Author of Article: DR. Gajanan Bhalerao (PT). PhD Scholar, MPTH Neuro, Certified Adult NDT therapist, HOD Physiotherapy & Rehabilitation Dept in Sancheti Hospital Shivajinagar Pune. Associate Professor & HOD PT in Neuro Rehabilitation Dept at Sancheti Institute College of Physiotherapy, Shivajinagar Pune
We all want to do good research and publish a paper our paper in scientific journal. But very few of us know how to write a research article. Writing of research paper is called as Manuscript writing. M.U.H.S. Nashik had arranged a workshop in Manuscript writing they and invited editors from Canada and US for training. Fortunately i got opportunity to attend the workshop. I am sharing with you what ever they trained us and given the guidelines for manuscript.
WRITING A BIOMEDICAL RESEARCH PAPER
Writing a scientific paper is a creative activity: designs for no two papers are the exactly the same. Nevertheless, there is a standard structure to biomedical scientific papers and many common elements, which are described below. Scientific papers written for other disciplines (e.g., the social sciences) and non-research papers on medical topics have somewhat different structures.
Disclaimer: no paper will follow all of the “rules” listed. The art of writing is to know how and when the research question, study design, data, journal, or its readers demand deviation from standard approaches for the sake of clarity, effective communication, and flow.
- Anatomy and Physiology of the Paper
Medical research papers have four main sections; an Introduction, Methods, Results, and Discussion. Each section should be separated by a prominent header (e.g., centered, in capitals and bold).
- Introduction is typically short: three to five paragraphs. Its purpose is to introduce the topic of the study, provide just enough background so the reader understands what the study is about, why it is important, how it fits into the literature, why it uses particular methods, and what its aims and/or purpose are. Get to the study aims quickly and directly or risk losing the reader’s attention.
- First paragraph introduces the general topic of the paper and its importance. Don’t go on and on about the general topic or you will lead the reader astray about the actual focus of the paper and bore him/her too. For example, don’t spend too much time addressing the general importance of pediatric immunizations, if the specific topic of the paper is physician compliance with national pediatric immunization guidelines. If possible, get to the specific topic of the paper in the first paragraph.
- The second through fourth paragraphs should (1) introduce or expand on the specific topic of the paper, (2) provide any key definitions the reader will need, (3) describe issues pertinent to the specific topic and/or other topics that will become important in how/why the study is designed the way it is, (4) review the important relevant lesson from the literature, (5) include references to substantiate the points made, and (6) identify a “problem” in the field. This last point is key, as it provides the rationale for the study. This crucial sentence often contains the words “however. . . .” or “unfortunately. . . .”
In reviewing the
- Last paragraph gives the study’s aims or goals, stated in one or a few sentences. These are the most important sentences of the paper and should be written first, but often must be tweaked later on. Typically, the goal statement starts, “The goals of this study are to . . . .” The rest of the paper can be assessed for relevance against these sentences: any parts which are not directly related likely should be dropped. If there are hypotheses, they are often presented immediately following the goal statement (alternatively, they can be specified in the Methods section). Sometimes this paragraph also gives a prelude to the study design, to prepare/orient the reader.
- The key to a good Introduction section is to be brief, crisp, to the point, and linear in reasoning. Save lengthy expositions for the Discussion section.
- The Methods should communicate to the reader how you went about addressing your research question. It is usually written in past tense. It should answer the questions “who” (study population), “what” (what you did), “how” (how did you accomplish it), “why” (explain why key methodological decisions were made the way they were), and sometimes “when” (to explain important time sequences).
- The key is to provide enough details for the reader to understand all important issues in the study design without getting bogged down in less important details. As a researcher, you spend most of your time with these details but don’t think that the reader needs or wants to know about them or that the editor will give you the space to address them. When you write your first research paper, you realize how much of the execution of a study is not reported in journals and, therefore, how much the integrity of medical science rests on the honesty of its researchers and trust that they know what they are doing in the many unreported methodological decisions they make. If more details will be important to some readers, consider including them as an appendix or providing a statement where this detailed information can be found, which is typically at a web site.
The goal in writing the Methods section is to provide readers with enough details so that they can feel confident that they understand generally how the study was performed, convinced of the integrity of the study’s design and execution, and replicate the major steps if so inclined.
- The typical components of the Methods are described below. It is often helpful to the reader to identify these components with subheadings. For randomized controlled trials and some other specific types of studies (e.g., assessments of diagnostic and therapeutic tools), there are international consensus documents (e.g., http://www.consort-statement.org/) listing the elements of study methods that should be included and how to report them. Components of the Methods:
- Introductory sentence or brief paragraph outlining the study’s overall methods (e.g., chart audit or mail survey) and study design (e.g., randomized control trial or cohort study).
