How to write the Discussion

Writing the Discussion section of a manuscript can be tricky. The Results and Methods are straightforward because you just write what you did and found during your experiments. However, the Discussion feels far less intuitive.

However, it doesn’t have to be complicated. Here are five tips for writing the Discussion part of your manuscript:

1. The beginning
   The last paragraph of the Introduction should contain the scientific question your study seeks to answer. In turn, the first paragraph of the Discussion should provide the answer to it. For instance:
   
   Last paragraph of the Introduction: 
   “In this phase 3 trial, we aimed to assess the safety and efficacy of Drug XYZ in patients with type 2 diabetes.”
   -> The question is: Is the drug safe and effective?
   
   First paragraph of the Discussion: “Here, we showed that Drug XYZ is effective in controlling the blood sugar levels in adult patients with type 2 diabetes, while maintaining an acceptable safety profile.”
   -> The answer is: Yes.

2. Order of topics
   Order your discussion points from most to least important. After answering the study question, quickly restate your most important result(s). Then, go into detail how your findings are in agreement or contrast with previous work. If you have some important secondary outcomes, treat them subsequently in the same fashion. Then, discuss less interesting/notable observations. Limitations usually appear at the end of the Discussion, directly preceding the Conclusions. 
   
   So, a rough structure of the Discussion could look like this:
   1) Answer to study question
   2) Restate main result
   3) Compare your main result to the existing literature
   4) Restate secondary result
   5) Compare your main result to the existing literature
   6) Repeat Steps 4) and 5) for each of your other secondary results
   7) Discuss any less interesting/notable observations
   8) Limitations
   9) Conclusions (if they don’t have their own section)

3. Organising and structuring
   The Discussion is usually 1,000–2,000 words long, which can be challenging to read. Thus, it would be best if you broke the text up into chunks. Many journals don’t allow subheadings in the Discussion, but you can still divide it into paragraphs without titles.

4. Discussing results
   The focus should be beyond the current study, meaning rather on interpreting the outcomes than on the actual data. The Results section has already covered the details of the findings. So, feel free to refer back to your results instead of repeating them in great detail. The same applies to discussing previous work/publications: A summary of the most relevant findings are enough. The most exciting part are the parallels and clashes between your current data and those preceding studies.

5. Vocabulary
   If your reader can follow your thought process, they are likely to understand your conclusions. Thus, you have to take your audience by the hand and lead them through your Discussion. The way to do this is to use transition words and phrases that signal your next sentence’s intention. For instance:

I hope these tips are useful to you. Let me know how you’re getting on, and:
Happy writing!