IMRAD manuscript writing is the foundation of biomedical research communication. Every original research article published in a peer-reviewed medical journal follows the same core architecture: Introduction, Methods, Results, and Discussion. This structure, standardized by the ICMJE (International Committee of Medical Journal Editors) and adopted by over 5,000 biomedical journals worldwide, ensures that research findings are presented in a consistent, reproducible, and evaluable format. Whether you are drafting your first clinical study report or preparing a multi-site trial for submission, mastering the IMRAD format determines whether your manuscript reaches peer review or lands in the desk rejection pile.
This section-by-section guide walks you through every component of how to write a medical manuscript using the IMRAD structure. You will learn what each section must contain, how long each section should be, which reporting checklists to use for your study design, and which mistakes trigger immediate rejection. The goal is a publication-ready manuscript that satisfies editors, survives peer review, and communicates your findings with precision.
What Is the IMRAD Format?
IMRAD stands for Introduction, Methods, Results, and Discussion -- the four mandatory sections of an original research article in biomedical journals. IMRAD is not a suggestion or a stylistic preference. It is a structural requirement enforced by the ICMJE Recommendations (2023) and adopted by virtually every indexed medical journal on the planet.
Definition: IMRAD stands for Introduction, Methods, Results, and Discussion -- the standardized structure for original research articles in biomedical journals. Adopted by the ICMJE (International Committee of Medical Journal Editors) and required by over 5,000 biomedical journals, IMRAD ensures consistent, reproducible research communication by separating the rationale, methodology, findings, and interpretation into distinct sections.
The IMRAD framework structures medical manuscripts by assigning each section a specific communicative function. The Introduction answers "why was this study done?" The Methods section answers "how was it done?" Results answers "what was found?" And the Discussion answers "what does it mean?" This separation is not arbitrary. It allows peer reviewers to evaluate methodological rigor independently of the authors' interpretation. It allows readers to jump directly to the section most relevant to their needs. And it provides editors with a predictable format that accelerates the review process.
The history of IMRAD traces back to the mid-20th century, but its dominance was cemented when the ICMJE formally adopted it in the 1970s. Today, the EQUATOR Network (Enhancing the QUAlity and Transparency Of health Research) builds on IMRAD by providing study-design-specific reporting checklists that specify what content must appear within each IMRAD section. IMRAD is the skeleton; reporting checklists are the muscles and organs.
Understanding IMRAD format explained at this structural level is the first step toward writing a manuscript that editors take seriously. The sections that follow break down each component in detail, with word targets, content requirements, and the specific mistakes that lead to rejection.
Before You Write -- Pre-Writing Preparation
Successful journal manuscript preparation begins before you type a single word of the manuscript. Pre-writing preparation determines whether your manuscript hits its target on the first submission or cycles through months of reformatting and resubmission. Three preparatory steps are non-negotiable: selecting your target journal, identifying the correct reporting checklist, and organizing your data and author contributions.
Select your target journal first. The journal you choose dictates word count limits, reference style, figure requirements, abstract format, and even which IMRAD subsections are required. Writing a manuscript without a target journal is like building a house without a blueprint -- you will inevitably tear parts down and rebuild. Read the journal's author guidelines before you outline. Check the aims and scope statement to confirm the journal publishes your study type. Review two or three recent articles in the same journal to understand the expected depth and tone. For detailed guidance, see our journal selection guide.
Identify the correct reporting checklist. The EQUATOR Network maintains reporting guidelines for every major study design. The four most commonly used are:
| Study Design | Reporting Checklist | Items |
|---|---|---|
| Randomized controlled trial | CONSORT 2010 | 25 items |
| Observational study (cohort, case-control, cross-sectional) | STROBE Statement | 22 items |
| Systematic review | PRISMA 2020 | 27 items |
| Meta-analysis of observational studies | MOOSE | 35 items |
Each checklist specifies exactly what information must appear in each IMRAD section for that study design. The CONSORT statement tells you what to include in the Methods section of a randomized trial. The STROBE checklist does the same for observational studies. PRISMA 2020 governs systematic reviews and meta-analyses of interventional studies. These checklists are not optional supplements -- most journals require a completed checklist at submission, and reviewers use them to evaluate your manuscript. Download your checklist before you start writing and use it as a structural blueprint.
