| Abstract|| |
A well-known unwritten law in institutions of higher learning is that of "Publish or perish". The duties of a University teacher, in order of priority are teaching, research and service. Reasons for writing clinical research papers are to get promoted, to get research grants and to make known, one's findings in order to improve patients' care. Writing papers is also a means of delivering continuous education, therefore publication is essential for any one pursuing an academic career. Research papers can be in the form of case reports, retrospective studies, prospective studies and laboratory or animal research. Two popular formats of writing papers are: The Vancouver Style and the Harvard System.
|How to cite this article:|
Ajao OG. Writing a clinical research paper. Saudi J Gastroenterol 1999;5:45-9
A review article on "Writing a clinical research paper" is justified for three reasons: firstly, to renew our knowledge of the subject, to correct a previously unknown unintentional errors in writing clinical research papers and to serve as guidelines for young researchers intending to write scientific articles.
In institutions of higher learning there is some justification for the phrase unwritten 'law',: "Publish or perish", bearing in mind that the duties of a University teacher, in order of priority, are teaching, research and service  . Some think research comes before teaching, but there is a consensus that service comes last, except for emergencies. Also academic scientists need to publish in peer-reviewed journals to get research grants, to get promoted and to increase their knowledge. Publishing therefore is also a means of continuous education. An educator is not only expected to teach students, but also to try improve on the established methods of patients' care and convey his findings to his colleagues. This can only be done through research and publications. Scientific misconduct is commonly seen as a case of falsification of data, however failure to publish an adequate account of a well-designed clinical trial for example is also a form of scientific misconduct because what can improve patients' care and treatment  is being deliberately suppressed. Many clinicians working in general rather than University Teaching Hospitals are more interested only in giving service and this should not imply that such clinicians are inferior to their counterparts in the university. Far from it! It is just a question of preference and personal interest.
This paper attempts to recount certain aspects of writing a clinical research paper.
| Historical note|| |
It is not easy to identify precisely the first clinical research paper. Exchange of letters between scientists was known after the Renaissance. Henry Oldenberg who edited the Philosophical Transactions of the Royal Society of London as far back as 1665 is generally credited with starting scientific journalism. By the eighteenth century medical journals were relatively well established, but were linked to medical societies and institutions , .
| Types of research paper|| |
Clinical research papers can be grouped arbitrarily into four major categories: case reports, retrospective studies, prospective studies and laboratory or animal studies.
Case reports, do not necessarily have to be low-keyed papers. It depends on the case and the presentation. Usually the format of writing this is a bit different from others , . This has a SICDC structure, that is, Summary, Introduction, Case report(s), Discussion and Conclusion. When there are more than one case being presented, each can be presented separately, one after the other, highlighting the essential points in each case. Because some journals do not publish case reports, one, has to be selective in the choice of journals. Some journals also do not require `Summary' for case reports, while others consider case reports as "inferior" publications. However, if there are no case reports we probably would not have known about Crohn's disease or Hirschprung's disease etc.
Retrospective studies are relatively common in clinical practice, especially among young consultants, who are just starting their academic life. Usually, this type of study relies on the records of previous cases managed by different individuals. Therefore the results tend to contain some flaws. Even then, meaningful deductions can still be made from such studies. In fact some very good papers have originated from retrospective studies. Many audit-related type of publications are best suited for this. One major drawback of retrospective studies is that proper randomization is usually not possible and the degree of accuracy of such studies depends on the Medical Records Department and how meticulous the clinical data were documented by the treating physicians.
