The scientific article (IMRAD)
- Concept
An original article is ordered as Introduction, Methods, Results and Discussion, with a title, abstract and references.
What it is
The original article follows the IMRAD structure because it mirrors the research process: why it was done (Introduction), how (Methods), what was found (Results) and what it means (Discussion). Around it sit a title, a structured abstract, keywords, tables and figures, funding and conflict-of-interest statements, and numbered references. Reporting guidelines (CONSORT, STROBE and others) say what each section must contain for each design.
Worked example
- Scenario (made up): a journal allows at most 3 000 words and a 250-word abstract for a cohort study.
- Text budget: Introduction 450 (15 %), Methods 900 (30 %), Results 900 (30 %), Discussion 750 (25 %); total 3 000.
- Abstract: background 40, methods 70, results 90 and conclusion 50 words; total 250.
- Study results (made up): of 240 patients followed, 36 had the outcome: 36 / 240 = 15 %.
- The results sentence gives the count and the percentage (‘36 of 240, 15 %’), not the percentage alone.
- Final check: the 15 % appears identically in the abstract, the text and table 2.
- Reading: the budget gives most room to methods and results, which is what readers need to judge and reproduce the study.
What goes in each section
| Section | Question it answers | Common error |
|---|---|---|
| Title and abstract | What is it about and what did it find? | Abstract that does not match the results |
| Introduction | Why was it done? | A long review that never reaches the question |
| Methods | How was it done? | Too little detail to reproduce the study |
| Results | What was found? | Interpreting instead of reporting |
| Discussion | What does it mean? | Conclusions beyond the data, no limitations |
| References | What does it rest on? | Unchecked citations or citations from predatory journals |
Markers
- IMRAD
- Structured abstract
- Vancouver
- CONSORT
- STROBE
In practice
- To read fast: title, abstract, table 1 (who took part), the main figure or table, and the limitations in the discussion.
- Dissemination does not end at the journal: posters, talks, plain-language summaries and open access bring results to the people who use them.
- Many journals require prior trial registration and the checklist of the matching reporting guideline.
Common pitfalls
- Mixing results with discussion.
- Giving percentages without the count or the denominator.
- An abstract whose figures differ from the text.
- Concluding causation from a cross-sectional or descriptive study.
Test yourself
Think of the answer, then tap the question to see it.
What does IMRAD stand for?
Introduction, Methods, Results and Discussion: the usual structure of an original article, following the order of the research process.
What must the methods section contain?
The design, population, variables and how they were measured, the sample size, the statistical analysis and ethics, in enough detail to repeat the study.
How are references numbered according to medical journal editors' recommendations?
Consecutively, in the order in which they are first mentioned in the text.
What is a common error in the results section?
Interpreting the data instead of reporting them, or giving percentages without the count and the denominator.
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: A. Preparing a Manuscript for Submission to a Medical Journal (International Committee of Medical Journal Editors (ICMJE), 2026); CONSORT 2025 Statement: updated guideline for reporting randomised trials (EQUATOR Network, 2025); The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies (EQUATOR Network, 2026).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.