Reporting Standards (EQUATOR-Aligned)

6.1 Purpose

  • Clear, complete reporting lets others evaluate, reproduce, and build on published work. We require authors to follow discipline-appropriate reporting guidelines and to submit completed checklists and supporting materials at submission and revision.

6.2 What We Require for Every Submission

  • A completed reporting-guideline checklist mapped to page and line numbers (or section labels) in the manuscript.
  • A brief deviations table explaining any items that are not applicable or not feasible, with justification.
  • A Data/Code/Materials Availability Statement (see Section 7) and citation of any datasets/software with persistent identifiers.
  • Versioned protocols, preregistrations, and analysis plans where applicable, with stable links.
  • Original figure/image files and a short description of any processing (uniform adjustments only).

6.3 Core Guidelines by Study Type

  • Use the most current versions. When a study spans types, include all relevant checklists.
  • Randomized trials: CONSORT (use extensions as relevant: cluster, non-inferiority, pragmatic, pilot/feasibility; CONSORT-Harms; CONSORT-AI for AI-involved interventions).
  • Systematic reviews/meta-analyses: PRISMA 2020 (add PRISMA-S for search strategies; PRISMA-IPD for individual participant data; PRISMA-ScR for scoping reviews).
  • Observational studies (cohort, case-control, cross-sectional): STROBE (plus STROBE-ME for molecular epidemiology).
  • Diagnostic accuracy: STARD.
  • Prediction/prognostic models: TRIPOD (and TRIPOD-AI when applicable).
  • Qualitative research: COREQ or SRQR, as suited to the methodology.
  • Case reports: CARE.
  • Animal research: ARRIVE (including details on randomization, blinding, housing, analgesia, humane endpoints).
  • Health economic evaluations: CHEERS.
  • Quality improvement studies: SQUIRE.
  • Intervention description: TIDieR to fully describe what was done, by whom, where, and with what materials.
  • Omics and specialized domains: MIAME (microarrays), MINSEQE (sequencing), REMARK (tumor markers), MIBBI/FAIR-consistent domain standards where relevant.

6.4 Minimum Reporting Elements (All Empirical Studies)

  • Research question and design (including any preregistration ID).
  • Eligibility/sampling criteria; recruitment setting; dates of data collection.
  • Interventions/exposures or index tests, comparators, and co-interventions; detailed so others can replicate (materials, doses, timing, training).
  • Outcomes (primary/secondary) with precise definitions and timing; any changes from protocol explained.
  • Sample size determination and assumptions (power parameters, interim analyses).
  • Randomization/blinding procedures (sequence generation, concealment, masking) or reasons not feasible.
  • Statistical methods, including model specifications, inference criteria, treatment of missing data, multiplicity adjustments, and sensitivity analyses.
  • Effect sizes with uncertainty (confidence/credible intervals) and exact p-values where appropriate.
  • Data integrity checks, software and versioning, and a reproducible pathway from raw to analyzed data.
  • Harms/adverse events reporting (definitions, ascertainment, severity grading).
  • Limitations and potential sources of bias, with direction and magnitude where possible.

6.5 Images, Figures, and Media

  • State the source of each figure (original experiment, external dataset, or illustration).
  • Disclose uniform adjustments (brightness/contrast) and avoid selective enhancement or splicing.
  • Keep and provide original, unprocessed files on request.
  • Clearly label illustrations or conceptual schematics to avoid implying empirical measurement.

6.6 Negative, Null, and Replication Findings

  • We welcome well-designed negative or null studies and replications. Reporting standards are identical to positive findings; emphasize precision, uncertainty, and power rather than “significance” alone.

6.7 Complex/Hybrid and Interdisciplinary Studies

  • When studies combine methods (e.g., trial + qualitative process evaluation; modeling + empirical data), include all relevant checklists and describe integration procedures (triangulation, mixed-methods design type, model validation).

6.8 AI/ML Studies and Software Papers

  • Provide dataset provenance, licensing, inclusion/exclusion criteria, and known limitations.
  • Report data splits (train/validation/test), cross-validation protocols, and measures to prevent leakage.
  • Present performance with confidence intervals and subgroup analyses; avoid deployment claims beyond evidence.
  • Supply model cards or equivalent documentation and code sufficient for independent verification (with environment details).

6.9 Systematic Search Transparency

  • Include full search strategies for at least one major database (search strings, limits, dates) and state the date of last search.
  • Describe screening, data extraction, risk-of-bias tools, and any automation used (with manual verification).

6.10 Data/Code/Materials Availability (Cross-reference)

  • Provide repository links (with DOIs where possible), access conditions, and licenses.
  • If access is restricted, explain why, outline how qualified researchers can request access, and provide a data dictionary and synthetic or example datasets when feasible.

6.11 Transparency for Deviations

  • Any post-hoc analyses or deviations from the protocol/preregistration must be labeled as such and justified. Primary outcomes cannot be redefined after unblinding without disclosure.

6.12 Unit-Level and Subgroup Reporting

  • Report the unit of analysis (cluster/individual) and any clustering adjustments.
  • Pre-specify subgroups; exploratory subgroup findings must be labeled exploratory.

6.13 Software, Hardware, and Reproducible Environments

  • Name software and package versions; share environment files (e.g., lockfiles, containers) when possible.
  • For custom hardware or devices, document specifications, calibration, and firmware versions.

6.14 Accessibility of Reports

  • Write plain-language summaries where appropriate. Expand acronyms on first use. Provide accessible formats for figures/tables to meet our accessibility standards.

6.15 Editorial Checks and Verification

  • Manuscripts are screened for the correct guideline and checklist completeness.
  • Editors may request raw data, analysis code, protocols, and image originals to verify compliance.
  • High-risk or complex designs may undergo methodological or statistical review in addition to peer review.

6.16 Non-Compliance

  • Incomplete or non-transparent reporting may lead to revise-and-resubmit, rejection, or—post-publication—correction, expression of concern, or retraction if the record is compromised.

6.17 Continuous Improvement

  • Reporting requirements are periodically updated to reflect evolving community standards. Updates appear with version histories and dates on this policy page.