How to write a discussion section [622521]

How to write a discussion section?
Öner Șanlı, Selçuk Erdem, Tzevat Tefik
Department of Urology,
İstanbul Faculty of Medicine,
İstanbul University, İstanbul,
Turkey
Submitted:
12.02.2013
Accepted:
02.04.2013
Correspondence:
Öner Șanlı
Department of Urology,
İstanbul Faculty of Medicine,
İstanbul University, 34093
İstanbul, Turkey
Phone: +90 535 300 12 36
E-mail: [anonimizat]
©Copyright 2013 by Turkish
Association of Urology
Available online at
www.turkishjournalofurology.comABSTRACT
Writing manuscripts to describe study outcomes, although not easy, is the main task of an academician.
The aim of the present review is to outline the main aspects of writing the discussion section of a manu –
script. Additionally, we address various issues regarding manuscripts in general. It is advisable to work on
a manuscript regularly to avoid losing familiarity with the article. On principle, simple, clear and effective
language should be used throughout the text. In addition, a pre-peer review process is recommended to ob –
tain feedback on the manuscript. The discussion section can be written in 3 parts: an introductory paragraph,
intermediate paragraphs and a conclusion paragraph. For intermediate paragraphs, a “divide and conquer”
approach, meaning a full paragraph describing each of the study endpoints, can be used. In conclusion, aca –
demic writing is similar to other skills, and practice makes perfect.
Key words: Academic writing; discussion section; writing a manuscript
Introduction
Sharing knowledge produced during academic
life is achieved through writing manuscripts.
However writing manuscripts is a challenging
endeavour in that we physicians have a heavy
workload, and English which is common lan –
guage used for the dissemination of scientific
knowledge is not our mother tongue.
The objective of this review is to summarize
the method of writing ‘Discussion’ section
which is the most important, but probably at
the same time the most unlikable part of a
manuscript, and demonstrate the easy ways
we applied in our practice, and finally share
the frequently made relevant mistakes. During
this procedure, inevitably some issues which
concerns general concept of manuscript writ –
ing process are dealt with. Therefore in this
review we will deal with topics related to the
general aspects of manuscript writing process,
and specifically issues concerning only the
‘Discussion’ section.
A) Approaches to general aspects of manu –
script writing process :
1. What should be the strategy of sparing
time for manuscript writing be? Two differ –
ent approaches can be formulated on this issue?
One of them is to allocate at least 30 minutes a day for writing a manuscript which amounts to
3.5 hours a week. This period of time is ade –
quate for completion of a manuscript within a
few weeks which can be generally considered
as a long time interval. Fundamental advantage
of this approach is to gain a habit of making
academic researches if one complies with
the designated time schedule, and to keep the
manuscript writing motivation at persistently
high levels. Another approach concerning this
issue is to accomplish manuscript writing pro –
cess within a week. With the latter approach,
the target is rapidly attained. However longer
time periods spent in order to concentrate on
the subject matter can be boring, and lead to
loss of motivation. Daily working require –
ments unrelated to the manuscript writing
might intervene, and prolong manuscript writ –
ing process. Alienation periods can cause loss
of time because of need for recurrent literature
reviews. The most optimal approach to manu –
script writing process is daily writing strategy
where higher levels of motivation are persis –
tently maintained.
Especially before writing the manuscript, the
most important step at the start is to construct
a draft, and completion of the manuscript on a
theoretical basis. Therefore, during construc –
tion of a draft, attention distracting environ –
ment should be avoided, and this step should Turkish Journal of Urology; 39(Supplement 1): 20-4 • doi:10.5152/tud.2013.049
Education20
PARTICULARS OF AN ACADEMIC WRITING

be completed within 1-2 hours. On the other hand, man uscript
writing process should begin before the completion of the study
(even the during project stage). The justification of this approach
is to see the missing aspects of the study and the manuscript
writing methodology, and try to solve the relevant problems
before completion of the study. Generally, after completion of
the study, it is very difficult to solve the problems which might
be discerned during the writing process. Herein, at least drafts of
the ‘Introduction’, and ‘Material and Methods’ can be written,
and even tables containing numerical data can be constructed.