- Describe who the study’s subjects were, how they were identified and recruited, and inclusion/exclusion criteria. Typically, you should mention how informed consent was handled and report that your IRB reviewed and approved or exempted your study.
Either in this section or in the beginning of the Results section, report participation/response, refusals, and ineligible rates. Account for all who dropped out from the study. The reader should be able to follow the numbers to understand how the subjects for whom data are complete compare to all individuals who were eligible for the study (“sampling frame”) and all individuals who were asked to participate.
- Data and Data collection. Describe how the study’s data were collected and what the data were. Describe the methods of data collection:
- If a questionnaire was used, describe how it was designed, what was learned from any pilot tests, provide test performance parameters for new measures or scales, and describe how mailings were conducted.
- For clinical studies, describe the mechanics of subject recruitment, specimen gathering and handling, equipment used, and how assays were conducted.
- For qualitative studies, describe the particular qualitative paradigm used, the interactions between study personnel and subjects, and any interview guides.
In quantitative studies, describe the outcome variables in some detail. Be certain they can be fully understood by the reader, who will be more interested in these variables than all others.
List any “secondary” data sources, e.g., appended billing data or data from national sources (e.g., community data from the US Census, hospital data from the American Hospital Association’s annual survey) and reference the sources.
- Data analysis. Describe how the data were analyzed. Describe these analytic methods in the order that their results will be reported in the Results section to follow. For quantitative studies, describe the statistical methods used to answer the various research questions. Mention how/why control variables were chosen as they were. For qualitative studies, describe the process of interpreting the data.
The analysis section typically ends with four pieces of information: (1) the statistical software package and version used when newer or complex statistical methods and study designs are employed and which may be handled differently by the various available statistical programs; (2) the level of statistical significance used and, if necessary, why; (3) a statement that the study received approval or exemption from a human subjects committee, and the name of that committee (alternatively, this statement is placed in the Subjects subsection); and (4) any competing interests or indicate that there are none.
- Here you report all of the study’s findings. The length of this section varies depending on the amount of data to be presented. Despite the temptation, save interpretation to the Discussion section.
- Typically the Results section starts by describing the study population, either in text or a table. The reader needs a clear understanding of who made it into your study. Typically one provides demographics (age, gender, race, SES indicators) and other general descriptors.
- Next, provide descriptive statistics characterizing features relevant to the study’s topic (e.g., in a cardiovascular study, the average weight and blood pressure of subjects) and the outcome measures (e.g., report the number of heart attacks or cardiovascular deaths).
- Finally, present the findings of analyses as they address each of study questions. These are generally between-group comparisons with statistical tests included (e.g., comparison of heart attack rates by age or by weight). Typically, one provides bi-variate associations first (e.g., t-tests, chi-squares), followed by multivariate (e.g., regression analysis) and subgroup analyses.
- Decide which of your findings are better presented in figures or tables rather than the text. Typically, data should be presented where they are easiest to understand and take up less space. Data presented in tables and figures should not be repeated fully in the text; only the key findings should be highlighted.
Typically a medical research paper will have five or fewer tables and figures total. Any more challenges the reader and amount of space the journal will allow. Keep the tables and figures as simple and relatively easy to understand. Tables and figures should be able to stand alone, that is the title, headings, and footnotes should allow the reader to understand them without needing to refer back to the text.
- When the data are available, add other “side analysis” which test possible interpretations of your main findings and answers various “but what about. . . .” questions that readers are quick to ask. This typically can be done in one short paragraph for each “side analysis” and will strengthen the paper and clarify the interpretation of the findings. These analyses should be clearly identified as post hoc.
- Add headings to subsections within the Results section if they will help orient the reader.
- You have most freedom with crafting this section. Length varies.
- The Discussion typically starts with a restatement of the study’s goals and the most important findings in summary. The following paragraphs then begin to interpret the findings, discuss their implications, and describe how they relate to the findings of previous studies. This is your chance to be more expansive, but don’t get too far from your findings. A common error is to simply restate the findings already given in the Results section. Instead, the Discussion is where findings are interpreted.
- Limitations are typically reported near the end of the Discussion section before the “Conclusions and Implications, and often labeled with a subheading. Alternatively, limitations can be presented at the end of the Results section or early in the Discussion. Find the place where it distracts least from the paper’s flow. Ending the paper with limitations weakens the Conclusion subsection and dilutes the study’s message.
The reader will judge your honesty and therefore, your paper’s integrity by what you “admit to” as a limitation. Be forthright about the study’s weaknesses; however, only include truly significant limitations and do not provide a laundry list of every possible study weakness.
