Manuscript Writing: An Editorial Perspective.

Artificial intelligence (AI) has revolutionised academic writing and medical research. These tools are now used across the full research cycle: literature searches, hypothesis generation, study design, data management, data analysis and manuscript drafting and editing.[ 20 ] As of early 2026, over 83% of high-impact academic journals have formal AI policies in place.[ 21 ] These policies mandate that authors disclose the following in the acknowledgement section: name of the AI software program or tool, manufacturer, version, date of use, a brief description of how the AI was used to develop the manuscript, and a confirmation that the author(s) take responsibility for the integrity of the content generated. One of the main drawbacks of Generative AI is that it can throw up incorrect information. It can produce medical data that sounds plausible but is entirely fabricated. The AI can carry forward biases embedded into its training data. Neither of these limitations will be acceptable in a peer-reviewed submission.[ 20 ] The consensus from the International Committee of Medical Journal Editors and World Association of Medical Editors is clear: AI systems cannot be listed as authors or co-authors.[ 22 ] AI cannot be used to fabricate or manipulate data. Input of sensitive patient data or confidential research data into public Generative AI tools is prohibited. Copying AI-generated text or analysis without human verification of accuracy is not allowed. Authors must take legal or ethical responsibility for their work. The author team must remain fully accountable for accuracy and originality of the research in the absence of plagiarism. The GAMER (Generative Artificial Intelligence tools in MEdical Research) statement, published in 2026, provides a structured 9-item checklist for transparent AI use.[ 23 ] The response of the editorial and publishing team is evolving on the handling of AI in medical writing. While the use of AI to improve grammar or language is acceptable with disclosure, its use for writing the manuscript, especially the introduction and discussion sections, is discouraged. The authors should make every effort to write up the manuscript using their own thought process. Currently, the plagiarism software is being upgraded to take screen for usage of AI in submitted manuscripts, and soon, there will be set cut-offs for its usage, similar to the similarity score cut-offs .

The rejection rate for most high to middle-impact journals is high, with an acceptance rate between 10% and 20%.[ 25 ] The common reasons for rejection are shown in Table 2 . Common reasons for manuscript rejection RCT: Randomised controlled trials Following rejection, the authors should address the editorial comments and revise the manuscript as per the inputs before considering a new journal. If no major methodological flaw is observed and additional or reanalysis is suggested, then the authors can revise the manuscript based on the revised analysis and explore other journals. Suboptimal write-up of the manuscript can also be addressed by approaching experienced researchers with extensive publishing experience or attending focused workshops or courses available online or offline.

The declaration of interests (DOI), also referred to as a conflict of interest statement or a competing interest, is an ethical requirement to maintain transparency in research. The authors must declare any potential competing financial and non-financial assistance in the conduct of the research that might influence a researcher’s objectivity. A transparent DOI is an important tool for scientific integrity, allowing readers, editors and peer reviewers to weigh the study’s findings against potential biases. It is a good practise to mention all the potential conflicts of interest, even if the authors think that it may not be relevant to the submitted work. For studies sponsored or funded by pharmaceutical companies, the authors should clearly state the extent of funding and whether the investigators had complete independent control over the study design, analysis and interpretation of the study results .

Sample size calculation is an important part of any study, especially those evaluating interventions. It is important to involve a biostatistician in the early phases of planning a study and to involve them in the protocol development. The sample size of a study affects funding, resources and the study timeline. Broadly, the requirements for sample size calculation include information on the acceptable alpha level, desired power, one or two-sided study and inputs regarding the baseline event rate and the expected difference between the two study arms (effect size).[ 14 ] The sample size is calculated for the primary outcome. While the alpha error ( P value) is usually set at 0.05 and the power of the study is set at 80%–90%, the baseline event rate and the expected difference between the two arms are the information the investigators need to provide to the statistician for calculating the sample size. For example, we are investigating a new oral ovulation induction agent for treating women with polycystic ovarian syndrome; we need to provide the success rate of the standard of care and the expected benefit. The cumulative live birth rate following 3 cycles of letrozole (standard of care or control) is approximately 30%, and the expected difference between the standard of care and new intervention (effect size) is 10% (based on an earlier preliminary study). The calculated sample size is approximately 360 in each arm, with a total size of 720. We will also need to account for attrition (usually between 10% and 20%) and finalise the sample size. A vast majority of the submitted studies are not adequately powered due to small sample size, and hence, no firm conclusions can be drawn from these inadequately powered trials. Small sample size with erroneous interpretation of the results remains an important reason for manuscript rejection, especially for a high-impact journal .

Plagiarism is a serious ethical violation that goes beyond copying text without an appropriate citation. It is defined as an unauthorised use of data, ideas, or unique techniques or methodologies without proper attribution or crediting the original investigators. Most journals and editors use software tools such as iThenticate or Turnitin to screen manuscripts for plagiarism before peer review or acceptance. If plagiarism or similarity score is significantly high (beyond acceptable cut-offs), manuscripts are immediately rejected and the author group can invite further scrutiny by the editorial team. In serious cases, the editorial team may contact the head of the department or dean of the academic institute and inform them about the research misconduct, which results in a tarnished academic reputation for the author group with potential lifelong implications on his/her academic career. Publishers may place authors on a blacklist, leading to rejection of future submissions for years. If plagiarism is reported after publication, a formal retraction notice is sent, and that label stays linked to the author’s name in databases. This lack of academic credibility can also lead to withdrawal of active grants and rejection of future funding applications. Similarity score or plagiarism is usually revealed following a similarity score check using software during the initial manuscript screening process. For a high similarity score, the editors will investigate the source, and in some instances, accept a higher than acceptable similarity score if it’s justifiable. Methodology description for common procedures/treatment or standard definitions may lead to a high similarity score, which can be acceptable as per the discretionary power of the editor. High similarity score arising from the introduction, results or discussion sections is less acceptable .

