Abstract
A sample of 1,000 women who gave birth at term in a London teaching hospital were investigated to establish, from the records, the pattern of attendance at ante-natal clinics and classes. The variables associated with good attendance confirm and extend previous research and include parity, age, ethnicity, marital status, occupational status and tobacco habit. The relationship between good attendance, medical interventions and other outcomes was found to be less marked than in previous studies. Statistical analysis, using ‘logit’, to control for linkage between variables, suggests that the only marked association with outcomes is that frequent class attenders are much more likely to breast feed. The implications are discussed in terms of a definition of natural childbirth.