictured in the mind’s eye.
As Emily explains, “People who read your journal article might be doctors, nurses, researchers, or students. Each reader comes with their own knowledge and personal experiences. When you use abstract language, you make it harder for these diverse readers to understand your work, which in turn makes it harder for them to remember, cite, or use your findings.”
When you edit your own writing, strive to introduce concrete language in the place of, or as a supplement to, abstractions by:
- using physical or sensory detail to help your reader to “see” what you “see”:
- Cardiovascular comorbidities in Type 2 diabetes can significantly impact quality of life is a sentence with few physical or sensory details.
- Cardiovascular comorbidities in Type 2 diabetes can lower – and even rein back – quality of life is a sentence with an abstract subject, “cardiovascular comorbidities,” that is brought to our senses through the animated, concrete verbs “lower” and “rein back.”
- ensuring that at least one sentence per paragraph has a human in the subject position of the sentence. So, for example:
- Cardiovascular comorbidities in Type 2 diabetes can significantly impact quality of life is a sentence with an abstract subject, “cardiovascular comorbidities.”
- People with Type 2 diabetes who develop cardiovascular comorbidities may have a lower quality of life is a sentence with humans, “people,” in the subject position.
- using active and passive voice strategically
- Passive: Cardiovascular comorbidities were reported. This sentence is short but unclear. Were these reports made by patients, clinicians, other researchers, or some other entity?
- Passive: Cardiovascular comorbidities were reported by patients with Type 2 diabetes. This sentence has all the information, and so is clear, but is unnecessarily long.
- Active: Patients with Type 2 diabetes reported cardiovascular comorbidities puts the person doing the action, “patients with Type 2 diabetes” in the subject position in the sentence – that’s what makes it an “active voice” sentence. This version also reduces the word count by 20 per cent.
- providing a specific example along with definitions or explanations of new or niche terminology:
- If the sentence “Lifestyle modifications can lower the risk of cardiovascular comorbidities” were included in a patient information PDF, it would be hard for the reader to know what to do differently, and what’s at stake.
- In contrast, consider the pair of sentences, “Lifestyle modifications such as physical activity and a balanced diet can lower the risk of developing cardiovascular comorbidities. When a person with Type 2 diabetes joins a baseball team, goes bowling regularly, and commits to meat-free Mondays, they reduce their chances of developing diseases in the arteries of their heart.” These sentences take up more space in the document, but make it easier for the patient to understand what they need to do to avoid developing dyslipidemia.
Want to dive deeper into editing your own work? Join Emily for the May/June section of “Becoming a Better Editor of Your Own Work,” facilitated by me!
Featured Jobs
- Vice-President Research & Scientific EngagementMS Canada
- Politics and Public Administration - Assistant Professor (Public Policy)Toronto Metropolitan University
- Economics - Associate/Full Professor of TeachingThe University of British Columbia
- Fashion - Instructional Assistant/Associate Professor (Creative & Cultural Industries)Chapman University - Wilkinson College of Arts, Humanities, and Social Sciences
- Public Policy - JW McConnell Visiting ScholarMcGill University
Post a comment
University Affairs moderates all comments according to the following guidelines. If approved, comments generally appear within one business day. We may republish particularly insightful remarks in our print edition or elsewhere.