Tables and graphs must not distort study relationships or trends
AI-verifiedwarningBinding lawRaised as a question — a person decides
Rule
Tables or graphs may be misleading if they distort or misrepresent the relationships, trends, differences, or changes among the studied variables or products (e.g., mislabeled axes).
Check
Review graphic/table axis labeling, scale, and presentation for distortion of the underlying data relationships.
Escalate when
Whether a graphic creates a distorted impression is a matter of visual/statistical judgment.
Where this comes from
21 CFR 202.1(e)(7)(iv). The highlighted words are quoted exactly; a script confirms they appear in the source before the rule can be published.
…ted to have clinical significance or validity, or fails to reveal the range of variations around the quoted average results. (iii) Uses statistical analyses and techniques on a retrospective basis to discover and cite findings not soundly supported by the study, or to suggest scientific validity and rigor for data from studies the design or protocol of which are not amenable to formal statistical evaluations. (iv) Uses tables or graphs to distort or misrepresent the relationships, trends, differences, or changes among the variables or products studied; for example, by failing to label abscissa and ordinate so that the graph creates a misleading impression. (v) Uses reports or statements represented to be statistical analyses, interpretations, or evaluations that are inconsistent with or violate the established principles of statistical theory, methodology, applied practice, and inference, or that are derived from clinical studies the design, data, or conduct of …
- Document
- 21 CFR 202.1
- Version
- 2026-09-16 — the link above is pinned to this date, so it shows the text as it stood then
- Retrieved
- 2026-09-21 from www.ecfr.gov
- Fingerprint
- a542640a872b3f2702d549e8…