Ask Mallard
Core concept

Choosing a reporting guideline

4 min read

Choose a reporting guideline from the study design and report type, then add relevant extensions and the journal requirements.

Fit

When it matters, and when it does not

It matters when

  • You plan a protocol, prepare a manuscript or check whether methods and results can be understood.
  • You need to distinguish a reporting checklist from design advice or a risk-of-bias tool.

It matters less when

Before you start

What to check in your study

Decide these first
  1. The design actually usedCheck assignment, sampling, data source and whether the work develops or evaluates a model.
  2. The document you are writingProtocols, full reports, abstracts and review reports can need different checklists.
  3. The current statement and relevant extensionsCheck scope and version, then journal instructions. An older extension may need to be read alongside an updated main statement.
It rests on these
  1. The checklist matches the actual work.How to check: Do not relabel an observational study as randomized to obtain a preferred checklist.
  2. A checklist location points to real information.How to check: State what was done, deviations and limitations. Explain genuinely inapplicable items rather than silently dropping them.

Plain language

How it works

Match the report to the statement scope. CONSORT 2025 covers randomized trial reports. STROBE covers cohort, case-control and cross-sectional reports. PRISMA 2020 covers systematic reviews. STARD 2015 covers diagnostic accuracy reports. TRIPOD+AI covers clinical prediction model development and evaluation using regression or machine learning. A prediction model is not simply a diagnostic test.

From the question to a defensible report
Match the report to the actual work. A complete checklist does not establish sound design, low risk of bias or a causal effect.Match the report to the actual workActual study design and data sourceProtocol, full report or other documentCurrent statement and relevant extensionsTransparent account with deviations and limitsA complete checklist does not establish sound design, low risk of bias or a causal effect.
Conceptual diagram. It shows the structure of the idea, not data from any study.
Use the checklist early

Read the statement and its explanations while planning data collection and writing. Check design-specific extensions and journal instructions before submission. For routinely collected observational data, consider RECORD alongside STROBE. For qualitative or other designs, use a guideline written for that work. A checklist cannot confer randomization, remove confounding or verify that an analysis was executed.

Worked example

A worked example

Invented for illustration
Setting
An invented audit checks a subset of 20 applicable reporting items in a cohort report.
Question
Can readers locate the information?
Approach
Choose STROBE for the cohort report. Fourteen checked items have clear manuscript locations.
Result
14 of 20 checked items can be located, 70 in 100. Six checked items still lack information.
What it means
This measures reporting completeness in the audit, not risk of bias or validity. A completed checklist does not repair confounding.

In practice

Recent studies PubMed search · not vetted

Searching PubMed…

Watch out

Common mistakes

The one to remember

A completed reporting checklist is not proof of sound design, low risk of bias or a causal effect. Choose the guideline for the actual study and report, then check relevant extensions.

  • Choosing CONSORT because the study compares treatments.Instead: Use CONSORT for randomized trial reports. STROBE covers cohort, case-control and cross-sectional reports; treatment comparison alone does not imply randomization.
  • Treating a reporting checklist as a quality score.Instead: Assess risk of bias separately. Transparent reporting makes flaws visible without removing them.
  • Using the same checklist for protocols and final results.Instead: Use SPIRIT for randomized trial protocols and CONSORT for trial reports. Check the current versions and applicable extensions.

Inside Mallard

How Mallard uses this

Mallard's crossover Learn article includes its design-specific reporting guideline. Plans do not link to this concept yet.

No Mallard code example points to this page.
Atlas node concept.reporting_guideline, the id a plan cue records.

Related Learn pages

Related designs and methods

Evidence

Sources & further reading

  1. CONSORT 2025 statement: updated guideline for reporting randomised trials (opens in a new tab) · BMJ · doi 10.1136/bmj-2024-081123
  2. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies (opens in a new tab) · BMJ · doi 10.1136/bmj.39335.541782.AD
  3. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews (opens in a new tab) · BMJ · doi 10.1136/bmj.n71
  4. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies (opens in a new tab) · BMJ · doi 10.1136/bmj.h5527
  5. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods (opens in a new tab) · BMJ · doi 10.1136/bmj-2023-078378
  6. SPIRIT 2025 statement: Updated guideline for protocols of randomised trials (opens in a new tab) · PLoS Med · doi 10.1371/journal.pmed.1004589
  7. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement (opens in a new tab) · PLoS Med · doi 10.1371/journal.pmed.1001885
Prepared with AI assistance. Each source was verified by PMID against PubMed on 2026-11-06, matching title, authors, year and DOI. Record verification does not establish support for every use. No independent professional or methodological approval is claimed.
Last source review: 2026-11-06
AI-generated educational content
Mallard uses reputable methodological and reporting sources, but this material may contain errors or omit important context. Verify important methodological decisions with the cited sources and, when appropriate, a qualified research methodologist or statistician.