Organize your data and clarify author contributions. Compile all statistical outputs, tables, and figures before drafting. Confirm that every co-author meets the four ICMJE authorship criteria: (1) substantial contribution to conception/design or data acquisition/analysis, (2) drafting or critical revision, (3) final approval, and (4) accountability for the work. Authorship disputes that surface mid-submission delay publication by months.
Establish your writing order. Most experienced medical writers recommend drafting in this sequence: Methods first (most concrete), then Results (data-driven), then Discussion (interpretation), then Introduction (framing), and finally the Abstract (summary). This order follows the logic of the research process rather than the reading order, and it prevents you from writing an Introduction that promises findings you cannot deliver.
Introduction -- Establishing the "Why"
The Introduction answers one question: why was this study necessary? It uses a funnel structure that moves from broad context to a specific knowledge gap to a precise research objective. A well-written Introduction is three to four paragraphs long -- roughly 400 to 600 words -- and ends with a clear statement of the study's aim or hypothesis.
Paragraph 1: Broad context. Open with the clinical or scientific problem your study addresses. Establish the topic's importance using epidemiological data, clinical burden, or public health significance. Cite foundational references but do not write a literature review. Two to three sentences are sufficient.
Paragraph 2: What is known. Summarize the current state of evidence. What have previous studies shown? What interventions, exposures, or approaches have been tested? Cite the most relevant and recent studies -- typically three to six references. This paragraph demonstrates that you understand the existing literature and positions your study within it.
Paragraph 3: The gap. Identify what remains unknown, unresolved, or inadequately studied. This is the most important paragraph in the Introduction because it justifies your study's existence. Be specific. "More research is needed" is not a gap statement. "No randomized trial has compared intervention X to intervention Y in population Z" is a gap statement. The gap must logically lead to your research question.
Paragraph 4: Objective and hypothesis. State the study's aim in one or two sentences. Use language that mirrors the study design: "We aimed to determine..." for hypothesis-testing studies, "We aimed to explore..." for exploratory studies. If your study has a formal hypothesis, state it here. This paragraph should also briefly mention the study design (e.g., "We conducted a randomized, double-blind, placebo-controlled trial...") to orient the reader before they enter the Methods section.
The Introduction must not include results, methods, or conclusions. It must not exceed four paragraphs in most journals. And it must cite the specific knowledge gap that your study fills -- vague justifications are a hallmark of weak manuscripts and a common trigger for desk rejection reasons. Medical manuscript follows ICMJE guidelines from the very first section: the ICMJE Recommendations (2023) explicitly state that the Introduction should "provide context but not review the subject extensively."
Methods -- Detailing the "How"
The Methods section is the most scrutinized section during peer review. In our experience preparing manuscripts for submission, the Methods section receives the most peer reviewer scrutiny -- and insufficient methodological detail is the single most common reason for desk rejection. ICMJE Recommendations require that Methods be described "in sufficient detail to allow others to replicate the work" (ICMJE, 2023). This section must be written in the past tense and organized under clear subheadings.
Study Design
State the study design in the first sentence (e.g., "This was a prospective, single-center, randomized controlled trial..."). Name the specific reporting checklist you followed. For an RCT, write: "This trial is reported in accordance with the CONSORT 2010 Statement." For a systematic review, reference PRISMA 2020. The study design declaration tells reviewers which checklist to use when evaluating your manuscript and demonstrates methodological awareness. CONSORT reports randomized controlled trials according to a standardized 25-item framework that specifies what belongs in each Methods subsection.
