Food allergy and infantile colitis.

it, the improvement in the survival rates from 1976 to 1983 is not so striking (Fig. 1). Indeed the overall survival of babies weighing 1251 to 1500 g from 1976 to 1983 (68-5%) would not have been significantly greater than it was from 1970 to 1975 (60-0%). The contribution of mechanical ventilation to the survival rates of babies is summarised in Fig. 2. We agree with Barson et al that good quality neonatal intensive care influences obstetrical management and leads to elective preterm delivery of high risk fetuses, thereby reducing the risk of stillbirth from asphyxia. Indeed, we submitted that idea as written and oral evidence to the House of Commons Social Services Committee in 1980.3 There are great pitfalls in using information from perinatal necropsy surveys as a basis for auditing neonatal care. Audit requires a mutlidisciplinary approach including the