jama-reporting-standards

Reporting Standards (jama-reporting-standards)

When to trigger

• You know the design and need the matching EQUATOR checklist
• A flow diagram is missing (participant flow for RCTs, study selection for reviews)
• You are unsure which guideline applies (CONSORT vs STROBE vs PRISMA vs STARD)
• Preparing the checklist file required at submission

Design → guideline map

| Design | EQUATOR guideline | Mandatory diagram |

|------------------------------------------|-------------------|--------------------------------------------|

| Randomized clinical trial | CONSORT | CONSORT participant-flow diagram |

| Observational (cohort, case-control, cross-sectional) | STROBE | Participant-flow diagram strongly expected |

| Systematic review ± meta-analysis | PRISMA | PRISMA study-selection flow diagram |

| Diagnostic-accuracy study | STARD | STARD participant-flow diagram |

| Study protocol | SPIRIT | Schedule of enrollment/assessments |

| Prediction model | TRIPOD | — |

| Quality-improvement study | SQUIRE | — |

| Other designs | The matching EQUATOR guideline | per guideline |

Always check the EQUATOR Network and JAMA's Instructions for Authors for the current guideline list and any JAMA-specific extensions or required checklist file.

What each core guideline forces you to report

CONSORT (RCTs)

• Title states "randomized"; structured abstract follows CONSORT-for-abstracts
• Trial design, allocation ratio, eligibility, settings; any changes after start
• Randomization (sequence, concealment, implementation) and blinding
• Numbers randomized, received intervention, analyzed, and lost — in the flow diagram
• Primary/secondary outcomes with estimated effect size and precision (CI)
• Harms, registration number, protocol availability, funding

STROBE (observational)

• Study design stated early; setting, eligibility, sources, follow-up
• All variables (outcomes, exposures, confounders, effect modifiers) defined
• Bias mitigation, sample-size logic, statistical methods incl. confounding
• Numbers at each stage, descriptive data, and unadjusted and adjusted estimates

PRISMA (systematic reviews / meta-analyses)

• Protocol/registration; explicit eligibility and information sources
• Full reproducible search strategy; selection and data-extraction process
• Risk-of-bias methods; synthesis methods; the study-selection flow diagram
• Certainty-of-evidence assessment (e.g., GRADE)

STARD (diagnostic accuracy)

• Index test and reference standard with rationale and thresholds
• Participant flow and timing; cross-tabulation of results vs reference
• Estimates of accuracy with CIs; handling of indeterminate results

Checklist

• [ ] Correct guideline chosen for the design
• [ ] Completed checklist file prepared with page/line numbers for each item
• [ ] Required flow diagram drafted (CONSORT / PRISMA / STARD / STROBE)
• [ ] Title and abstract conform to the abstract-level extension where one exists
• [ ] Registration number and protocol availability reported where required
• [ ] Adjusted and unadjusted estimates both reported (observational)
• [ ] Certainty-of-evidence / risk-of-bias reported (reviews, observational)

Anti-patterns

• Submitting an RCT with no CONSORT flow diagram
• A systematic review with no PRISMA diagram or no reproducible search
• Filling the checklist with page numbers that do not match the text
• Picking STROBE for a randomized trial (or CONSORT for a cohort)
• Reporting only adjusted estimates in an observational study
• Ignoring the abstract-level extension (e.g., CONSORT-for-abstracts)

Worked example: matching design to checklist (illustrative)

Vignette (illustrative): a multicenter randomized clinical trial, N = 2,800 adults with sepsis, primary outcome 90-day mortality, absolute risk difference -3.1 percentage points (95% CI, -5.8 to -0.4). At the Journal of the American Medical Association this maps to CONSORT with a mandatory participant-flow diagram accounting for everyone screened, randomized, and analyzed; the completed CONSORT checklist must carry page/line locations, the title must state "randomized," and the abstract follows CONSORT-for-abstracts. Had the same question been answered by a cohort, the map would shift to STROBE with both unadjusted and adjusted estimates; a systematic review would shift to PRISMA with a study-selection diagram and a GRADE certainty assessment.

Reviewer pushback and the JAMA fix

• "Reporting incomplete (CONSORT/PRISMA item X)." Fix: add the item and update the checklist's page/line location to match the revised text.
• "Picked STROBE for a randomized trial." Fix: re-map to CONSORT; the design, not convenience, fixes the guideline.
• "Only adjusted estimates reported (observational)." Fix: report unadjusted and adjusted estimates per STROBE.

Calibration anchors (hedge where uncertain): the design-to-guideline mapping and the mandatory flow diagram for RCTs and reviews are durable; JAMA-specific extensions and required checklist files evolve — confirm against the EQUATOR Network and current author guidelines.

Reporting pass for JAMA

Run this as a concrete capability pass. First lock the clinical question, patient population, estimand or endpoint, safety/ethics issue, and reporting checklist; then test whether the manuscript addresses clinical reviewers who ask whether the evidence changes patient care, policy, or medical decision-making while satisfying reporting standards.

• Primary move: Tie every reporting checklist item to a manuscript location; missing registration, denominator, consent, data, or code statements must be explicit blockers.
• Decision ledger: return claim / evidence / blocker / next edit rows so the next pass can patch the manuscript directly.
• Sibling comparison: compare against NEJM for field-changing clinical medicine, Lancet for global-health breadth, specialty journals for narrower clinical domains; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.
• Verification floor: before submission-ready advice, re-open resources/official-source-map.md for volatile rules and name the one unresolved fact that could change the recommendation.

Output format

【Design】...
【Guideline】CONSORT / STROBE / PRISMA / STARD / other
【Flow diagram drafted】yes / no
【Checklist file complete with locations】yes / no
【Items still unreported】...
【Next skill】jama-statistics