Writing a medical research paper follows a well-established structure that has been standard in health sciences for decades: Introduction, Methods, Results, and Discussion, known as IMRaD. Understanding this framework and the specific expectations within each section is the difference between a manuscript that moves through peer review and one that gets rejected at the editor’s desk. The single most common reason for rejection is lack of novelty, but close behind are inadequate methods and poor writing, both of which are entirely fixable.
Choose Your Journal Before You Write
This sounds counterintuitive, but selecting your target journal first shapes everything from word count to citation format to how you frame your findings. Start with journals you already read and cite in your own work. Look at where similar studies in your field have been published recently. Colleagues and mentors who know the landscape of your specialty can point you toward journals where your topic fits the stated aim and scope.
Don’t rely solely on impact factor when choosing. A more practical approach is to evaluate whether the journal is indexed in major databases like MEDLINE, Scopus, or Web of Science, since indexing indicates quality and ensures your work will actually be discoverable. Check the journal’s editorial board, peer review process, and copyright policies. Transparency in these areas is a strong signal of a legitimate journal. Be wary of journals that advertise vague impact scores from sources that don’t contain actual citation data.
Once you’ve identified two or three target journals (a first choice and backups), download their author guidelines. These will specify word limits, reference style, abstract format, and any reporting checklists required for your study design.
Match Your Reporting Guideline to Your Study Design
Medical journals increasingly require authors to follow standardized reporting checklists, and using the right one from the start will save you from major revisions later. The main ones to know:
- CONSORT for randomized controlled trials: a 25-item checklist plus a flow diagram showing participant allocation, follow-up, and analysis.
- STROBE for observational studies (cohort, case-control, and cross-sectional): a 22-item checklist with extensions for medical subspecialties.
- PRISMA for systematic reviews and meta-analyses: a 27-item checklist with a four-phase flow diagram documenting your search and selection process.
- MOOSE for meta-analyses of observational studies: a 35-item checklist.
If you’re planning a randomized trial, the SPIRIT checklist (33 items) guides protocol development and is designed to transition smoothly into a CONSORT-formatted paper. Using SPIRIT for your protocol makes the final manuscript significantly easier to write.
Register Clinical Trials Before Enrollment
If your study is a clinical trial, registration in a public registry is not optional. Journals affiliated with the International Committee of Medical Journal Editors (ICMJE) require registration at or before the first patient is enrolled as a condition of publication. ClinicalTrials.gov is the most commonly used registry, but any primary register of the WHO International Clinical Trials Registry Platform works, as long as it includes the minimum 24-item registration data set.
Getting ethics board approval alone does not satisfy this requirement. If you miss the window and register late, you’ll need to explain in your manuscript when registration was completed and why it was delayed. Some editors will still consider the paper, but many won’t. Register early.
Writing the Introduction
The introduction is a funnel. Start with the broad clinical problem, narrow to the specific gap in current knowledge, and end with your study’s objective or hypothesis. Every sentence should move the reader closer to understanding why this study needed to be done. A weak study rationale is flagged in roughly 28% of peer-reviewed rejections, so this section carries real weight.
Keep it concise. Three to five paragraphs is typical. Cite recent, relevant literature to establish what is already known, then clearly articulate what remains unknown. Your final paragraph should state the study’s purpose in specific, testable terms.
Writing the Methods Section
The methods section is where most rejected manuscripts fall apart. In one analysis of rejection reports, inadequate elaboration of methodology appeared in over 50% of papers rejected after peer review, and fatal design flaws in about 26%. Reviewers need enough detail to judge whether your findings are trustworthy, and other researchers need enough to replicate your work.
Study Population
Describe who was included and why. Report basic demographics: age, sex, and general health status (healthy volunteers versus patients with a specific condition). Spell out your inclusion and exclusion criteria explicitly. For example, if you studied adults 65 and older with COPD admitted to specific hospitals during a defined time frame, say exactly that. If you excluded patients who died during admission, entered hospice, or refused home care, list those exclusions. For animal studies, report the species, weight, age, and sex.
Equipment, Drugs, and Procedures
The first time you mention any device, provide the specific product name, model number, and manufacturer. Include instrument settings, because these details are essential for replication. Describe calibration procedures and how often calibration was performed. For drugs, report the concentration, dose, frequency, and route of administration. For gases, include flow rates. For chemicals, note the name and concentration.
