Research methods guide
How to write a medical research Methods section
Describe design, source population, enrollment, measurements, models, and main analysis decisions. Give readers enough information to reproduce the key choices.
Compare two outpatient management strategies and 90-day readmission in a synthetic two-center cohort. Another analyst should be able to reconstruct the sample and main results from Methods.
Methods sections and audit trail
Tie each passage to the dictionary, scripts and results.
| Item | Worked example | Verification |
|---|---|---|
| Design and eligibility | Retrospective new-user cohort indexed at first eligible initiation after 180 days of observed history. | Give sites, dates and handling of ties and repeated visits, with the flow diagram. |
| Variables and measurement | First unplanned admission on days 1–90, identified by reviewed admission fields. | Specify planned-admission exclusions, external coverage, coding and deaths. |
| Analysis and sensitivity | Use pre-exposure covariates and standardized predicted risks to report a risk difference and interval. | Specify model, missing-data and interval methods plus alternate outcome definitions. |
Apply the method to your materials
Match the actual analysis version
Compare data cutoff, software and script parameters. Distinguish planned and later analyses and document deviations.
Reconstruct one main table row
Trace each denominator, estimate and interval. Ask a colleague to reproduce one row and fill any undocumented rule.
Questions that arise in practice
Can we say standard regression was used?
Specify model, contrast, variable forms, references, missingness and interval method. Put longer code in a supplement while retaining interpretive details.
Verified sources
The official sources below were checked on 2026-11-12. Access conditions and update dates are rechecked before publication.
RigorMed workspace
Start with your research materials
Bring your research question, available materials, and the steps you want to check.
