Ascertainment of pregnancies terminated because of birth defects: effect on completeness of adding a new source of data

BACKGROUND: When evaluating preventive programs such as folate promotion and rubella vaccination, it is critically important to include terminations of pregnancy for neural tube defects and congenital rubella syndrome. Data from birth defects registries are often used for this purpose. The Western Australian Birth Defects Registry ascertains cases of birth defects in livebirths, stillbirths, and terminations of pregnancy for fetal abnormality, using multiple sources of ascertainment. METHODS: Data on terminations of pregnancy for fetal abnormality from the Western Australian Hospital Morbidity Data System 1980-1997 (not previously available to the Registry) were used to estimate the completeness of ascertainment of such cases by the Registry. Ascertainment-adjusted prevalences were calculated using capture-recapture methods. RESULTS: A total of 702 terminations with birth defects were identified among hospital discharges, most of which were already known to the Registry (87.9%). Of the 85 new cases, seven had a neural tube defect, 23 had a chromosomal defect, and 12 had confirmed maternal rubella infection during pregnancy. The ascertainment-adjusted prevalence was not importantly corrected] different for birth defects overall or for these individual conditions, although the 95% confidence intervals for all birth defects, and for all chromosomal defects, did not include the prevalence based on registered cases only. CONCLUSIONS: The Western Australian Birth Defects Registry ascertains a high proportion of pregnancies terminated for fetal abnormality, and should therefore be a reliable source of data with which to assist in monitoring the effectiveness of preventive programs.