The introduction of a research paper is a short funnel of four moves: what is already known, what is still unknown, why that gap matters, and the specific objective your study answers. In a clinical research article, three or four paragraphs can carry all four moves. The final sentence should state an aim your Methods can actually test.
A reviewer reads the introduction to find one thing: the question. Every sentence before the objective should make that question look necessary.
Our overview of the Introduction, Methods, Results and Discussion (IMRaD) structure covers all four sections in brief. This guide covers the introduction alone, paragraph by paragraph.
The introduction of a research paper moves from known to gap to aim
The funnel is a sequence of claims, each narrower than the one before. The Recommendations of the International Committee of Medical Journal Editors (ICMJE) ask the introduction to give the context of the study and state its purpose or hypothesis. They also ask authors to cite only directly pertinent references and to leave out the data and conclusions of the work being reported. Reporting guidelines listed by the EQUATOR Network, such as CONSORT and STROBE, ask for the same pair: background and rationale, then specific objectives.
| Move | Question it answers | Working length | It fails when |
|---|---|---|---|
| 1. Known | What does the field already accept about this problem? | 3–5 sentences | It starts with the disease in general, not the problem you studied |
| 2. Gap | What is still not known, and how do you know? | 2–4 sentences | The gap is asserted, not shown from a search |
| 3. Why it matters | Which decision or patient outcome depends on the answer? | 1–3 sentences | The consequence is vague, such as "further work is needed" |
| 4. Aim | What exactly did this study set out to find? | 1–2 sentences | The aim names an outcome or design the Methods do not contain |
Paragraph one: state what is known, already narrowed to your problem
Open at the level of your question, not the level of the disease. A study of urinary retention after hernia repair does not need a sentence on how many hernias are repaired worldwide. It needs the established facts that make retention a relevant outcome: who is at risk, what it costs the patient and how it is handled now.
Each factual sentence carries a citation, and each citation should be one you have opened. Background sentences are where references get copied from other papers' introductions without being read. Check each one against its source before you keep it, using the routine in how to check a paper before you cite it.
Paragraph two: show the gap from the search you ran
The gap is the centre of the introduction. A research gap is a specific question that the published literature has not answered for your population, setting or outcome. Stating it well means saying what earlier studies did and exactly where they stop. They may have studied a different population, used a smaller sample, left your outcome unmeasured or used a design that cannot answer the question.
Avoid "no study has ever examined" and "this is the first study" unless you have searched properly and can report the search. A novelty claim concerns the whole literature, and one reader holding an earlier paper can disprove it. Describe the search instead: the databases, the date and what it found. Our guide to testing a first-reported-case claim sets out a search you can adapt for other study types.
Paragraph three: say why the gap matters to a decision
A gap is not automatically worth filling. Paragraph three names what changes once the question is answered: a consent conversation, a discharge pathway, the sample size of a future trial, or a choice between two tests. Keep it concrete and in proportion to your design. A single-centre retrospective cohort can inform the planning of a trial, but it cannot settle practice.
If you cannot write this paragraph, say plainly that the work is descriptive or exploratory.
Paragraph four: end on an objective the Methods can answer
The last sentence of the introduction is a contract with the rest of the paper. Every element it names should reappear in the Methods: the design, population, exposure or intervention, comparator and primary outcome.
- Aim or objective — what the study set out to estimate, describe or compare. It suits descriptive and exploratory work.
- Hypothesis — a testable prediction fixed before the data were collected or analysed. State one only when the study was designed to test it, never after seeing the results.
- Verb — match it to the design. A cohort study can estimate an association; a randomised controlled trial can test whether an intervention causes a difference.
Check the match line by line once the Methods are final. An objective that promises to show effectiveness, placed above a Methods section with no comparison group, is a mismatch a reviewer will see on the first read.
A paragraph-by-paragraph template for a clinical research introduction
Copy the template below and replace each slot in brackets. Delete any sentence you cannot support with a reference or with your own search.
Paragraph 1 — what is known
[problem or outcome] occurs in [population] after [procedure, exposure or condition] [reference]. It matters to patients because [consequence] [reference]. At present, [current approach] [reference].Paragraph 2 — what is not known
Previous studies of [problem] have examined [population or setting] [references]. They did not report [missing outcome, subgroup or setting], and [design limitation, if any]. A search of [databases] on [date], using [search strategy location], identified [number] studies in [your population], none of which [specific gap].Paragraph 3 — why it matters
Without this information, [clinicians, patients or services] cannot [decision that depends on the answer]. An estimate of [quantity] would inform [specific use].Paragraph 4 — the objective
In a [study design] at [setting], we aimed to [estimate, describe or compare] [primary outcome] in [population] between [start date] and [end date], and to examine its association with [factors]. Confirmatory designs only: we hypothesised that [prespecified prediction].
Worked example: an invented introduction that fills the template
The introduction below is invented to illustrate the structure. Its statements and citation numbers are placeholders, not a summary of real literature, and it contains no clinical advice.
Urinary retention can follow day-case inguinal hernia repair and can turn a planned same-day discharge into an unplanned overnight admission [1]. For patients it can mean catheterisation, a longer stay and an unexpected night in hospital [2]. At the study centre, patients must pass urine before discharge [3].
Earlier studies of retention after hernia repair reported inpatient cohorts treated before day-case pathways were introduced [4,5]. They did not report how often retention led to unplanned admission. A search of PubMed and Embase in March 2026, using the strategy in Supplementary File 1, found two day-case series, and neither reported admission as an outcome [6,7].
Without that figure, the centre cannot give patients a local estimate of the chance of an overnight stay when they consent to day-case repair.
In a retrospective cohort study at one centre, we aimed to estimate the proportion of adults admitted overnight with urinary retention after day-case inguinal hernia repair between January 2021 and December 2024. We also aimed to examine its association with age and type of anaesthesia.
The objective names the design, setting, population, outcome, period and factors, each of which will reappear in the Methods. Its verbs suit a cohort. The gap is a dated search result, not a first-ever claim, and there is no hypothesis because the study is descriptive and its analysis of associations is exploratory.
Five faults in an introduction and what causes each
| Fault | What causes it | Fix |
|---|---|---|
| Opening with global disease burden | The writer starts where the topic starts, not where the question starts | Begin with the problem your study measured |
| A literature review in miniature | Every related paper is summarised because it was read | Keep only studies that establish what is known or define the gap |
| An asserted gap | The gap comes from belief rather than a search | Report the databases, the date and the result |
| Your own findings in the introduction | The introduction is drafted after the Discussion and borrows from it | Remove every result and conclusion of your own study |
| An objective the Methods do not match | The outcomes changed during the study, or the aim was rewritten after analysis | State the prespecified objective and report any change to it in the Methods |
Keep the introduction short by moving material elsewhere
Extra length comes from material that belongs in another section. Comparisons between your findings and earlier studies go in the Discussion section. Technical detail goes in the Methods, and possible mechanisms go in the Discussion, labelled as speculation.
A reviewer should be able to underline the gap and the objective without rereading. The advice in writing clear scientific English applies here with particular force, because the introduction is read first. Check the target journal's author guidelines for any word limit before you settle the length.
If you spot an error in this guide, tell us how to report a problem and we will review it.