These tables can be written down in the ‘Results’ section.[1]
2. How should the manuscript be written? The most impor –
tant principle to be remembered on this issue is to obey the
criteria of simplicity, clarity, and effectiveness.[2] Herein, do not
forget that, the objective should be to share our findings with
the readers in an easily comprehensible format. Our approach
on this subject is to write all structured parts of the manuscript
at the same time, and start writing the manuscript while reading
the first literature. Thus newly arisen connotations, and self-
brain gyms will be promptly written down. However during this
process your outcomes should be revealed fully, and roughly the
message of the manuscript which be delivered. Thus with this
so-called ‘hunter’s approach’ the target can be achieved directly,
and rapidly. Another approach is ‘collectioner’s approach.[3]
In this approach, firstly, potential data, and literature studies
are gathered, read, and then selected ones are used. Since this
approach suits with surgical point of view, probably ‘hunter’s
approach’ serves our purposes more appropriately. However,
in parallel with academic development, our novice colleague
‘manuscripters’ can prefer ‘collectioner’s approach.’
On the other hand, we think that research team consisting
of different age groups has some advantages. Indeed young
colleagues have the enthusiasm, and energy required for the
conduction of the study, while middle-aged researchers have
the knowledge to manage the research, and manuscript writ –
ing. Experienced researchers make guiding contributions to the
manuscript. However working together in harmony requires
assignment of a chief researcher, and periodically organizing
advancement meetings. Besides, talents, skills, and experi –
ences of the researchers in different fields (ie. research methods,
contact with patients, preparation of a project,fund-raising,
statistical analysis etc.) will determine task sharing, and make
a favourable contribution to the perfection of the manuscript.
Achievement of the shared duties within a predetermined time
frame will sustain the motivation of the researchers, and prevent
wearing out of updated data.
According to our point of view, ‘Abstract’ section of the manu –
script shoul d be written after completion of the manuscript. The
reason for this is that during writing process of the main text, the
significant study outcomes might become insignificant or vice versa. However, generally, before onset of the writing process of
the manuscript, its abstract might be already presented in vari –
ous congresses. During writing process, this abstract might be a
useful guide which prevents deviation from the main objective
of the manuscript.
On the other hand references should be promptly put in place
while writing the manuscript, Sorting, and placement of the refer –
ences should not be left to the last moment. Indeed, it might be
very difficult to remember relevant references to be placed in the
‘Discussion’ section. For the placement of references use of soft –
ware programs detailed in other sections is a rational approach.
3. Which target journal should be selected? In essence, the
methodology to be followed in writing the ‘Discussion’ section
is directly related to the selection of the target journal. Indeed,
in compliance with the writing rules of the target journal, limi –
tations made on the number of words after onset of the writing
process, effects mostly the ‘Discussion’ section. Proper match –
ing of the manuscript with the appropriate journal requires
clear, and complete comprehension of the available data from
scientific point of view. Previously, similar articles might have
been published, however innovative messages, and new per –
spectives on the relevant subject will facilitate acceptance of the
article for publication. Nowadays, articles questioning available
information, rather than confirmatory ones attract attention.
However during this process, classical information should not
be questioned except for special circumstances. For example
manuscripts which lead to the conclusions as “laparoscopic
surgery is more painful than open surgery” or “laparoscopic
surgery can be performed without prior training” will not be
accepted or they will be returned by the editor of the target jour –
nal to the authors with the request of critical review. Besides the
target journal to be selected should be ready to accept articles
with similar concept. In fact editors of the journal will not
reserve the limited space in their journal for articles yielding
similar conclusions.
The title of the manuscript is as important as the structured sec –
tions* of the manuscript. The title can be the most striking or the
newest outcome among results obtained.
Before writing down the manuscript, determination of 2-3 titles
increases the motivation of the authors towards the manuscript.
