Scientific Paper Writing for Pediatric Surgeons: Why, What, and How?

Why Write a Scientific Paper

You don't write because you want to say something, you write because you have something to say-Scot Fitzgerald

The advancement of surgical science is the result of contributions from many researchers over many centuries, starting with the great Indian surgeon, Susruta. About 2500 years ago, Susruta compiled a monumental treatise in surgery, the Susruta Samhita, which is the earliest text for the systematic teaching of surgery.[1] Medicine (and surgery) is an imperfect science, and there is a constant improvement/appraisal of knowledge on a daily basis. The aphorism “today's science is tomorrow's nonsense” is true for medical science. Diagnostic methods and treatment protocols keep changing constantly, because of the scientific research published by clinical researchers. High-class research when published in peer-reviewed, reputed journals constitutes new evidence which may alter existing treatment protocols for the better. Thus, to advance surgical science, surgeons should not be just “doers” but also “writers and teachers.” Surgeons may also have a certain degree of moral obligation to publish any novel observation, be it positive or negative, whether it resulted in success or failure.

Apart from contribution to progress of surgical science, there are several other reasons to research and publish;[2] academic and professional promotion, competing for fellowship positions, developing professional contacts, and becoming famous are some of them. Research and writing also improves one's knowledge and judgment and makes one a master in the subject. It helps one to understand and interpret the data presented in conferences and journals, to recognize “good evidence” and not get carried away with flashy presentations representing poor evidence.

WHAT TO WRITE

This is the next question. There are five main areas where the young surgeon can contribute.

Case report and case series: This is usually the first and the simplest scientific writing. An interesting or rare case, especially when successfully managed in a novel way, can be published. Although many international journals do not publish case reports, our JIAPS does publish them primarily to encourage the youngsters to start writing. Sometimes, a case report can be a start to great research, as exemplified by Burkitt's report of 1958[3]

Letter to the editor/correspondence: This is an easy and quick way to get a publication in a leading journal, since most top journals also publish them.[4] A letter is usually written referencing a recently published article. It usually puts forth a different interpretation and critical analysis of the article, backed by published evidence whenever possible. It should be written in a polite language. It also goes through a peer-review process, and the authors of the original article are given a chance to clarify the points raised in the letter. A published letter shows that the author keeps himself/herself regularly updated with recent journal articles, can interpret given data independently, and can write convincingly with logical and scientific basis

Retrospective studies (surgical audit): This constitutes the simplest form of 'original research' and is accepted by all journals. Rather than writing as “our experience with management,” any useful new aspect of management should be highlighted.[5] If possible, the results of two management techniques can be compared, and adding statistical analysis gives further credence to the research paper. A properly done surgical audit can alter clinical practice, when giving the researcher an insight into writing a paper. This requires good record keeping and follow-up data by the department

Prospective clinical studies:[6] Although difficult in private practice, they are not impossible to do. They can be nonrandomized (cohort study) or randomized (randomized control trial). A well-written protocol for the study is essential. Most teaching institutions have a standard research protocol, which forms a template. Once all the necessary columns (objectives, design, inclusion and exclusion criteria, intervention, assessment and safety evaluations, statistics, rights of participants, and any committees involved) are filled satisfactorily, approval is sought from the institutional review board. Once written approval is obtained, the study is registered under the Clinical Trial Registry of India (www.ctri.in). It is important to publish both positive and negative (statistically insignificant) results

Technical innovations: Most surgical journals consider technical reports, and some have a separate section like “how I do it/point of technique.” To be accepted and published, the technical modification should be brief, clinically useful, and well documented with pictures, videos, and illustrations.[7] Most technical breakthroughs are a result of individual ingenuity and perseverance and the ability to think out of the box.

Apart from these, surgeons can write other types of articles also

Review articles: Systematic reviews and meta-analyses allow compilation and interpretation of published literature to update and educate the journal readers. Writing these require special skills and advanced statistical analyses

Editorials/expert opinions: These are usually invited articles on general or specific topics by senior academic surgeons. They should be as objective and unbiased as possible, but the personal opinion of the author is also invariably expressed. Hence, they constitute low level of evidence.

