Prediction error subgroup audit =============================== StatMedAnalysis | Original downloadable worksheet | 6 November 2026 Investigate whether predictive performance differs across relevant groups. How to use this worksheet Define groups and the reason they matter before inspecting results. Use the same outcome horizon and metric definitions across groups. Report denominators, outcome frequencies and uncertainty alongside point estimates. Consider measurement and selection differences before attributing a disparity to the model alone. 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. Group definitions and rationale Response/evidence: _________________________________________________________ ___________________________________________________________________________ 2. Sample and event counts Response/evidence: _________________________________________________________ ___________________________________________________________________________ 3. Outcome ascertainment comparability Response/evidence: _________________________________________________________ ___________________________________________________________________________ 4. Metric and threshold definition Response/evidence: _________________________________________________________ ___________________________________________________________________________ 5. Performance estimates with uncertainty Response/evidence: _________________________________________________________ ___________________________________________________________________________ 6. Intersectional or sparse-group limits Response/evidence: _________________________________________________________ ___________________________________________________________________________ 7. Potential mechanisms and follow-up Response/evidence: _________________________________________________________ ___________________________________________________________________________ 8. Scope of fairness or safety claim Response/evidence: _________________________________________________________ ___________________________________________________________________________ Hypothetical worked case A fictional subgroup has three outcome events and a high observed sensitivity. The estimate is too imprecise to establish equal performance; pooling away the group would also conceal the uncertainty. Decision rule and limitations Avoid declaring parity or inequity from point estimates alone. Specify the metric, population and uncertainty, and plan additional evidence where groups are sparse. 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/model_selection.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.