How to Write a Good Abstract
A structured summary that lets a reader grasp a study — and judge its validity — at a glance.
A good abstract is structured (IMRaD): Background/Objective, Methods, Results, Conclusion. State the question; in Methods give the design, setting, population, intervention/comparator and the primary outcome; in Results report the key numbers with a measure of uncertainty (a confidence interval) for that primary outcome; and a Conclusion that answers the objective without over-reaching beyond the data.
Read top to bottom: a clear question → appraisable methods → numbers with a CI → an honest conclusion.
The four IMRaD sections every strong abstract carries, and the one thing each must get right. A missing or vague section is a red flag before you reach the full paper: no named primary outcome, results without a confidence interval, or a conclusion that quietly switches to a secondary endpoint.
Check that the Conclusion answers the Objective using the primary outcome reported in Results — and that the effect comes with a confidence interval, not just “significant” or a bare p-value. If the conclusion celebrates a subgroup or secondary finding while the primary outcome was negative, that is spin.
Most clinicians read only the abstract, and many decisions are made on it alone. A well-built abstract lets you appraise validity in seconds; a spun one can change practice on a benefit the data never showed. Spotting spin in an abstract is a core FRCEM critical-appraisal skill.
- Structure:
IMRaD— Background/Objective, Methods, Results, Conclusion - Methods name the
primary outcome; Results give it+ a 95% CI - Conclusion answers the objective —
no over-reach, no spin
Spin, side by side — imagine a trial whose primary outcome (28-day mortality) shows no significant difference: 14.8% vs 15.1%, RR 0.98 (95% CI 0.86–1.11). A clean abstract concludes “the intervention did not reduce 28-day mortality.” A spun abstract instead leads with “significantly fewer ICU days in the pre-specified sickest subgroup” and concludes the drug “improves outcomes” — quietly burying a negative primary result. Reviews of trial abstracts repeatedly find spin: a conclusion favouring a treatment whose primary outcome was non-significant. Always check the conclusion against the primary outcome and its CI.
- Spin / over-reach — a conclusion the primary result does not support.
- Missing the primary outcome, or reporting it without a measure of uncertainty.
- Conclusions resting on a secondary or subgroup finding rather than the results as a whole.
Quick check
What is the biggest red flag in an abstract’s conclusion?
Answer: Spin — claiming a benefit that the results, especially the primary outcome, don’t actually support (e.g. leading with a secondary/subgroup finding when the primary outcome was negative).
Ready to build your plan? EMF Premium gives you all 40,000+ questions, 20 mocks and 1,215 OSCE stations from £29/month — or a one-off 3- or 6-month pass.