During writing process of the manuscript one of these can be
selected based on the intensity of the discussion. However the
suitability of the title to the agenda of the target journal should
be investigated beforehand. For example an article bearing the
title “Use of barbed sutures in laparoscopic partial nephrectomy
shortens warm ischemia time” should not be sent to “Original
Investigations and Seminars in Urologic Oncology” Indeed the
topic of the manuscript is out of the agenda of this journal.21Șanlı et al.
How to write a discussion section?

4. Do we have to get a pre-peer review about the written
manuscript? Before submission of the manuscript to the tar –
get journal the opinions of internal, and external referees should
be taken.[1] Internal referees can be considered in 2 categories
as “General internal referees” and “expert internal referees”
General internal referees (ie. our colleagues from other medical
disciplines) are not directly concerned with your subject matter
but as mentioned above they critically review the manuscript
as for simplicity, clarity, and effectiveness of its writing style.
Expert internal reviewers have a profound knowledge about the
subject, and they can provide guidance about the writing process
of the manuscript (ie. our senior colleagues more experienced
than us). External referees are our colleagues who did not con –
tribute to data collection of our study in any way,but we can re –
quest their opinions about the subject matter of the manuscript.
Since they are unrelated both to the author(s), and subject mat –
ter of the manuscript, these referees can review our manuscript
more objectively. Before sending the manuscript to internal, and
external referees, we should contact with them, and ask them if
they have time to review our manuscript . We should also give
information about our subject matter. Otherwise pre-peer review
process can delay publication of the manuscript, and decrease
motivation of the authors. In conclusion, whoever the preferred
referee will be, these internal, and external referees should re –
spond the following questions objectively. 1) Does the manu –
script contribute to the literature?; 2) Does it persuasive? 3) Is
it suitable for the publication in the selected journal? 4) Has a
simple, clear, and effective language been used throughout the
manuscript? In line with the opinions of the referees, the manu –
script can be critically reviewed, and perfected.[1]**
Following receival of the opinions of internal, and external ref –
erees, one should concentrate priorly on indicated problems, and
their solutions. Comments coming from the reviewers should be
criticized, but a defensive attitude should not be assumed during
this evaluation process. During this “incubation” period where
the comments of the internal, and external referees are awaited,
literature should be reviewed once more. Indeed during this time
interval a new article which you should consider in the ‘Discus –
sion’ section can be cited in the literature.
5. What are the common mistakes made related to the writ –
ing process of a manuscript? Probably the most important
mistakes made related to the writing process of a manuscript
include lack of a clear message of the manuscript , inclusion
of more than one main idea in the same text or provision of
numerous unrelated results at the same time so as to reinforce
the assertions of the manuscript. This approach can be termed
roughly as “loss of the focus of the study” In conclusion, the
author(s) should ask themselves the following question at every
stage of the writing process:. “What is the objective of the
study? If you always get clear-cut answers whenever you ask this question, then the study is proceeding towards the right
direction. Besides application of a template which contains the
intended clear-cut messages to be followed will contribute to the
communication of net messages.
One of the important mistakes is refraining from critical review
of the manuscript as a whole after completion of the writing
process. Therefore, the authors should go over the manuscript
for at least three times after finalization of the manuscript based
on joint decision. The first control should concentrate on the
evaluation of the appropriateness of the logic of the manu –
script, and its organization, and whether desired messages have
been delivered or not. Secondly, syutax, and grammar of the
manuscript should be controlled. It is appropriate to review the
manuscript for the third time 1 or 2 weeks after completion of
its writing process. Thus, evaluation of the “cooled” manuscript
will be made from a more objective perspective, and assessment
process of its integrity will be facilitated.
