Predictive value and completeness of the registration of congenital abnormalities in three Danish population-based registries

Aims: The predictive value and completeness of data on congenital abnormalities (CAs) collected in three administrative health registries in the County of North Jutland, Denmark were compared. Methods: The study included all singleton liveborn infants in the county during the period 1991-94 (n = 24,147). All infants recorded as having a CA in either the Medical Birth Registry (MBR), the Hospital Discharge Registry ( HDR), or the National Registry of Congenital Abnormalities (NRCA) were identified, and the recordings in each registry were compared. Infants recorded in at least two registries were considered correctly diagnosed with a CA for the sake of the analyses. The predictive value was defined as the number of infants correctly diagnosed with a CA in the registry divided by the total number of infants recorded with a CA in the registry. In all cases with recording in one registry only, the predictive value of the registration with CA diagnosis was assessed through a review of a sample of medical records. The completeness was defined as the number of correctly diagnosed CAs in the registry divided by the total number of identified CAs. Results: The predictive value and completeness were calculated as 89.1% (85.3-92.8) and 32.3% (28.9-35.7) in the MBR; 88.2% (85.9-90.5) and 89.9% (87.7-92.1) in the HDR; and 99.6% (98.9-100.0) and 36.0% (32.5-39.5) in the NRCA. Conclusions: The HDR seems to have a predictive value and completeness that are acceptable for general surveillance and epidemiological research regarding CAs. The NRCA may be suitable for case-control studies owing to a high predictive value.