Participants
Describe eligibility criteria (inclusion and exclusion), recruitment setting, recruitment period, and how participants were identified. Report the sample size calculation, including the expected effect size, alpha level, power, and the statistical test used to determine the required n. A missing sample size calculation is a red flag that reviewers consistently identify.
Intervention or Exposure
For trials, describe the intervention and comparator in enough detail for replication -- dosing, frequency, duration, mode of delivery, and who delivered it. For observational studies, define the exposure variable, its measurement, and any categorization used. If the intervention is complex, consider referencing a published protocol or supplementary materials.
Outcome Measures
Define the primary outcome measure and all secondary outcomes. Specify how each outcome was measured, who measured it, and at what time points. Validated measurement instruments should be cited. If composite outcomes were used, define the individual components. Clear outcome specification prevents post-hoc outcome switching -- a form of reporting bias that peer reviewers are trained to detect.
Statistical Analysis
Describe every statistical test used, the software and version, the significance threshold, and how missing data were handled. Report whether intention-to-treat or per-protocol analysis was used (for trials). For regression models, list all covariates and the rationale for their inclusion. Statistical analysis details are where reviewers most often request additional information, so err on the side of completeness.
Ethical Approval
State that the study was approved by an IRB (Institutional Review Board) or ethics committee, including the approval number. Confirm that informed consent was obtained from all participants. For studies involving human subjects, ICMJE Recommendations (2023) require ethical approval documentation. State trial registration (if applicable) with the registry name and number.
The Methods section typically runs 800 to 1,200 words. Every reporting checklist item that applies to the Methods section of your study design must be addressed. If in doubt about whether a detail is necessary, include it -- reviewers can always ask you to trim, but they will reject a manuscript with missing methodological information.
Results -- Presenting the "What"
The Results section presents findings objectively, without interpretation. State your primary outcome in the first sentence. Do not bury the lead -- readers and reviewers want to know the main finding immediately, and everything else is context for that finding. The Results section typically runs 600 to 1,000 words, excluding tables and figures.
Report the primary outcome first. Begin with participant flow (how many were enrolled, randomized, analyzed) and then state the primary outcome with its effect size, 95% confidence interval, and p-value. For example: "The intervention group showed a 15% reduction in 30-day readmission rates compared to the control group (RR 0.85, 95% CI 0.74-0.97, p = 0.02)."
Use tables and figures strategically. Tables present exact values; figures show trends and comparisons. Every table and figure must have a descriptive caption that allows it to be understood without reading the main text. Number tables and figures in the order they are first mentioned. Do not duplicate data -- if a finding appears in a table, reference the table in the text rather than repeating the numbers. For systematic reviews, the Results section should include a PRISMA flow diagram showing study identification, screening, eligibility, and inclusion. You can create one using our free PRISMA flow diagram generator.
Report secondary outcomes and subgroup analyses. After the primary outcome, present secondary outcomes in the order listed in the Methods section. Subgroup analyses should be clearly labeled as pre-specified or exploratory. Pre-specified subgroup analyses carry more weight than post-hoc explorations, and reviewers will check whether your reported subgroups match those described in Methods.
Present adverse events and missing data. For clinical trials, the CONSORT 2010 Statement requires reporting all adverse events by group, including serious adverse events. Report the amount of missing data and the method used to handle it (complete case analysis, multiple imputation, etc.). Transparency about missing data builds reviewer confidence.
What to avoid in Results. Do not interpret findings -- save interpretation for the Discussion. Do not use words like "significant" without specifying "statistically significant" and providing the test statistic. Do not selectively report favorable outcomes while omitting unfavorable ones -- this constitutes selective reporting and violates the ICMJE guidelines. Do not introduce methodological details that belong in Methods.
A clean Results section follows the order established in the Methods section, presents every pre-specified outcome, uses tables and figures to reduce text density, and maintains strict objectivity. Peer review evaluates manuscript quality most harshly when Results and Discussion bleed into each other.