Other erroneous issues consist of superfluousness of the
manuscript with unnecessary repetitions, undue, and recurrent
references to the problems adressed in the manuscript or their
solution methods, overcriticizing or overpraising other studies,
and use of a pompous literary language overlooking the main
objective of sharing information.[4]
B) Approaches to the writing process of the ‘Discussion’
section:
1. How should the main points of ‘Discussion’ section be
constructed? Generally the length of the ‘Discussion ‘ section
should not exceed the sum of other sections (ıntroduction, mate –
rial and methods, and results), and it should be completed within
6-7 paragraphs.. Each paragraph should not contain more than
200 words, and hence words should be counted repeteadly. The
‘Discussion’ section can be generally divided into 3 separate
paragraphs as. 1) Introductory paragraph, 2) Intermediate para –
graphs, 3) Concluding paragraph.
The introductory paragraph contains the main idea of perform –
ing the study in question. Without repeating ‘Introduction’
section of the manuscript, the problem to be addressed, and
its updateness are analysed. The introductory paragraph starts
with an undebatable sentence, and proceeds with a part address –
ing the following questions as 1) On what issue we have to
concentrate, discuss or elaborate? 2) What solutions can be
recommended to solve this problem? 3) What will be the new,
different, and innovative issue? 4) How will our study contrib –
ute to the solution of this problem An introductory paragraph
in this format is helpful to accomodate reader to the rest of the
Discussion section. However summarizing the basic findings
of the experimental studies in the first paragraph is generally
recommended by the editors of the journal.[5]22Turkish Journal of Urology 2013; 39(Supplement 1): 20-4
doi:10.5152/tud.2013.049

In the last paragraph of the Discussion section “strong points”
of the study should be mentioned using “constrained” , and
“not too strongly assertive” statements. Indicating limitations
of the study will reflect objectivity of the authors, and provide
answers to the questions which will be directed by the review –
ers of the journal. On the other hand in the last paragraph,
future directions or potential clinical applications may be
emphasized.
2. How should the intermediate paragraphs of the Discussion
section be formulated? The reader passes through a test of bore –
dom while reading paragraphs of the Discussion section apart from
the introductory, and the last paragraphs. Herein your findings
rather than those of the other researchers are discussed. The previ –
ous studies can be an explanation or reinforcement of your find –
ings. Each paragraph should contain opinions in favour or against
the topic discussed, critical evaluations, and learning points.
Our management approach for intermediate paragraphs is
“divide and conquer” tactics. Accordingly, the findings of the
study are determined in order of their importance, and a para –
graph is constructed for each finding (Figure 1). Each paragraph
begins with an “indisputable” introductory sentence about
the topic to be discussed. This sentence basically can be the
answer to the question “What have we found?” Then a sentence
associated with the subject matter to be discussed is written.
Subsequently, in the light of the current literature this finding
is discussed, new ideas on this subject are revealed, and the
paragraph ends with a concluding remark.
In this paragraph, main topic should be emphasized without
going into much detail. Its place, and importance among other
studies should be indicated. However during this procedure
studies should be presented in a logical sequence (ie. from past
to present, from a few to many cases), and aspects of the study
contradictory to other studies should be underlined. Results
without any supportive evidence or equivocal results should not
be written. Besides numerical values presented in the Results
section should not be repeated unless required.
Besides, asking the following questions, and searching their an –
swers in the same paragraph will facilitate writing process of the
paragraph.[1] 1) Can the discussed result be false or inadequate?
2) Why is it false? (inadequate blinding, protocol contamination,
lost to follow-up, lower statistical power of the study etc.), 3)
What meaning does this outcome convey?
3. What are the common mistakes made in writing the Dis –
cussion section?: Probably the most important mistake made
while writing the Discussion section is the need for mentioning
all literature references. One point to remember is that we are
not writing a review article, and only the results related to this paragraph should be discussed. Meanwhile, each word of the
paragraphs should be counted, and placed carefully. Each word
whose removal will not change the meaning should be taken out
from the text.” Writing a saga with “word salads”*** is one of the
reasons for prompt rejection. Indeed, if the reviewer thinks that
it is difficult to correct the Discussion section, he/she use her/
his vote in the direction of rejection to save time (Uniform re –
quirements for manuscripts: International Comittee of Medical
Journal Editors [http://www.icmje.org/urm_full.pdf])
The other important mistake is to give too much references , and
irrelevancy between the references, and the section with these cit –
ed references.[3] While referring these studies, (excl. introductory
sentences linking indisputable sentences or paragraphs) original
articles should be cited. Abstracts should not be referred, and re –
view articles should not be cited unless required very much.