V.V.S.Chandrasekharam

EVIDENCE-BASED MEDICINE AND LEVELS OF EVIDENCE

The judicious use of current best available evidence for the care of individual patients constitutes evidence-based medicine. All available evidence does not fall into the same category, but is listed according to its authenticity. This is not a new concept. Ancient Sanskrit texts mention three levels of evidence, Pratyaksha (direct evidence), Anumana (indirect evidence), and Pramana (published and accepted). Evidence in modern medical science [Table 1] is categorized into five levels (level 1 representing the highest and level 5 the lowest evidence) with some subclassification within.[8] Although a detailed discussion of levels of evidence (LOE) is beyond the scope of this article, it is important to remember that in clinical practice, the best LOE available may be applied to the individual patient. The authors of clinical research papers should understand the LOE associated with their article as some journals (e.g., BJU International) regularly publish the LOE associated with each original article. However, LOE also have certain limitations, hence grading systems are used for clinical recommendations based on evidence.[910]

Table 1

Levels of Evidence for therapeutic studies

Level	Type of study
1A	Systematic review (with homogeneity) of RCTs
1B	Individual RCT (with narrow confidence intervals)
1C	All or none study
2A	Systematic review (with homogeneity) of cohort studies
2B	Individual Cohort study (including low quality RCT, <80% follow-up)
2C	“Outcomes” research; Ecological studies
3A	Systematic review (with homogeneity) of case-control studies
3B	Individual Case-control study
4	Case series (and poor quality cohort and case-control study
5	Expert opinion

RCT: Randomized control trial

BASIC BIOSTATIATICS AND DATA MANAGEMENT

In the ideal situation, every clinical research should involve a biostatistician from the beginning. This is especially true for prospective studies. However, in the practical world, this may not be feasible. Hence, the clinician who is authoring a scientific paper should understand basic biostatistics and statistical methods, so that they can do basic statistics themselves. Some of the basic statistical methods every clinician should know are given in Table 2. A knowledge of data management is necessary to collect, store, and analyze data. Meticulous primary data collection and entry is necessary, including storage of data in electronic systems for easy retrieval and analysis. Many statistical softwares are available to calculate “p” values, most of them free of cost. Most basic clinical research (retrospective and nonrandomized prospective studies) can be performed with these statistical methods. More complex studies should involve a biostatistician, who assists in protocol development and data management (data collection and analysis) and helps in reporting the results in the best possible way. It is important to remember that although P value is the most commonly used tool for calculating statistical significance, it has many limitations and is frequently misinterpreted.[11]

Table 2

Basic statistical methods every surgeon should know

Sample size

Randomization

Softwares for data entry and analysis

Mean/median, SD, SE

Statistical methods (Student’s t-test, Chi-square test, Fischer exact test, Mann-Whitney U-test, CI)

P value (basic idea, calculation, significance)

SD: Standard deviation, SE: Standard error, CI: Confidence interval

CONDUCTING A CLINICAL STUDY THE IDEAL FRAMEWORK

The IDEAL framework consists of forming an idea/innovation followed by its development, evaluation, assessment, and long-term follow-up.[12] This provides the skeleton on which any clinical study can be built. The author's practical steps for beginners are given in Table 3. It is very important to perform a thorough literature search before embarking on further study because it may be a waste of time and effort to pursue an idea that has already been published!

Table 3

Practical steps to conduct basic clinical research

Identify a question/problem

Perform thorough literature search

Re-examine the question for relevance

Formulate hypothetical solution

Test the hypothesis

Apply in clinical setting

Collect adequate follow-up data

Analyze the data

Present and publish the results

Continue to follow-up

HOW TO WRITE

Begin with the end in mind

There are many general rules for writing a scientific paper. The following steps may provide a practical roadmap

It is important to choose the appropriate journal for the work. This choice is based on the quality of the work and the type, quality, and readership of the journal. It is better to avoid publishing in predatory journals, which charge a fee to publish all articles; hence, the peer-review process is unreliable [13]

The authors' instructions of the journal should be read thoroughly

Most medical manuscripts are written in the standard format (IMRAD): Introduction, Methods, Results, and Discussion

A practical sequence when writing a manuscript is to write the aim, methods, and results first. These are specific for the article and writing them is relatively easy straightforward. Then, the introduction and discussion are written, staying true to the results, without straying into additional areas. Abstract and conclusions are generally the last parts to be written.

Let us now examine the specific parts of the manuscript:

Abstract: It is a miniature paper, basically précis-writing, best written at the end. Most journals want structured abstract, following introduction, methods, results, and discussion format (except discussion). It is important to cover all salient points, yet follow the required word limit.