Discussion -- Interpreting the "So What"
The Discussion section interprets your findings, contextualizes them within the existing literature, acknowledges limitations, and states implications. It answers the question "so what?" and is typically the longest section at 1,000 to 1,500 words. The Discussion follows a predictable internal structure organized around five functions.
Key Findings
Open the Discussion by restating the primary finding in one to two sentences. Do not repeat the statistical values from Results -- summarize the finding in plain language. For example: "This trial demonstrated that the telehealth intervention significantly reduced 30-day readmission rates compared to usual care." This opening sentence anchors the entire Discussion and tells the reader what to expect.
Comparison with Existing Literature
Compare your findings with those of previous studies. Where do your results agree? Where do they diverge? If your findings contradict prior work, propose explanations -- differences in population, intervention, measurement, or analysis. Cite at least three to five relevant studies and engage with them substantively rather than simply listing them. This section demonstrates your command of the field and positions your contribution within the broader evidence base.
Strengths and Limitations
Every study has limitations. Reviewers expect you to identify them honestly. Common limitations include small sample size, single-center design, short follow-up, self-reported outcomes, and potential confounding. For each limitation, explain how it might affect the interpretation of your results. Then state the study's strengths -- rigorous design, validated measures, adequate power, long follow-up, or novel population. Presenting strengths and limitations together shows balanced scientific judgment.
Clinical and Research Implications
State what your findings mean for clinical practice, policy, or future research. Be specific and proportionate. A single RCT rarely changes practice guidelines; it contributes evidence that may inform future guideline updates. Suggest specific next steps for research -- larger trials, different populations, longer follow-up, mechanistic studies. Avoid vague statements like "more research is needed" without specifying what kind of research and why.
Conclusion
End the Discussion with a concise conclusion (two to three sentences) that restates the main finding and its primary implication. Some journals require a separate Conclusion section; others embed it as the final paragraph of the Discussion. Check your target journal's guidelines.
Critical rules for the Discussion. Do not introduce new data or results that were not presented in the Results section. Do not overstate your findings -- language like "proves" or "demonstrates conclusively" is inappropriate for a single study. Do not repeat the Results section verbatim. Do not ignore findings that did not support your hypothesis.
The Abstract -- Written Last, Read First
The Abstract is the single most important piece of text in your manuscript. It is the first element editors read when deciding whether to send your paper to reviewers. It is the only part most readers will ever see. It must function as a standalone document that accurately summarizes the entire study.
Most biomedical journals require a structured abstract format with four labeled sections that mirror IMRAD: Background (or Objective), Methods, Results, and Conclusions. Word limits typically range from 150 to 350 words. The ICMJE Recommendations (2023) specify that abstracts should be structured and should not exceed the journal's word limit.
Background/Objective (2-3 sentences). State the clinical problem and the study's specific aim. Extract this from the final paragraph of your Introduction.
Methods (3-4 sentences). Describe the study design, setting, participants, intervention, primary outcome measure, and main analytical approach. Extract this from your Methods section, compressing to essentials.
Results (3-4 sentences). Report the primary outcome with the effect size, confidence interval, and p-value. Include the sample size and key secondary outcomes if space permits. Extract this from your Results section.
Conclusions (1-2 sentences). State the main conclusion and its primary implication. This should match the final paragraph of your Discussion -- not overstate it.
Write the Abstract last, after all other sections are complete. Draft it by extracting one key sentence from each IMRAD section, then refine for flow and word count. Ensure the Abstract does not contain information that is absent from the main text -- abstract-body mismatch is a common desk rejection trigger. Do not include references, abbreviations (except universally recognized ones), or table/figure citations in the Abstract.