Prospective studies tend to have more validity than the retrospective because the short comings of the latter are largely eliminated. However, it requires a meticulous, orderly planning before the study is embarked upon. There has to be a clear aim or objective of the study. It is advisable to gather more data than needed, than to find out during the writeup of the paper that some necessary data were not included initially. Unlike in retrospective study, prospective studies make randomization and clinical trials possible. The clinical trials can involve a group of patients randomly selected, or using the same patients as their own controls. For example, in a clinical trial on methylprednisolone sodium succinate (Solu-Medrol) in the treatment of typhoid perforation, envelopes labelled "Yes" for the steroids usage and "No" for without-steroids can be sealed  . To avoid bias, the nurse or anyone not directly involved in the study can be made to select any envelope blindly for any patient admitted with typhoid perforation. This will determine the modality of treatment. Of course, this type of study requires approval by the Ethical Committee of the institution or hospital, which is responsible for preventing unethical practice in clinical research. A patient can also serve as his or her own control. For example, in testing vagotomy for the relief of pain in inoperable upper-gastrointestinal carcinoma, pain experienced by the patient can be graded before the procedure, and then after the performance of the procedure  . The main drawback in such a study is that it is difficult to eliminate completely, the suggestive effect of the treatment on the patient. Many academic institutions regard laboratory or animal studies as more important than the three previously mentioned above. There is no doubt that majority of laboratory and animal studies produced break-throughs in clinical medicine. Therefore, such studies should be encouraged and supported.
Ignorance of previous studies can affect the design and execution of trials  . In case of clinical trials, for example, searching MEDLINE alone may not be adequate. Other sources such as Oxford Database of Perinatal Trials and "Current Contents" : Clinical Medicine should be sought. Some institutions don't give much recognition to "Letters to the Editor" and "Brief or Short Reports" and therefore such publications are usually not considered for promotion. However, the importance of an article lies in the content and not the length! Some important scientific information has emerged from "Letters to the Editor" and "Brief or Short Reports". The article by J.D. Watson and F.H. Crick on the structure on DNA which appeared in "Nature" in 1953, occupied just over half a printed page , . Yet this is regarded as a break-through article.
Review articles are usually presented by an expert or a senior professional colleague in a particular field. Essentially such articles carry the views of many experts in a particular field. In general review articles, are usually invited by the editor of the journal or are submitted for publication after consultation between author and the editor on the topic to be reviewed. However, the practice nowadays is that if a review article is good enough and addresses an important scientific subject, most editors will accept it for publication with or without prior consultation with author.
| Study design in medical research|| |
There are certain terminologies that any medical researcher should be familiar with. (1) Randomized control trial refers to comparing at least 2 groups: experimental and a control group for comparison  . (2) Crossover trial refers to a situation in which 2 or more treatments are given one after the other to a group, and each serves as his own control for comparison of treatments. (3) Cohort study refers to a situation in which a group is followed for a period of time to assess the outcome of a disease or a form of therapy. This can be prospective or retrospective. (4) Case-control study is a retrospective study of the risk factors for a disease in which the cases are compared with the controls. (5) Cross-sectional study refers to the study of a group at a given point or period of time to determine if variables are associated with each other. (6) Case series describes a group with a similar medical condition or similar medical procedure. (7) Meta-analysis refers to a study that pools the results of 2 or more studies of a particular disease or problem. (8) Cost-effectiveness analysis compares the financial benefits or otherwise of a treatment or procedure. In all these the studies could be laboratory experiments, epidemiologic studies or controlled clinical trials.
| Choice of journal|| |
Scientific journals will accept any manuscript for publication if they fulfil the requirements of the journal. Some journals are very selective, and highly specialised. For example, a paper on "Chest Trauma" will be more appropriat for a journal of "Thoracic and Cardiovascular system" or a journal of "Trauma" than a journal of general surgery. In the initial submission of a paper, journal prestige, readership composition and relevance to the topic should all be born in Mind , . Top journals are defined by their impact factor, that is a figure related to the total number of citations they achieve, divided by the total number of citable articles they publish  . Besides, for articles submitted for publication after one or two rejections, the likelihood of acceptance becomes paramount in journal selection, by the author  , fueled on by the "publish or perish" syndrome  .