For any trial or study involving humans, the investigators must seek approval from the institutional ethical board or committee. For interventional trials (e.g. RCTs), the ethics committee must be registered with the Central Licensing Authority (CLA) under the Ministry of Health and Family Welfare. Government of India, under the New Drugs and Clinical Trials Rules, 2019.[ 17 ] If the institutional ethics committee is not registered with the CLA, the approvals from such committees for clinical trials and trials involving new drugs are not valid. Even retrospective studies need ethics approval for waiver of consent. For interventional trials, it is important that after the ethics approval is obtained, the investigators initiate the process of prospectively registering the trial in the Clinical Trials Registry of India.[ 18 19 ] Here, it is worth noting that ‘prospective’ means that the recruitment of the first participant should be conducted only after the trial is prospectively registered, which is now mandatory. Prospective registration and the availability of the broad protocol in the public domain help increase transparency, mitigate selective reporting and inform other groups about ongoing trials, thereby avoiding duplication of efforts. Many submitted manuscripts do not pay due attention to ethical and regulatory approvals, which invites immediate desk-top rejections. For many high-impact-factor journals, uploading of an ethical approval document is mandatory during the submission process. Lack of information and details about ethical and regulatory approvals in the methods section is often an indicator of partial or non-adherence to the ethical and regulatory requirements and may invite rejection at the initial screening stage itself. Regarding registration, in any ambiguous situations involving human studies, it will be advisable to register the study with a trial registry .

The manuscript writing model has gradually evolved and improved over the years, and since the 1970s, it has become the standard to categorise a manuscript using the IMRAD format (Introduction, Methods, Results And Discussion). The IMRAD format provides a structured and consistent framework that guides authors in systematically addressing the key questions necessary for understanding and interpreting a scientific study.[ 12 ] a) I – Introduction – why did you ask this question? b) M – Methods – what did you do? c) R – Results – what did you find? d) D – Discussion – what might it mean? a) I – Introduction – why did you ask this question? b) M – Methods – what did you do? c) R – Results – what did you find? d) D – Discussion – what might it mean? Authors could create subheadings within these sections to further organise their work. The anatomy of a manuscript varies depending on the type of article: systematic or narrative review, original research, case reports or editorials. The EQUATOR network defines a reporting guideline as ‘A checklist, flow diagram or structured text to guide authors in reporting a specific type of research, developed using explicit methodology’. Reporting guidelines help to improve clarity, transparency, reproducibility, and completeness of presenting scientific research. There are different reporting guidelines for different study designs, and they can be accessed through the EQUATOR network website.[ 13 ] Common examples include CONSORT for RCTs, STROBE for observational studies and PRISMA for systematic reviews. They usually comprise a checklist and a flow chart and follow the IMRAD format. Most indexed journals now mandate that authors enclose these respective reporting guidelines along with their manuscript for publication.

Primary clinical research aims to advance knowledge within a specialty for the benefit of humankind, improve health outcomes and identify better methods to diagnose and treat disease. Publishing medical research is, therefore, a natural extension of conducting it. Although numerous guides and publications on scientific writing already exist, this review does not attempt to present anything entirely new. However, our experience as members of editorial boards suggests that the art of scientific writing remains insufficiently developed among many researchers. Strengthening this skill is essential to improve the quality of published research in reproductive medicine. A sound understanding of basic research methodology is vital for every practising clinician – not only to plan clinically relevant research projects but also to critically interpret scientific literature that informs outpatient decision-making, surgical planning, patient counselling and practices within Reproductive Medicine. This review is intended as a concise ready reckoner to help young researchers, particularly in reproductive medicine, develop their scientific writing skills, while also serving as a practical revision resource for experienced researchers. The primary focus of the review is on clinical research questions around therapeutic interventions.

The current narrative review addresses all the important aspects of manuscript writing and gives an editorial perspective. Manuscript writing serves as a packaging, but the conducted study is the ‘core product’ that needs to be presented to the wider scientific community for scrutiny. The scientific merit and quality of the study ultimately will decide the outcome of the manuscript journey. SG – Conceptualisation, Literature search, Writing– original draft, final approval, SK – Conceptualisation, Literature search, Writing– original draft, final approval, MB – Conceptualisation, Writing – review and editing, Reviewed and gave critical inputs, final approval, MSK – Conceptualisation, Writing – original draft, review and editing, Reviewed and gave critical inputs, final approval. All authors are on the Editorial Board of Journal of Human Reproductive Sciences . However, the manuscript has undergone blinded and independent peer review. Not applicable. AI was used to improve language and grammar.

Following a discussion among the contributory authors of the editorial board, topics were identified and divided among the authors. The search for appropriate studies relevant to the topic was performed through PubMed using appropriate keywords. Since the review was narrative and was not a systematic review, a formal search strategy was not planned. The topics broadly covered in the current review are the following: Framing a research question, appropriate study designs, manuscript writing, ethical and regulatory approvals, plagiarism, and the use of Artificial Intelligence (AI).