How to Write a Scientific Abstract

A scientific abstract is a short, self-contained summary of a study that states the question, methods, main results and conclusion. It is the part most people read, and often the only part indexed, so it must be accurate and consistent with the paper.

Typical structure

• Background/Objective: the problem and the aim.
• Methods: design, setting, participants, intervention or exposure, main outcome, analysis.
• Results: number analysed, primary outcome result with effect size and confidence interval, key secondary findings.
• Conclusions: what the results support, within the limits of the study.
• Optional: registration number, funding.

How to write it

1. Write the abstract after the paper is complete.
2. Copy the key sentence of each section, then shorten.
3. Include numbers, not only words like "significant".
4. Use the journal's headings and word limit (often 150 to 300 words).
5. Check that every statement is in the paper.
6. Avoid references, unexplained abbreviations and new information.

Illustrative example

Invented for teaching. "Objective: To compare 30-day readmission after hip fracture surgery between early and delayed operation. Methods: Retrospective cohort of 1,000 adults at one hospital (2022–2024). Results: Readmission occurred in 90/400 (22.5%) early and 210/600 (35.0%) delayed patients (adjusted odds ratio 0.65, 95% CI 0.49 to 0.86). Conclusions: Early surgery was associated with lower readmission; residual confounding is possible." (Invented numbers.)

A quick consistency check

After the abstract is written, compare it with the paper in this order: objective, design and sample size, primary outcome, main result, conclusion. Each should appear in both places with the same numbers and wording. If the abstract contains a claim that the Discussion hedges, change the abstract, not the Discussion. Submit the abstract to a colleague who can see only the abstract and ask what they would do differently as a result.

Common pitfalls

• A vague results section ("results were significant").
• Conclusions that go beyond the data.
• Mismatches with the full text.

Key takeaways

• State question, methods, key numbers and a cautious conclusion.
• MRN offers a free abstract template.

This page is general educational information about research methods. It is not medical, legal or regulatory advice; follow your institution's ethics board and your target journal's instructions.