Statistical Analysis
Name every statistical test you used and explain why it was appropriate for the data type. Specify your threshold for statistical significance (typically p < 0.05). State which software package performed the analysis. This subsection acts as a contract with your reader: it tells them exactly how you moved from raw data to the numbers in your results.
Writing the Results
Present your findings without interpretation. Report what you found, not what it means. Start with descriptive data about your study population (how many participants enrolled, how many completed the study, baseline characteristics), then move through your primary outcome, secondary outcomes, and any subgroup analyses.
How you handle numbers matters. Report p-values to two decimal places when they’re above 0.01, and to three decimal places when below (for example, p = 0.03 or p = 0.004). For very small values, “p < 0.001” is acceptable. Always pair p-values with the actual size of the difference between groups, whether that’s a difference in means, an odds ratio, or a hazard ratio, along with a confidence interval. A p-value alone tells the reader whether a difference exists, but the confidence interval tells them how large it probably is and how precise your estimate is.
For percentages, use one decimal place when your sample size is above 200 (e.g., 34.7%). Name the specific test behind each comparison: chi-square, Fisher’s exact test, or whichever you used. Tables and figures should stand on their own, meaning a reader should be able to understand them without reading the full text.
Writing the Discussion
Open with a brief statement of your main finding. Then put it in context: how does it compare to previous studies? Where does it agree, and where does it diverge? This is the section where you interpret, but interpretation should stay grounded in your data. About 30% of post-review rejections cite an inadequate discussion, often because authors either repeat their results without adding insight or make claims their data can’t support.
Distinguish between statistical significance and clinical importance. A bench study might find a statistically significant difference in a measurement that is so small it would never change how a clinician treats a patient. If your finding is statistically significant but clinically marginal, say so. Conversely, if the difference is large enough to matter in practice, explain which patients it applies to and which it does not.
Every discussion needs a limitations paragraph. Be honest and specific. If your sample was small, say so. If your study was retrospective and therefore couldn’t establish causation, say that. Reviewers already see the limitations; acknowledging them demonstrates scientific rigor rather than weakness.
Constructing the Abstract
Most medical journals require a structured abstract with labeled sections mirroring the paper itself: Background (or Introduction), Methods, Results, and Discussion (or Conclusions). The ICMJE mandates structured abstracts for original research, systematic reviews, and meta-analyses, though the exact format varies by journal. Word limits typically range from 250 to 350 words.
Write the abstract last. It should contain your study’s objective, the design and setting, key methods, primary results with specific numbers, and a one- or two-sentence conclusion. Think of it as a self-contained miniature of the entire paper. Many readers, including reviewers, will form their first impression here.
Authorship and Ethical Disclosures
The ICMJE defines authorship through four criteria that must all be met: substantial contribution to the study’s conception, design, or data analysis; drafting or critically revising the manuscript; final approval of the published version; and agreement to be accountable for the work’s accuracy and integrity. Everyone who meets all four criteria should be listed as an author. Everyone listed must meet all four. Gift authorship (listing someone who didn’t meaningfully contribute) and ghost authorship (excluding someone who did) both violate these standards.
The corresponding author is responsible for ensuring that ethics committee approval, clinical trial registration, conflict of interest disclosures, and other administrative requirements are properly reported, though these tasks can be delegated to co-authors. Prepare these disclosures early so they don’t become a bottleneck at submission.
What Happens After You Submit
Once submitted, your manuscript first goes through an editorial screening. Editors typically make an initial decision within about 10 days. The most common reasons for desk rejection are that the topic isn’t novel enough (flagged in about 46% of desk rejections) or the study falls outside the journal’s scope (about 17%). Papers that pass this screening go to peer reviewers.
Peer review takes considerably longer. The median time from submission to a first peer-reviewed decision is roughly 60 days, though it can stretch well beyond that. If revisions are requested, the timeline to a final decision extends to a median of about 198 days from the original submission. When revisions come back, address every point raised by each reviewer. Write a detailed response letter mapping each comment to the specific change you made, or explaining clearly why you respectfully disagree. Papers that don’t clear review at your first-choice journal can be revised and submitted to a backup journal, and this is completely normal.