Sometimes journals carry supplements. Supplements are frequently underwritten by pharmaceutical manufacturers and are less likely to undergo the peer review process; yet, supplement articles are listed in databases such as MEDLINE. It has been shown that manuscripts published in journal supplements are generally inferior to those published- in the parent journal  .
| Format of manuscript|| |
With the exception of case reports and review articles, most papers have the SIMRDC structure  , that is Summary, Introduction, Materials and Methods, Result, Discussion and Conclusion. All journals dictate specific methods of presenting the references and these fall in two categories: The Vancouver Style , and the Harvard System  . In January 1978 a group of editors from major biomedical journals published met in Vancouver, British Columbia, and decided on uniform technical requirements for manuscripts, submitted to their journals , . This is the origin of the Vancouver Style of citing references. In the body of the paper, references are indicated by numbers and references listed numerically in the order in which they are cited in the text. Most journals follow the Vancouver style.
The Harvard System  is different. In the text the names of the authors and the year of publication are given. In the reference section, the authors are listed in alphabetical order. In both systems names of journals cited are abbreviated in accordance with the style in Index Medicus. "Unpublished" references, "Personal Communications", and "In press" are not acceptable as references in some journals.
| Causes of paper rejection|| |
There are many reasons for rejection of a manuscript sent for publication. One is (1) lack of clear objectives. The main aim of a paper must be clearly stated in the "Introduction". The paper should focus on attaining the stated objectives. Another cause of paper rejection is (2) error in calculations or statistical analysis. Usually when a reviewer detects such errors, the credibility of the whole paper is immediately called into question. (3) Failure of adherence to the accepted structure of writing a research paper also mitigates against acceptance. It is not the duty of a reviewer or editor to teach the author(s) how to write a research paper. It is easier and quicker for him to just write "rejected". (4) Typographical and grammatical errors may not lead to outright rejection of a scientific paper, but they don't enhance its acceptance either. The tendency to include "Findings" or "Results" or "Discussion" under "Methods and Materials" should be avoided. Each heading should address what it is supposed to do  . Discussion should be restricted to the study being presented. Any statement outside this and without been backed by a reference is usually regarded as speculation. Authors should not only check the list of references, examine each reference carefully and not just copy supposedly relevant references wtihout reading them. Most common citation errors are incorrect spellings of authors' names and partial omissions of titles of articles  . This is one indication that the reference in question was not read, but only copied from another article.
Most journals require an "Abstract" to go along with the paper. This is usually a summary of the key points in the paper and usually contains no more than 250 words. "Letters to the editor" or "Brief Communication" usually don't require an abstract. "Keywords" which often follow the "Abstract" can be regarded as the most important words of the theme of the paper.
Kassirer and Campion  gave a list of reasons for rejection of manuscript.
| How to deal with a rejected paper|| |
When a paper is rejected, in some cases, reasons for the rejection are given. In this situation, it may not be too difficult to rectify the flaws and re-write the paper. When it is outright rejection, it is better to send the revised paper to a different journal. However, in cases of "provisional acceptance subject to certain changes", the revised paper should be sent to the same journal. When a paper is rejected without any reason, it is better for the author to put the paper away for a couple of weeks. When the author goes over the paper again, certain flaws in the paper which were not obvious initially tend to become obvious.
| Are all rejections justified ?|| |
When faced with the harsh critical remarks of a petulant, disdainful and arrogant reviewer, a novice who presents a poorly written article can be totally demoralised. Yet not all articles rejected are justified. It has been shown that about 62% of rejected manuscripts in a major journal were also turned down by other good indexed medical journals solicited  , which means that the remaining 38% of the articles initially rejected were publishable. Edwin G. Erdos' paper on angiotesin I - converting enzyme was initially rejected before it was finally published in 1975 , . Glen Davis case report on Parkinson's disease was rejected twice by two top journals before it was finally accepted by a third journal in its first year of publication  . It is now known that this paper has given enormous contribution to our understanding of Parkinson's disease. Hans Krebs paper on what is now known as the Krebs (or tricarboxylic acid) cycle for which he has been awarded the Nobel prize, in 1953, was initially rejected by a top journal before being published by the Dutch Enzymologia .
