OBJECTIVES: To assess the ascertainment of spina bifida and cleft lip/palate (CLP) in newborns and in fetuses from terminated pregnancies (ToPs) in the Swedish Birth Defects Registry (BDR) and to estimate the true incidences of these two anomalies. DESIGN: Retrospective register study. SETTING: Center for Epidemiology at the Swedish National Board of Health and Welfare, and Uppsala University Hospital. POPULATION: Newborns and fetuses from ToPs with spina bifida (1999-2004) and CLP (1999-2002) in Sweden. METHODS: Data from four registries/sources were used to estimate ascertainment in BDR and incidences of spina bifida and CLP. Main outcome measure: Ascertainment, under-ascertainment, and true incidence. RESULTS: For newborns, under-ascertainment of spina bifida and CLP were 6 and 13%, respectively, in BDR after record linkage with the Medical Birth Registry. Ascertainment of cleft palate increased when accompanied by cleft lip. The under-ascertainment of spina bifida in ToPs after 18 gestational weeks was 27%. Ascertainment of CLP in all ToPs and of spina bifida in ToPs before the 18th gestational week could not be estimated. The majority (109/155, 70%) of ToPs with spina bifida occurred before the 18th week. The estimated incidence of spina bifida per 10,000 births was 6.1 (2.4 newborns and 3.7 ToPs) and of CLP 20.1 (18.9 newborns and 1.2 ToPs). CONCLUSION: The ascertainments are relatively high for newborns in BDR, but lower or unknown for ToPs, which has an impact on the surveillance of spina bifida in view of the high proportion of ToPs.