Objective: To ascertain whether there is variation in obstetric practice within a defined subgroup of primiparous women.
Design: Analysis of routinely collected data.
Subjects: Ten thousand two hundred and ninety-five primiparous deliveries defined as 'normal' in Scotland during 1990 and 44,820 primiparous deliveries defined as 'normal' in England between April 1990 and April 1991.
Results: Little variation was found in the distribution of mothers' ages and gestational ages at delivery, and babies' birthweights. In both England and Scotland there was considerable variation between regions in instrumental delivery rates and caesarean section rates. There were many deficiencies in the quality of the data provided by the English Maternity Hospital Episode System.
Conclusions: There is regional variation in instrumental delivery rates and caesarean section rates in England and Scotland. The poor quality of data for England makes interpretation of the cause of variation difficult because the extent to which variation may reflect deficiencies in the data, rather than differences in practice, is unknown. Improvements need to be made in Maternity Hospital Episode system data, increasing both coverage and data quality. Nevertheless, similar variations in instrumental delivery and caesarean section rates may be associated with differences in population characteristics not measured in these data sets of differences in obstetric practice.