Common IMRAD Mistakes That Lead to Desk Rejection
Desk rejection occurs when an editor rejects a manuscript without sending it to peer review. Rates vary by journal but commonly range from 30% to 70% of submissions. Many desk rejections are caused by structural and formatting errors -- problems that are entirely preventable. Here are the most common IMRAD-related mistakes that trigger immediate rejection.
Vague or incomplete Methods. Insufficient methodological detail is the leading cause of desk rejection at biomedical journals, ahead of scope mismatch and statistical errors. Missing sample size calculations, undefined outcome measures, unreported statistical tests, and absent ethical approval statements all signal a manuscript that is not ready for review.
Results that include interpretation. When authors write "the significant improvement demonstrates that the intervention is effective" in the Results section, they have crossed into Discussion territory. Results must be objective. Interpretation belongs exclusively in the Discussion.
Discussion that introduces new data. Presenting data in the Discussion that was not reported in Results is a structural violation. If you discover an important finding while writing the Discussion, add it to the Results section first, then discuss it.
Introduction that is too long. An Introduction that reads like a literature review signals that the author does not understand the section's purpose. Four paragraphs, 400 to 600 words. If your Introduction exceeds 800 words, you are almost certainly including material that belongs elsewhere.
Abstract-body mismatch. The Abstract reports a different primary outcome, different sample size, or different conclusion than the main text. This happens when authors revise the manuscript but forget to update the Abstract. Reviewers and editors check for consistency, and mismatches destroy credibility.
Missing reporting checklist. Many journals require a completed CONSORT, STROBE, or PRISMA checklist at submission. Submitting without it results in administrative rejection before the manuscript reaches an editor. Even when not explicitly required, including a completed checklist signals methodological rigor.
Incorrect formatting. Wrong reference style, figures at incorrect resolution, exceeding word count limits, missing title page elements, or incorrect file formats. These are administrative errors, not scientific ones, but they trigger desk rejection just the same. For a comprehensive list, read our guide to common reasons for rejection.
Reporting Checklists That Complement IMRAD
IMRAD provides the section structure for your manuscript. Reporting checklists specify what content must appear within each section for your particular study design. The two systems are complementary, not interchangeable. Using IMRAD without a reporting checklist produces a well-structured but potentially incomplete manuscript. Using a reporting checklist without IMRAD produces complete content in a non-standard format. You need both.
CONSORT 2010 for randomized controlled trials. The CONSORT 2010 Statement includes a 25-item checklist and a participant flow diagram. CONSORT specifies what to report in each IMRAD section: title and abstract identification (Item 1a-1b), introduction rationale and objectives (Items 2-2b), and detailed methodological elements including trial design, participants, interventions, outcomes, sample size, randomization, blinding, and statistical methods (Items 3a-12b). The Results section covers participant flow, baseline data, outcomes, and harms (Items 13a-19). The Discussion covers limitations, generalizability, and interpretation (Items 20-22). CONSORT reports randomized controlled trials with a level of specificity that IMRAD alone cannot provide.
STROBE for observational studies. The STROBE Statement provides a 22-item checklist for cohort, case-control, and cross-sectional studies. STROBE specifies reporting requirements for study design, setting, participants, variables, data sources, bias, study size, quantitative variables, and statistical methods within the Methods section. It requires specific content in Results (participants, descriptive data, outcome data, main results, other analyses) and Discussion (key results, limitations, interpretation, generalizability).
PRISMA 2020 for systematic reviews. PRISMA 2020 (Preferred Reporting Items for Systematic reviews and Meta-Analyses) includes a 27-item checklist and a flow diagram showing the study selection process. SR follows PRISMA 2020 reporting guidelines to ensure transparent reporting of the search strategy, study selection, data extraction, risk of bias assessment, and synthesis methods. PRISMA 2020 updated the original 2009 statement to address advances in systematic review methodology, including new items for automation tools, certainty of evidence, and registration. For systematic review manuscripts, see our complete systematic review guide.
