External performance and precision
Estimate validation metrics and uncertainty
Assess calibration across the risk range, discrimination and relevant utility. Case-mix changes can affect performance summaries. For time-to-event outcomes, align the prediction horizon and use methods that address censoring.
Flow diagram
Steps
Define evaluation metrics
Prespecify calibration, discrimination and overall error for the outcome.
Plan uncertainty intervals
Choose methods respecting clustering, censoring and sample structure.
Evaluate intended subgroups
Assess relevant settings and groups with precision limitations.
Interpret case-mix differences
Distinguish changes in population from changes in predictor-outcome relationships.
Preserve original-model results
Report untouched performance before any updating comparison.
External performance and precision
End flow
Tables and supporting information
Validation precision plan
Plan the required information and interval methods; no performance values are prefilled.
Assumptions and rationale
Performance estimates need uncertainty and intended-use context; no single metric establishes transportability.
Method and protocol evidence
Attach the protocol, methodological sources and versioned software specification supporting these decisions. Record citations and provenance; do not enter participant identifiers.