Evidence literacy
“No significant difference” does not establish aquarium compatibility
Compare a fictional confidence interval with a declared equivalence region and show why including zero is insufficient.
By FishyMag · Published
Worked arithmetic example · Not a clinical review or treatment recommendation.
Quick answer
The fictional 90% interval is −8 to +4 percentage points. It includes zero, but extends beyond the stipulated equivalence region of −5 to +5. Under the stated equal-tail two-one-sided-test interval criterion, equivalence is not demonstrated. The illustrative region has no established aquarium acceptance justification.


Figure data and explanation
Original fictional equal-tail interval exercise. The −5 to +5 region is stipulated solely for the containment calculation.
- Illustrative region
- The region runs from −5 to +5 points.
- Reported interval
- The interval runs from −8 to +4 points.
- Containment result
- The lower limit lies outside the region.
A no-significant-difference summary cannot replace a completed equivalence assessment.
Specify which question the interval is supposed to answer
This original fictional report gives an adverse-outcome difference of −2 percentage points and an equal-tail 90% confidence interval from −8 to +4 for its day-14 endpoint. For the exercise, a predeclared region of −5 to +5 points is stipulated. No biological justification is claimed for that region. FDA’s May 2026 bioequivalence guidance explains the statistical connection between two one-sided tests and an applicable equal-tail 90% interval; its drug acceptance limits are not transferred to aquarium animals.
Check containment rather than only zero
Assume for this illustration that the reported interval method is appropriate for the parameter and test. The entire interval would have to lie inside the declared region for this numerical criterion to be met.
| Check | Fictional values | Finding |
|---|---|---|
| Point estimate | −2 points | Inside the illustrative region |
| Does interval include zero? | −8 to +4 | Yes; not the equivalence criterion |
| Lower interval limit versus lower margin | −8 versus −5 | Extends three points below margin |
| Upper interval limit versus upper margin | +4 versus +5 | Inside that boundary |
| Entire interval contained? | No | Equivalence not demonstrated |
| Biological acceptability of region | Not justified | No aquarium compatibility conclusion |
Keep inconclusive distinct from proven different
Failure to demonstrate equivalence is not proof that the true effect lies outside the region. The packet is insufficient for that conclusion too. Nor does including zero prove equality or absence of a meaningful effect. In a real paper, retain the prespecified margin rationale, interval method, endpoint and follow-up with the result. A margin chosen after inspecting the interval changes the question and cannot be described as predeclared. The useful output here is a completed containment decision with its assumptions, not a new manufacturer verdict or a recommended tolerance for adverse outcomes.
Use your own details
“No significant difference” does not establish aquarium compatibility worksheet
Download the editable text file, fill it in on your device and keep it with the relevant record. This blank worksheet records your information; it does not verify entries or approve treatment.
- Endpoint and comparison direction
- Observation horizon
- Predeclared equivalence region
- Scientific justification for both limits
- Interval level and equal-tail method
- Lower endpoint versus lower margin
- Upper endpoint versus upper margin
- Numerical finding versus biological acceptance
Put it into practice
Your checklist
- Recover the prespecified equivalence region and its justification.
- Check both interval limits against that region.
- Keep numerical containment separate from animal-acceptance evidence.
Sources and scope
This article distinguishes manufacturer statements from FishyMag’s own workflows and arithmetic examples. Linked formulation records retain their own evidence dates; publishing this article does not refresh those records.
- FDA, Statistical Approaches to Establishing Bioequivalence, May 2026, section II.C.1; reviewed November 9, 2026 ↗
- NIST handbook: confidence limits and their repeated-sampling meaning; reviewed November 9, 2026 ↗
AI-assisted writing and development. Not hands-on product testing or veterinary review. How we publish →
FishyMag reports what manufacturers say, quoted and linked. It does not diagnose illness or recommend a treatment, and it never infers that a product is safe from the absence of a warning — where no statement exists, it says so.
Formulations change. Always read the label on the bottle in your hand, and contact the manufacturer if your tank contains anything valuable or unusual.