Considerations for writing the Discussion section

Rossella Ferrari is a Biology graduate and a consultant medical writer for medical communication and editorial projects in many specialistic medical fields. She has been a member of the European Medical Writer Association (EMWA) since 2012.

In this blog post Rossella Ferrari gives tips for writing the Discussion section of scientific manuscripts, i.e., original research articles, case reports, and literature review articles. 

The Discussion section of papers often seem to be the most difficult to write. These considerations also attempt to identify a fil rouge (common thread) running throughout the texture of the manuscripts.

The overture: the value of the findings. The Discussion section provides the meaning of the research work and is the place where the research question - covered swiftly in the Introduction - can find answers. For this reason, the findings can be summarized as statements at the beginning of the Discussion, bridging it to the previous Results section. The findings are composed of data stemmed from the research question, and the Methods used to produce them. The Discussion should start highlighting the strengths of the findings that answer the research question based on novelty or additional or contradictory evidence.  The weaknesses can be mentioned but are expanded in the final part of the Discussion dedicated to the limitations.

The corpus: a) the findings drive the Discussion. The tenet of the Discussion section is comparing the findings of the study with those found in the literature. To this aim, when available, the recent publications that provide advancement compared to older literature, are the most relevant. The comparisons of the findings highlight differences and similarities with other studies. These analyses can also include methodologic issues, such as the type and size of samples or populations, protocols, techniques, and the rationales chosen in the various settings. Also, statistical tests can be discussed. 

The corpus: b) advancement, not competition. While rigour, completeness, and relevance versus the research question characterize the descriptions of Methods and Results, the critical approach is the heart of the Discussion. Consequently, the more critical analyses are in the Discussion, the more substantial the contribution is to the final implications or recommendations, and the more effective is the research. 
Outlining the context of the research is also of paramount importance for generalizing or inferring the insights achieved during the study. The findings of the previous studies should report the features of their settings. A finding that is aligned with results observed in similar settings or conditions can strengthen and built on the previous evidence. Findings that are different or contradictory with the previous studies can drive further analysis of the sources of differences. 

The corpus: c) list & rank. Of course, the composition of the Discussion may vary, but it is comfortable to follow a pathway. The findings to discuss can be ranked during the writing of the Results section. Incidentally, the sequence of the Results (and Discussion) should also be parallel to sequence in the Methods. Then, grouping the main findings (primary endpoints) in blocks can ease the organization and the structure of the Discussion. 
In the Introduction section, some publications may have been already used as background references to set the scenario. However, in the Discussion, the findings of previously published researches are directly compared with those of the current study. A proper literature review can yield additional research articles for the Discussion. 
When the findings are numerous, they can be marked with subheadings to drive the pathway of the text. In some cases, I suggest citing tables or graphs or images as it can make it easier to follow the discourse.                       
The secondary findings (secondary endpoints) of the study can be treated after the series of the main topics. At this level, the most acknowledged guidelines and recommendations can support the assessment of the findings.

The final part: limitations and implications. The limitations of the study should be clearly recognized. As this can be used for transparency, the limitations help the readers understand the context, and ease the work of the reviewers of the manuscript. The limitations can also orient towards further specific researches to complete the answers to the primary question of the current study. 
The implications of the findings that can be practical, i.e., providing suggestions for public health administrators, or theoretical that lead to new research fields. Implications and recommendations should consider the target audience(s) of the manuscript that will be published and spread. 

See figure 1 below to visualize the steps:

The final part of the Discussion ends with the distilled statements derived from the previous parts and allowing a natural transition to the next Conclusion section. 

The Discussion section brings to light the value of the results by a structured comparison with the previous findings in the field. A balanced assessment of strengths and weaknesses of the study is the best tool to pave the way for the next research.