Reproducible output comparison
==============================

StatMedAnalysis | Original downloadable worksheet | 6 November 2026

Distinguish meaningful analytical changes from harmless output differences.

How to use this worksheet
Define which outputs must match exactly and which permit numerical tolerance. Separate
timestamps, file ordering and presentation metadata from scientific results. Compare
structured values before visual summaries where possible. Record every tolerated
difference and its justification.

Save a working copy and keep the original blank version. Use public, hypothetical or
aggregate information only; do not enter patient identifiers, credentials or confidential
records. Keep source evidence in your approved project workspace. Mark unresolved text
fields UNKNOWN or NOT REPORTED with a reason; never substitute zero for an unknown count.

Project/version __________________  Prepared by role/date __________________

Working fields
1. Output artifact and revision pair
Response/evidence: _________________________________________________________
___________________________________________________________________________

2. Exact-match fields
Response/evidence: _________________________________________________________
___________________________________________________________________________

3. Numerical tolerance fields
Response/evidence: _________________________________________________________
___________________________________________________________________________

4. Nonsemantic metadata exclusions
Response/evidence: _________________________________________________________
___________________________________________________________________________

5. Structured comparison result
Response/evidence: _________________________________________________________
___________________________________________________________________________

6. Visual comparison result
Response/evidence: _________________________________________________________
___________________________________________________________________________

7. Unexpected scientific difference
Response/evidence: _________________________________________________________
___________________________________________________________________________

8. Resolution and approval scope
Response/evidence: _________________________________________________________
___________________________________________________________________________

Hypothetical worked case
Two fictional reports differ only in generation timestamps, while all numerical cells
match. Ignoring the timestamp is reasonable if documented; ignoring changed denominators
or confidence limits would not be.

Decision rule and limitations
Do not normalize away differences that can change interpretation. A passing comparison
must state exactly what was compared and which differences were intentionally excluded.

Decision _______________________  Remaining uncertainty ____________________
Next action ____________________  Owner / review date ______________________
Independent check and discrepancy resolution _______________________________

Official method or reporting reference checked 6 November 2026
https://docs.python.org/3/library/math.html#math.isclose

This is an original planning aid, not an official checklist, validated instrument,
clinical recommendation or certificate of expert approval. Apply the referenced
guidance to the specific design and record any additional requirements.