4. What points should be paid attention about writing rules,
and grammar? As is the case with the whole article, text of the
Discussion section should be written with a simple language,
as if we are talking with our colleague.[2] Each sentence should
indicate a single point, and it should not exceed 25-30 words.
The priorly mentioned information which linked the previous
sentence should be placed at the beginning of the sentence,
while the new information should be located at the end of the
sentence. During construction of the sentences, avoid unneces –
sary words, and active voice rather than passive voice should be
used.**** Since conventionally passive voice is used in the sci –
entific manuscripts written in the Turkish language, the above
statement contradicts our writing habits. However, one should
not refrain from beginning the sentences with the word “we”.
Indeed, editors of the journal recommend use of active voice so
as to increase the intelligibility of the manuscript.23Șanlı et al.
How to write a discussion section?
Figure 1. Divide and Conquer tactics1. Result 1. paragraph of the
Discussion section Introduction-Why has
this study performed?*
Introduction-link
Literature +Opinions
Outcome
Introduction-link
Literature + Opinions
Outcome
Introduction-link
Literature + Opinions
Outcome
Introduction-link
Literature + Opinions
Outcome
Introduction-link
Literature + Opinions
Outcome
Limitations-plans
for future
* İf no mentioned in the Introduction section 2. paragraph of the
Discussion section
3. paragraph of the
Discussion section
4. paragraph of the
Discussion section
5. paragraph of the
Discussion section2. Result
3. ResultResults
4. Result
5. Result

In conclusion, the major point to remember is that the manu –
script should be written complying with principles of simplic –
ity, clarity, and effectiveness. In the light of these principles,
as is the case in our daily practice, all components of the
manuscript (IMRAD) can be written concurrently. In the
‘Discussion’ section ‘divide and conquer’ tactics remarkably
facilitates writing process of the discussion. On the other
hand, relevant or irrelevant feedbacks received from our col –
leagues can contribute to the perfection of the manuscript.
Do not forget that none of the manuscripts is perfect, and one
should not refrain from writing because of language prob –
lems, and related lack of experience. References
1. Welch HG. Preparing manuscripts for submission to medical jour –
nals: The paper trail. Eff Clin Prac 1999;2:131-7.
2. Tompson A. How to write an English medical manuscript that will
be published and have impact. Surg Today 2006;36:404-9.
3. Setiati S, Harimurti K. Writing for scientific medical manuscript: A
Guide for preparing manuscript submitted to biomedical journals.
Acta Med Indones 2007;30:50-5.
4. Murray R. Finding a topic and developing an argument. In Murray
R editor Writing for Academical Journals. Open University Press
2005.p.67-98.
5. International Committee of Medical Journal Editors. Uniform Re –
quirements for Manuscripts Submitted to Biomedical Journals: Wri –
ting and Editing for Biomedical Publication. Updated April 2010.
Instead of structured sections of a manuscript (IMRAD): Introduction, Material and Methods, Results, and Discussion
Instead of in the Istanbul University Faculty of Medicine posters to be submitted in congresses are time to time discussed in Wednesday meetings, and opinions of
the internal referees are obtained about the weak, and strong points of the study
Instead of a writing style which uses words or sentences with a weak logical meaning that do not lead the reader to any conclusion
Instead of “white color”; “proven”; nstead of “history”; “to”. should be used instead of “white in color”, “definitely proven” , “past history”, and “in order to” ,
respectively (ref. 2)
Instead of “No instances of either postoperative death or major complications occurred during the early post-operative period” use “There were no deaths or major
complications occurred during the early post-operative period.
Instead of “Measurements were performed to evaluate the levels of CEA in the serum” use “We measured serum CEA levels”24Turkish Journal of Urology 2013; 39(Supplement 1): 20-4
doi:10.5152/tud.2013.049