MOOSE for meta-analyses of observational studies. The MOOSE (Meta-analysis Of Observational Studies in Epidemiology) guidelines provide a 35-item checklist tailored to the unique challenges of synthesizing observational data, including heterogeneity assessment, confounding, and bias evaluation.
The EQUATOR Network as your central resource. The EQUATOR Network maintains an online library of over 500 reporting guidelines organized by study design. Before you start writing, visit the EQUATOR Network website, identify the guideline that matches your study design, and download the checklist. Use it as your writing template -- not as a post-hoc checklist you fill in after the manuscript is complete.
| Reporting Guideline | Study Design | Checklist Items | Flow Diagram |
|---|---|---|---|
| CONSORT 2010 | Randomized controlled trials | 25 | Yes |
| STROBE | Observational studies | 22 | No |
| PRISMA 2020 | Systematic reviews / meta-analyses | 27 | Yes |
| MOOSE | Meta-analyses of observational studies | 35 | No |
| EQUATOR Network | All study designs | 500+ guidelines | Varies |
When to Get Professional Medical Writing Support
Not every researcher needs professional help with IMRAD manuscript writing. But certain situations make a medical writing service not just helpful but strategically important.
First-time authors. If you have never published an original research article, the gap between knowing your results and knowing how to present them in a publication-ready manuscript is significant. A professional medical writer structures your manuscript for IMRAD structure compliance, formats it to your target journal's specifications, and ensures your reporting checklist is complete. This accelerates your learning curve and dramatically reduces the risk of desk rejection on your first submission.
Non-native English speakers. Language quality is not supposed to influence peer review, but in practice it does. Reviewers who struggle to parse unclear sentences may interpret language problems as scientific problems. A medical writing for journals service ensures that your science is communicated with the clarity and precision that English-language journals expect, without altering your scientific contribution.
Complex multi-site or multi-arm studies. Studies with multiple intervention arms, sites, or endpoints produce complex Results sections that are difficult to organize without experience. A professional writer knows how to structure these sections, design clear tables, and present subgroup analyses in a way that reviewers can follow. Clinical research writing for complex designs is a specialized skill that takes years to develop.
Time pressure. Grant deadlines, fellowship applications, promotion timelines, and co-author availability all create pressure to publish quickly. A medical writing service compresses the writing timeline from months to weeks. Researchers who attempt to rush the writing process independently often produce manuscripts that require multiple revision cycles, ultimately taking longer than if they had engaged professional support from the start.
Systematic reviews and meta-analyses. These study types involve unique reporting requirements -- PRISMA flow diagrams, forest plots, risk of bias tables, GRADE evidence profiles -- that many clinical researchers have limited experience preparing. Research Gold specializes in evidence synthesis manuscripts and can structure your systematic review for PRISMA 2020 compliance from protocol through publication. For more information, see our medical writing services guide.
In every case, the researcher retains full intellectual ownership and authorship. Professional medical writing for journals is acknowledged in the manuscript per ICMJE Recommendations (2023) and follows the ethical standards established by GPP3 (Good Publication Practice 3). The writer structures, formats, and polishes -- the science remains yours.
Typical word count targets by IMRAD section. The table below provides general targets for a standard original research article. Journal-specific limits take precedence.
| IMRAD Section | Word Count Target | Primary Function |
|---|---|---|
| Abstract | 150-350 words | Standalone summary |
| Introduction | 400-600 words | Context, gap, objective |
| Methods | 800-1,200 words | Design, participants, analysis |
| Results | 600-1,000 words | Objective findings, tables, figures |
| Discussion | 1,000-1,500 words | Interpretation, context, limitations |
| Total manuscript | 3,000-5,000 words | Complete original research article |
Mastering how to structure a medical manuscript for journal submission is a career-long skill. The IMRAD format provides the architecture. Reporting checklists fill in the details. And professional support is available when you need it. The goal is always the same: a manuscript that communicates your research clearly, survives peer review, and advances knowledge in your field.