Making Your Career through Publishing Original Papers
To present a quick, simplistic overview of how the Discussion should be structured. I recommend that the first paragraph, especially the first sentence, should describe the main findings of the study or what was attempted. The second part should consist of supporting evidence from other papers correlating with your findings and explaining any discrepancies. The third part should handle and discuss the results of papers with results differing from yours, and again explain why that might be. The fourth section contains the limitations of your studies, for example, if it were done only in animals, or if the number of human subjects were limited, or if there was a bias towards a certain age group. And the fifth section generally presents the conclusion, which is more or less a reiteration, perhaps in slightly stronger language, of the message of first section.
To return to the first section, by initially restating the purpose of the study and what your findings were or what you set out to do, you are allowing the reviewer and reader to reset their minds, after the Materials and Results, regarding what your discussion will be about, and where you are going, so it is easy for them to follow. In addition you should remember that most reviewers are doing review work as volunteer scientists and as such they usually do it after the day’s work is finished, perhaps after dinner, when they are perhaps sleepy and tired, and it helps them to refocus their mind on the subject at hand. When I give talks to audiences containing fairly large numbers of people who review papers, and I ask them what time they do their reviewing, usually about 90% say late at night. Thus simplicity and clarity are essential.
In the second section, although it is not necessary to go into great detail about your own results you have already described in the Result section, you should give a thorough review of papers that are similar to yours and especially if, as I recommend, you are intending to suggest possible reviewers in your cover letter, be sure to include papers by those investigators in your Discussion. When the reviewers see his/her works cited in the paper, it disposes them more positively toward the paper they are reviewing. Again, if there are any differences in the findings between those papers and yours, you should try to explain why. But if possible, emphasize the positive similarity of findings. And you should also try to present mechanisms or explanations for the differing findings in your study and also in papers in similar or non-contradictory results.
It is also important to show that you are not biased by thoroughly discussing any papers that present evidence directly conflicting with yours, and this should be as thorough as possible within the limits of common sense. You should try to explain why these differences occurred and why these do not detract from your findings. In discussing your findings, unless you are specifically devoting your paper to medico-economics, e.g. a cost benefit study, you should not say the messages are cheaper etc. because in different countries, equipment and drugs can vary widely in price, and insurance systems etc. are extremely different internationally.
Do not make statements that are not supported by the data in your results, and also avoid claiming priority stating that this is the first study to show that. If you are trying to establish priority, you can say that this, to the best of your knowledge, is the first study to show that….
If you are citing personal communications, you should first obtain in writing permission from the person in question authorizing you to make their statements public in your paper, assuming that you keep these and all other documents, research notes, lab. notes etc. relevant to your study in a single file that can be accessed easily, should there be any investigation into the veracity of your findings. It goes without saying that no unpublished work should be cited, except in the case of when a paper has been already accepted for publication. The fact that a paper has been submitted for consideration for publication does not allow it to be cited as a reference. If a paper has been accepted for publication by the same or another journal, then please list it in the References as “in press”, and if you know the date of anticipated publication, you can write “to be published in …..”, giving the month and year. You may wish to state a new hypothesis, but if you do so, it should be emphasized that this is merely a hypothesis. Avoid meaningless concluding statements such as “we intend to conduct further studies on ABC”, because this gives no real important information to the readers, thus is a waste of both their and your time.