PD-B02 · internal validation precision
Classification and audit outcome: internal validation precision; supported with reporting ambiguity.
Finding
Gervasoni’s Methods describes leave-one-out cross-validation, but Results do not identify separate apparent and cross-validated performance.
Current amendment or qualification
Replace “apparent” with “reported”; explain that leave-one-out cross-validation was described, while separation of estimates and nesting of PCA/selection were unspecified.
Integration note: Use reported performance, with LOOCV described. Do not certify that all PCA/selection steps were nested.
Remaining limitation: Use reported performance, with LOOCV described. Do not certify that all PCA/selection steps were nested.
Evidence and provenance
- Gervasoni E, Cattaneo D, Messina P, et al. Clinical and stabilometric measures predicting falls in Parkinson disease/parkinsonisms. Acta Neurologica Scandinavica. 2015;132:235–241. DOI 10.1111/ane.12388 · Original source
Affected reports
Original audit locator(s): balance1:P0116; balance1:P0146
Audit corrections describe the report only where that classification is explicit. A disagreement within a study, missing protocol, statistical query, access upgrade or optional addition is not silently converted into a report error or a new clinical threshold.
Audit register · Unresolved evidence queue
Linked knowledge pages
- COND-PD Parkinson’s disease
Located in this vault
Mapping status: exact. An exact location match is not an additional empirical verification.