Expensive and time consuming research projects are of no use to the scientific community if no one knows about their results. Therefore, the publication of their results should be given as much effort as the research work itself. To those whose papers "place the author at the mercy of malignant jealousy of anonymous rival" , , I would like to quote Leland Clark 13 ' 22) who invented Platinum electrode for the oximeter: "Young people, and young at heart, take heed. It is nice to publish articles in prestigious journals ... but if they won't accept them, do not be discouraged. Publish them elsewhere. Mendel published in a little monastery newsletter".
| References|| |
|1.||Ajao OG. Peer Review in Medicine and Biomedical Sciences. Saudi J Gastroenterol. 1997;3:107-12. |
|2.||Chalmers I. Underreporting research is scientific miscondut. JAMA. 1990;263:1405-8. |
|3.||Breathnack CS. The golden heresy of truth. J Irish Coll of Physicians and Surgeons. 1995;24:114-21. |
|4.||Boorstin DJ. The discoverers. London. Dent. 1984;386-394. |
|5.||International Committee of Medical Journal Editors. Uniform requirements for manuscript submitted to biomedical journals N Engl J Med 1991;324:424-8. |
|6.||International Committee of Medical Journal Editors. Uniform requirements for manuscript submitted to biomedical journals. BMJ 1991;302:338-41. [PUBMED] [FULLTEXT]|
|7.||Ajao OG, Ajao OA, Tinuke Johnson. Methylprednisolone sodium succinate (Solu-Medrol) in the treatment of typhoid perforation (a preliminary report). Transactions. 1984;78:573-6. |
|8.||Ajao OG. Vagotomy for relief of pain in some upper gastrointestinal neoplasms. J Natl Med Assoc. 1977;69:655-8. |
|9.||Chalmers TC, Frank CS, Reitman D. Minimizing the three stages of publication bias. JAMA. 1990;263:1392-5. |
|10.||Maddox J. Is the literature about to be readable ? Nature 1988;335:665. |
|11.||Lore W. Peer review and refereeing in science. East Afr Med J. 1995;72:335-7. |
|12.||Hill AB. The reasons for writing. Br Med J. 1965;ii:870-71. |
|13.||Frank E. Authors' criteria for selecting journals. JAMA. 1994;272:163-4. |
|14.||Rochon PA, Gurwitz JH, Cheung CM, Hayes JA, Chalmers TC. Evaluating the quality of articles published in journal supplements compared with the quality of those published in the parent journal. JAMA. 1994;272:108-13. |
|15.||No author. Instructions to authors. Br J Radiol. 1993;66:98-9. |
|16.||Lee ST. Annals Editorial: Reviewing process improving the quality of publication (Editorial). Ann Acad Med Singapore 1992;21:301-3. [PUBMED] |
|17.||Evans SJ, Nadjari HI, Burchell SA. Quotation and Reference Accuracy in Surgical Journals. A continuing peer review problem. JAMA. 1990;263:1353-7. |
|18.||Kassirer JP, Campion EW. Peer review. Crude and understudied, but indispensable. JAMA. 1994;272:96-7. |
|19.||Abby M, Massey MD, Galandiuk S, Polk HC. Peer reviews is an effective screening process to evaluate medical manuscripts. JAMA. 1994;272:105-7. |
|20.||Erdos GE. The Angiotensin I - converting enzyme. Laboratory Investigation. 1987;56:345-8. |
|21.||Altman DG. The scandal of poor medical research. BMJ. 1994;308:283-4. |
|22.||Severinghaus JW, AHA! In Astrup.O.Severinghaus J.W. The history of blood gases, acids and bases. Copenhagen. Munksgaard. 1996:282-6. |
Oluwole Gbolagunte Ajao
College of Medicine, King Khalid University - Abha, P.O.Box 641, Abha
Source of Support: None, Conflict of Interest: None