Introduction (Why? 2–3 paragraphs): This should cover four key points: significance of topic in question (what is known), information gaps in available literature (what is unknown), literature review to support key questions, and the hypotheses/objectives the paper wishes to address. It is important to stick to the specific topic in question. An extensive literature review should be avoided, and the introduction should end with a solid statement of the purpose/hypothesis of the study. The key word is “clarity of thought.”

Methods (How? 4–5 paragraphs): This is a clear description of what was used (study population and equipment and inclusion and exclusion criteria), what was done (surgical intervention), how it was done (specific surgical technique and protocols), how outcomes were measured, and how data were collected and analyzed (statistical analysis). A thorough description of all the above points is essential. The key word is “attention to detail.”

Results (What? 3–4 paragraphs): This is a description of the major findings of the study, including measurements and data on the stated primary and secondary endpoints of the study. Results should be described “as they were found,” without judging or interpreting them at this stage. Wherever possible and necessary, tables and figures may be used to show results more comprehensively and clearly. Duplication of the same information in text and tables should be avoided. The key word is “accuracy.” It is important to publish negative (statistically insignificant/inconclusive) results also, because reporting only positive results can cause bias, which wrongly impacts systematic reviews and treatment protocols.[14]

Discussion (So what? 5–6 paragraphs): Discussion summarizes the paper and should put the results into a broader context. Discussion from literature that is not relevant to the results of the present study should be avoided. Overstating the findings of the study or making strong statements should be avoided. Discussion compares the results to similar past studies and discusses why they are similar or different. It also enumerates the strengths and limitations of the study and may suggest any potential changes of practice or prospects for future research. The last paragraph (a separate subheading may be used) should enumerate the conclusions of the study. Conclusions are final, summative statements that reflect the flow and outcomes of the entire paper. The conclusions should come directly from the results, without any speculative statements. Since the discussion is the last and usually the lengthiest part of the article, it is important to keep within the word limit of the article as prescribed by the journal.

Title: The title of the paper is best finalized after writing the entire manuscript. The title should be short, simple, convey the essence of the article, and capture the reader's interest. It can be statement of the purpose or conclusion of the paper, or a question that is answered in the paper.

Keywords: They are words/short phrases that represent the salient topics in the article. They assist in cross-indexing and literature search and usually appear immediately after the abstract.

References: These are written according to the referencing style of the journal. It is important for the authors to read the full text of all articles that are referenced in the manuscript. References should be “Recent, Relevant, and Reliable.” The type of article determines the number of references that can be included. Reference managers (Mendeley/Zotero, etc.) can be utilized for this purpose. A detailed description of citing various types of references can be found online.[15]

Tables, figures, and additional supporting material: Wherever appropriate, methods or results may be arranged in tables to convey a lot of information in a nutshell. A small amount of data that can be written in a few lines do not necessitate the use of tables. Similarly, figures must be used to convey specific information, which may otherwise take a lot of text. Tables and figures must be easy to understand, clear, and well labeled. They should be numbered in the order in which they appear in the manuscript and are accompanied by appropriate legends. Footnotes are used to expand any abbreviations used in the table. When reporting surgical innovations, short, well-edited videos may be submitted to explain the concept clearly. This is especially important for endoscopic and minimally invasive techniques.

Ethical approval, funding, and conflicts of interest: These three are essential subheadings to be mentioned after the discussion, before the references. Any specific concerns and information regarding these should be mentioned. If not relevant for the article, that also must be mentioned.

Reporting guidelines: There are specific guidelines, which set international standards for reporting each type of research study. The guidelines provide a checklist to standardize the trial design and also facilitate the accurate reporting and correct interpretation of the research results.[16]

Publication ethics: The Committee on Publication Ethics (COPE) was founded in 1997 as a voluntary organization to define ethics in scientific publishing. The COPE guidelines address study design, ethical approval, data analysis, conflicts of interest and authorship, peer-review process, redundant publication and plagiarism, duties of editors, advertisements, media relations, and dealing with misconduct. Thus, they are invaluable for authors, editors, reviewers, readers, owners of journals, and publishers. A detailed description of the guidelines is beyond the scope of this article, and the reader is encouraged to go to the COPE website for details.[17]

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.