Model recalibration decision plan ================================= StatMedAnalysis | Original downloadable worksheet | 6 November 2026 Define how a prediction model may be updated for a new setting. How to use this worksheet Evaluate original predictions before updating the model. Separate overall risk differences from changes in calibration slope or predictor effects. Choose the least complex update justified by data and intended use. Reserve independent evidence or appropriate resampling to evaluate the updating procedure. 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. Original model and target setting Response/evidence: _________________________________________________________ ___________________________________________________________________________ 2. Unchanged-model calibration evidence Response/evidence: _________________________________________________________ ___________________________________________________________________________ 3. Candidate update: intercept/slope/refit Response/evidence: _________________________________________________________ ___________________________________________________________________________ 4. Update sample and information Response/evidence: _________________________________________________________ ___________________________________________________________________________ 5. Predictor measurement compatibility Response/evidence: _________________________________________________________ ___________________________________________________________________________ 6. Validation of updating procedure Response/evidence: _________________________________________________________ ___________________________________________________________________________ 7. Comparison with original predictions Response/evidence: _________________________________________________________ ___________________________________________________________________________ 8. Deployment and monitoring decision Response/evidence: _________________________________________________________ ___________________________________________________________________________ Hypothetical worked case A fictional model systematically predicts higher probabilities than observed in a new setting. Updating an intercept may address overall calibration, but the same update data cannot provide an unbiased assessment of its new performance. Decision rule and limitations Do not present recalibrated in-sample performance as independent validation. Record exactly what was changed and whether ranking, calibration and clinical utility were re- evaluated. Decision _______________________ Remaining uncertainty ____________________ Next action ____________________ Owner / review date ______________________ Independent check and discrepancy resolution _______________________________ Official method or reporting reference checked 6 November 2026 https://scikit-learn.org/stable/modules/calibration.html 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.