Rules/misleading presentation/US-MIS-039

Do not use statistics/techniques to suggest significance beyond what the data supports

AI-verifiedblockerFDA guidance · non-bindingRaised as a question — a person decides

Rule

A firm-generated presentation should not use statistical analyses or techniques to indicate clinical significance or validity for a finding that the source publication's data does not support.

Check

Check whether any statistical presentation (e.g., a recalculation, a new comparison) implies significance or validity beyond what the source data actually supports. A violation is such an unsupported statistical implication.

Escalate when

Assessing statistical validity claims requires biostatistical judgment.

Where this comes from

Section V, Q5 (prohibited practices) · page 29. The highlighted words are quoted exactly; a script confirms they appear in the source before the rule can be published.

…ion and without immediately following it with the statement identifying any authors, editors, or other contributors to the source publication who were employees of or consultants to or who received compensation from the firm at the time of writing, editing, or contributing to the source publication Present information (e.g., excerpts, quotes, paraphrases, conclusions) from the source publication out of context Use statistical analyses or techniques to indicate clinical significance or validity of a finding not supported by the data or information in the source publication Use presentational elements to obscure or distort the scientific content, such as by using textual features and graphic design elements to emphasize only positive information or distract attention from unfavorable information, or by using tables or graphs to distort or misrepresent the relationships, trends, differences, or changes among the outcomes evaluated in the source publication Nevertheless, it is consis…

Open the original at page 29 on fda.gov ↗Read our full copy →

Document: FDA guidance: Communications From Firms to Health Care Providers Regarding Scientific Information on Unapproved Uses of Approved/Cleared Medical Products: Questions and Answers (final)
Version: as retrieved — the publisher does not version this document; the fingerprint below identifies the exact file
Retrieved: 2026-09-21 from www.fda.gov
Fingerprint: a36a07e1efd950b7a2a8f6b3…

FDA guidance describes the agency's current thinking. It is not binding law, and this rule is a recommendation, not an obligation.

Applies when

Audiencehcp

Basisunapproved-use-scientific

How this rule was checked

Citation proven by script and passed an independent verifier running on a different model. Not yet confirmed by a human.

Drafted
claude-code:sonnet · 2026-09-21
Independent verification — pass
claude-code:opus · 2026-09-21
Reviewer's reasons
- The quote is a complete enumerated item from the Q5 'should not... do any of the following' list, verbatim, and the rule restates it with the subject restored and the recommendation modality preserved.
- Not overstated. The check's examples ('a recalculation, a new comparison') are illustrations of how the failure mode arises and do not broaden the requirement, which remains tied to what the source publication's data supports.
- Atomic: one requirement about statistical analyses or techniques indicating clinical significance or validity not supported by the source data.
- Determinism 'judgment' is right: assessing whether a statistical technique indicates unsupported significance or validity requires biostatistical evaluation, not a literal match.
- Authority 'guidance' matches the source document; nothing statutory is restated.
- applies_when tracks the SIUU-to-HCP scope, with the firm-generated-presentation limit stated in the rule text.
- Severity 'blocker' is defensible: fabricated statistical significance for an unapproved use is a serious misleading claim.

Nearby rules

US-MIS-036 US-MIS-037 US-MIS-038 US-MIS-040