Writing an original research paper

It is good to know how to write an original research article because research is finished only when published. Original research articles follow the IMRaD structure, incorporating the introduction, methods, results, and discussion. The introduction briefly outlines your research problem and explains the rationale behind your study. The methods section describes how you conducted your research and analyzed your data. The results report your findings, whereas the discussion puts them in a broader context. Although you may like to start reading research articles from the introduction, you better begin writing your paper from the results and follow up with the methods, introduction, and discussion.

Results

Results are the essence of an original research article, with all article parts revolving around your results. Thus, start writing your paper from the results section. Please see this post for tips on writing a good results section before reading on. Your results should have a main message, which will be the backbone of your paper. The main message in our example report may be put like this: 

High baseline monoclonal protein concentrations were not associated with overall survival but increased the risk for multiple myeloma.

Having your main message will help you organize all your manuscript sections, which should have a presentation parallel to that in the results section. For example, if the results section first reports your cohort characteristics and later the findings on overall survival and the risk for multiple myeloma, you should keep that same order in other manuscript parts.

Methods

The methods describe your study design and all its procedures. In clinical studies, you usually explain how you selected participants, collected the data, and carried out statistical analyses. The methods in pre-clinical studies might be more complex depending on your experiments. In any case, the description should provide sufficient detail for other researchers to replicate your study. Your methods should mirror your results. For example, start by describing your recruitment criteria and later explain how you ascertained the date of death and diagnosed multiple myeloma. Likewise, keep the same order when explaining the statistical methods you used to analyze these outcomes.

Introduction

The introduction presents a linear chain of arguments justifying your research. This part of your paper is not an exhaustive monograph, so mention only the facts that serve your reasoning. Readers from English-speaking countries expect explicit, linear reasoning that goes straight to the point. In other countries, however, people might use circular reasoning and expect the reader to distill the meaning themselves. Whatever country you come from, when writing in English, always use linear reasoning and take full responsibility for readability, never making the reader grind away at your paper to get your message. Begin the introduction with an outline of your research but do not mention commonly known facts, like defining monoclonal gammopathy for the readers of Multiple Myeloma. Instead, start your introduction with relevant information: 

Monoclonal gammopathy of undetermined significance affects 5% of people over 70 years and is associated with a shorter life expectancy and an increased risk for multiple myeloma. 

Then, identify gaps in the existing evidence that justify your study: 

Because predictors of overall survival and conversion to multiple myeloma remain unknown, we investigated the association of monoclonal protein concentrations at diagnosis with these outcomes. 

Try to keep the introduction parallel to your main message. When citing previous research, start from studies on overall survival and then mention the ones on multiple myeloma risk. 

Discussion

The discussion is the best section to read but the hardest to write. In the discussion, which is longer than other manuscript parts, show you are knowledgeable about your research problem. One feature of poorly written papers is a short discussion, which only repeats the results and gives no insights. Suggestions to shorten the introduction and expand the discussion are common in peer review. The discussion first summarizes your main findings and then puts them in a broader context. The first paragraph usually repeats your main message: 

In our study with over 10,000 person-years of follow-up, higher concentrations of monoclonal protein were unrelated to overall survival but increased the risk for multiple myeloma.

The following paragraphs compare your findings with those from previous studies, keeping the presentation parallel to your main message. In one paragraph, compare your sample with other cohorts, and in the next, refer to studies on overall survival and multiple myeloma, respectively. When bringing up your findings in the discussion, be more general than in the results section. Write one-third instead of 36.6% or over two-fold instead of 2.22. The second-to-last paragraph lists the limitations of your study, like a retrospective design, and its strengths, like a large sample or long follow-up. The last paragraph presents your conclusions, which usually mirror your main message: 

Higher monoclonal protein concentrations do not explain shorter life expectancy but are associated with a greater risk for multiple myeloma in patients with monoclonal gammopathy of undetermined significance. 

The last sentence of the discussion typically suggests improvements for future studies, like longer follow-up or a prospective design. 

Bringing all manuscript parts together 

Once you have written all sections of your manuscript, put it aside for a few days and revise it afresh. A good manuscript describes a logical history but must also read well, which you can achieve by avoiding verbosity, using unambiguous language, and writing sentences in the active voice and first person. All manuscript parts should flow from one to another without repeating information or requiring the reader to backtrack. Remember that you write to inform, not to impress, so keep it short and simple. When editing your manuscript, you should be able to cut it down by 10% to 20% without affecting your message. An informed reader must understand your manuscript on first reading, so ask such a person to read your manuscript and find what parts seem unclear. Be sure that if a reviewer cannot understand your manuscript on first reading, they will likely reject it because the author, not the reader, is responsible for readability.

Abstract

Write your abstract when your manuscript is nearly ready, making sure it is the most polished part of your paper because, for most readers, including editors and reviewers, it is the first or the only piece they read. Your abstract must be standalone so readers can understand it without the main text. Typically, you have only 250 words to describe why and how you conducted your study, report your main findings, and give your conclusions. More and more journals require now structured abstracts with the following headings:

• Purpose: Briefly explain your study rationale.
• Methods: Describe your study design, participant recruitment, endpoints, and statistical methods.
• Results: Present your key findings, including specific numbers, like means, standard deviations, effect size measures, confidence intervals, and p-values.
• Conclusions: Briefly explain the relevance of your findings.

Conclusions

It is best to write an original research article in the following order:

• Results with a clear main message
• Methods explaining your study procedures and analyses 
• Introduction that justifies your research 
• Discussion putting your findings in a broader context
• Abstract that is a standalone summary of your research 

Read this editorial for more tips on writing a good original research article.

Language tips:

• Be brief
  • “Already the ancient Egyptians knew that dementia was common in old age.”
  • Better: “Dementia is common in old age.”
• Use simple language instead of jargon
  • “Patients with stroke are considered as experiencing a longer duration of falling asleep.”
  • Better: “Patients with stroke may need more time to fall asleep.”
• Do not make the reader backtrack
  • Hospitalization risk was greater in group A than in groups B and C.” Which group is which?
  • Better: “Hospitalization risk was greater in the placebo group than in the groups receiving 10 mg and 20 mg of the wonder drug.”
• Be unambiguous
  • “Men chose surgery more frequently than women.” Were women an alternative to surgery?   
  • Better: “Men chose surgery more frequently than did women.”
• Avoid abbreviations
  • “BP decreased after 6 months of CR in patients with AF and IVS width  > 1.1 cm.“
  • Better: “BP decreased after 6 months of cardiac rehabilitation in patients with AF and interventricular septum width > 1.1 cm.”  
• Use active voice
  • “Patients were given insulin injections.“
  • Better: “Patients received insulin injections.”
• Use first-person sentences
  • “The authors of the study hypothesized… “
  • Better: “We hypothesized…”
• Avoid negation
  • “Patients were not included.“
  • Better: “Patients were excluded.”
• Mind the commas around that and which
  • “All scientific journals that are worth publishing in send manuscripts for peer review.” Meaning: All respectable scientific journals use peer review. Correct!
  • “All scientific journals, which are worth publishing in, send manuscripts for peer review.” Meaning: All scientific journals are respectable and use peer review. Incorrect!