East and Central African Journal of Surgery Association of Surgeons of East Africa and College of Surgeons of East Central and Southern Africa ISSN: 1024-297X EISSN: 2073-9990 Vol. 13, Num. 1, 2008, pp. 4-6

East and Central African Journal of Surgery, Vol. 13, No. 1, March-April 2008, pp. 4-6

Writing  a  Convincing  and  Effective  Discussion  Section  of  a Biomedical  Research  Paper.

N. S.   Motsitsi, M.MED   (Orth.),  F.C.S.

University of  Pretoria, Kalafong  Hospital, Department  of  Orthopaedic  Surgery, Pretoria.,South  Africa. E-mail:  silas.motsitsi@up.ac.za. Fax:  +2712  373  9031.

Code Number: js08001

Writing a biomedical research paper is hard work.  The paper must conform to the orthodox:  Introduction, Method, Results, and Discussion.  This  format  brings  some  order  and  logic  to  the  research  paper. The  primary  aim  of  writing  a  biomedical   research  paper  is  to  contribute  to  scientific  knowledge.  This  contribution  must  be  explicit  and  it  must  occupy  a  prominent  position  in  the  paper. That position   is the discussion section. Regrettably,  the  discussion  is  usually  poorly  written   in  most  biomedical  journals.  The author’s contribution is often widely scattered throughout the discussion.  Sometimes it is overwhelmed by the cited literature.   This    weakens the message and the impact of the contribution.  The  approach  suggested  by  this  author  is  aimed  at  structuring  the  discussion  in  such  a  way  that  the  author’s  contribution  stands  out.  This  will  enable  the  author  to  write  a  compelling  and  effective  discussion.

Introduction

A   biomedical research paper follows a well-defined format; Introduction, Methods,  Results, and  Discussion (an IMRAD mnemonic).  The  discussion  section  is  the ‘trumpet’  of  the  paper:  it  announces  the  author’s  contribution  to  scientific  knowledge,  elaborates  on  the  significance  of  the  results,  and  gives  direction  for  future  research. Unfortunately, the discussion section is often poorly written by most authors.  The  reason  could  be  that  there  is  no  generally  accepted  format  on  how  to  structure  the  different  sections  of  the  discussion¹. Many authors have suggested guidelines for writing the discussion^2,3.  Unfortunately, these guidelines are often not followed. The  purpose  of  this  educational  article is  to  assist  young  authors  in  structuring  and  organizing  the  discussion  so  that  they  can  write  compelling  and  effective  discussion.  Well-heeled authors can use this article as a revision.

Discussion

What  is  the  main  function  or  purpose  of  the  discussion  section?  The  function  is  to  answer  the  question  posed  in  the  Introduction  section⁴.  Check the research question again before writing this section⁵. For  clarity  of  thought,  divide  the  section  into  paragraphs   as  indicated  in  table 1.  This structure helps the author to focus.  Bear  in  mind  that  a  paragraph  represents  a  unit  of  thought.  However, all paragraphs must be linked to provide coherence.  Start each paragraph with a topic sentence.  All  subsequent  sentences  in  that  paragraph  must  elaborate  on  the  topic  sentence.  Let me elaborate on each paragraph.

A.   Paragraph 1

The reader expects an answer to your research question. This paragraph represents direct declaration of your results³.  Give the answer to your main research question.  You may state  your  answer  by  saying,  for  example,  ‘The  results  show  that  spinal  tuberculosis  has  a  good  prognosis’.  This is a direct declaration of your results. Do  not  summarize  your  results  or  parrot  what  is  in  the  Results  section⁵.   Do not introduce new information. If  you  did  statistical  analysis,  give  the  interpretation  and  not  only  the  p-value. Confidence interval (C.I.) gives interpretation to the results and it must occupy a prominent position.  Elaborate on the significance of the results.What   about   secondary, interesting, or unexpected results?  These results are peripheral to your research question.  They must be mentioned, and if necessary, plausible explanation(s) offered. Do not dwell on them for they are not the focus of the research question or paper.

Table 1. Divisions of Paragraphs in a Discussion

Be careful with speculation; it has a place in the discussion provided there is some hint of support to it in your data or in cited studies⁶. Speculation is a trap for the naive, particularly in descriptive studies.   Avoiding  making  priority, because  most  of  the  time  such  claims  are  false⁷. Do not cite references in this paragraph:  your work must take the center-stage. The citations will weaken your contribution and steal the limelight. This is your opportunity to contribute and be visible; grab it.

B. Paragraph 2

This paragraph deals with the literature. It shows the reader that the author is aware of the contributions of other scientists in this area. This gives credibility to the work of the author. Generally, there are three categories of literature that the author must deal with in this paragraph¹:

1. The literature consistent with your findings.
2. Less compatible literature.
3. The literature that is in disagreement with your findings.

Discuss the literature critically.  Critique the literature equally.

C. Paragraph 3

Discussion of the study limitations is often not well done in most biomedical journals. All research work has some limitations⁸. The author must review the Method and Results sections for possible shortcomings of the study; the shortcoming may be in the selection of the study group, treatment allocation, assumptions in statistical analysis, lost to follow-up, or even the duration of follow-up. Acknowledging limitations of the study shows the reader that the author has a critical and unbiased scientific thinking. It helps future researchers to improve on the quality of the studies.  Your paper will be cited in that research: a great honour indeed.

D.    Paragraph 4

This paragraph leads to the conclusion of the research paper.  This paragraph must indicate that.  The author must summarize his results together with the relatable literature. Finally the author must suggest the type of a research that must be done to address a specific question. The author must not promise the reader that he/she (the author) will do the research because most of the time such a promise is not fulfilled.

Conclusion

The discussion section is the most difficult to write.  I hope these  guidelines  will of great help to ease this difficulty.   

References

1. Bjorn   Gustavii.  Discussion.  In:  How  to  Write  and  Illustrate  a  Scientific  Paper.  Cambridge:  Cambridge  University  Press,  New  York;  2003.  p  67-69.
2. Michael  Doherty  and  Richard  Smith.  The  case  for  structuring  the  discussion  of  scientific  papers.  BMJ  1999;  318;  1224-1225.
3. Dean  R.  Hess.  How  to  write  an  effective  discussion. Respiratory  Care 2004 ;  49(10): 1238 – 1241.
4. Mimi  Zeiger.  Discussion.  In:  Essentials  of  Writing  Biomedical  Research  Papers, 2^nd  Edition ;  McGraw-Hill;  New  York;  2000.  p  176-202.
5. The  Health  Care  Communication  Group.  The  scientific  journal  article:  approaching  the  first  draft.  In:  Writing,  Speaking,  and  Communication  Skills  for  Health  Professionals ;  Yale;  Yale  University  Press;  New  Haven  and  London;  2001. p  85-87.  
6. Vernon  Booth.   Writing  a  scientific  paper.  In:  Communicating  in  Science;  2^nd  Edition;  Cambridge;  Cambridge  University  Press;  United  Kingdom; 1993.  p  6.
7. Robert  L.  Iles.  How  to  write  a  report  of  a  clinical  study.  In:  Guidebook  to  Better  Medical  Writing;  Iles  Publications;  Revised  2003.  p  119 – 131.
8. John  P. A.  Ioannidis.  Limitations  are  not  properly  acknowledged  in  the  scientific  literature.  Journal  of  Clinical  Epidemiology 60: ( 2007);  pp  324 – 329.

© 2008 East and Central African Journal of Surgery