Use of local neural tube defect registers to interpret national trends.

To conduct a number of studies into the prevalence of neural tube defects (NTD) in the area covered by the Oxford Record Linkage Study (ORLS), multiple sources were used to build a local register of cases occurring in Oxfordshire and West Berkshire between 1968-1990. One source of potential cases--namely, termination and congenital malformation monitoring data available for the locality from the Office of Population Censuses and Surveys (OPCS) data--were kept separate. Comparison of the local cases recorded by OPCS and those known to the register from 1974-1990, using the method of capture-recapture, suggested that national data are only about two thirds complete, but that this underreporting is likely to be reasonably consistent from year to year. OPCS data can therefore be used to study NTD trends if not absolute risks. The local register seemed, by the same yardstick, to be very complete and is being used in a variety of studies of the occurrence of NTD. Survival to one year in this area, over the period 1968-1990, has only improved in the recent past, if at all. Most NTD pregnancies now end in termination rather than birth, and there has been a true decline in the occurrence of NTDs, and likewise the different subtypes.