- The paper should conclude with one to a few paragraphs stating the study’s conclusions and implications, often with a subheading “Conclusions” or “Conclusions and Implications”. Include implications for all important relevant parties, e.g., physicians, policy makers, and educators. Don’t remain insular in your thinking here. Physician-researchers too often think that their readers are only other physicians. Anticipate what readers in other fields should learn from your study and communicate this.
You can address the need for further research in this subsection. However, don’t conclude the paper with a statement that simply indicates more research is needed, as this tells the most readers—those who are not researchers—that they just wasted their time reading your study, as it was unable to teach them anything other than to wait for the work of others.
Do’s, Don’ts and Pearls in Writing a Research Paper
- Never forget that the object of writing is to communicate. Write plainly. Avoid words which are needlessly long, obscure, or jargon. While they may seem like they make you and the paper more erudite, they probably make it less accessible and therefore less influential with the reader.
- Decide which journal (or a couple of possible journals) to which you will submit your paper before you begin writing. Get a copy of that journal’s “Instructions for Authors” before hand to guide your choices in constructing the paper. Ask an experienced colleague to help think through the choice of journals.
- Writing a paper is like a putting together a jigsaw puzzle: there are lots of pieces that must be moved around and tried out here and there until the right order becomes clear. For example, often a point made in the Introduction section fits the flow of the paper better if moved to the Methods or Discussion. This is part of the “art” of writing.
- Footnotes are rarely used in biomedical journals; readers are not accustomed to them and many journals don’t allow them. Instead, fit the information into the text. Bracket the information if necessary.
In the paper, directly address issues or questions that will arise for many readers: don’t leave those “but what about. . . .” questions unanswered or left hanging until answered several pages later. Unanswered questions will keep readers from being truly convinced of your study’s integrity, the meaning of its findings and its implications. If possible, deal with these issues earlier in the paper rather than later to remove readers’ doubts so they will accept the remainder of the paper. Rely on colleague-reviewers to identify these questions that must be answered promptly for readers.
- It is usually better to understate than overstate your point. Understating generally gives a more learned tone to your paper. Avoid hyperbole, dramatic statements and tenuous leaps of faith, as they come over as obvious for what they are, which is unconvincing and often silly.
- Reviewing the relevant literature succinctly challenges many writers. In a medical research paper, one usually briefly mentions the lessons or issues from previous studies, then adds a few references to support each point. This allows the text to flow more smoothly and quickly than naming and describing specific studies and the particulars of their execution. Provide details only for specific studies that are particularly influential in the field or if they are particularly important to understanding how your study fits into the literature. Be sure to repeat your literature search as you write your paper: the search you did when first designing your study is probably out of date.
- Ask three to six or more colleagues to review your paper. Learn who among your colleagues makes a good editor, and who provides which perspectives and skills as reviewers then choose individuals with complementary skills. Learn to value lots of red marks on your papers from your colleague-reviewers: it shows they care about you and your paper. If you don’t get this constructive feedback from them, then you are likely to get the feedback they didn’t provide from the journal’s reviewers, perhaps with a rejection decision from the editor. Every weakness identified by a colleague-reviewer is another chance to improve the paper. Polish your paper somewhat before asking colleagues for a review: don’t force them to wade through a rough, half-finished draft unless you have a specific, basic question on which you are looking for feedback early, and if so, tell them the specific area of feedback you need.
- When getting feedback from colleague-reviewers, you needn’t take every suggestion made by each reviewer. Use reviewers to point out areas that need more attention, but don’t feel obligated to follow the specific advice the reviewers have provided. For example, the appropriate response to a recommendation to drop mention of analyses of a certain subgroup may be to do a better job in the text of describing this subgroup and justifying its importance. If a given issue is pointed out by more than one reviewer, it very likely means that many readers will question this point and it deserves a response.
- Don’t hesitate to delete whole paragraphs or remove major themes if it strengthens the paper overall. Be emotionally ready to cut the paper by a third or half if your colleagues or the editor suggest (or opt for a different journal). Typically there is not enough room in a paper for authors to make all the points they are certain are of earth-shattering importance. Pare down the points to those most important to the most readers. One or two messages are all you can hope the reader will carry away from your study and paper. These most important messages are likely to be lost if you, the author, try to squeeze in more messages of secondary importance. It is a case of “less is more.”
- Put real polish on your paper before submitting it for publication. Editors and reviewers look more favorably on manuscripts that require less work from them to get into shape for publication. Ten or twenty drafts on your part may be about right!
- Whenever possible, let papers sit for a month or longer after they are written before final editing and submission. After getting some distance from your paper, you are more likely to spot its rough areas and find ways to improve it.