From: Validation of de-identified record linkage to ascertain hospital admissions in a cohort study
 | Linkage using Medicare details (V1) |
---|---|
Number of true CVD events (medical record review) | 101 |
Number of admissions identified by VAED using NEWID | 98 |
Number of CVD events correctly identified by VAED using NEWID | 94 |
Sensitivity | 94/101 = 93% |
Number of incorrect matches in VAED | 4 |
Non-matched rate | 4/101 = 4% |