Acute antenatal transfer to specialist centres is an accepted practice but few or no regular data are collected regarding the numbers of transfers performed or subsequent pregnancy outcome. We wished to determine the numbers, and the maternal and fetal outcomes following acute antenatal transfer between consultant obstetric units in the former Northern Region of the UK over a 12-month period (1 January-31 December 99). This is a geographically defined population in terms of provision of perinatal services. All acute antenatal transfers were notified centrally. Data pertaining to each transfer were collected at the time of transfer. Subsequent maternal and fetal outcomes were determined from patient records and neonatal databases. The regional annual acute antenatal transfer rate was 3.7 per 1000 deliveries. Most were for fetal reasons, although transfer rates varied between hospitals. The decision to transfer was influenced by distance and availability of paediatric staff. Even units that have similar characteristics show considerable variation in their transfer rates. No adverse incidents occurred during transfer and no major changes in maternal management occurred following transfer. Twenty-four per cent of women remained undelivered following transfer. Women with preterm labour in the absence of ruptured membranes were less likely to deliver than those transferred for other reasons and if they did deliver, their infants were also less likely to need intensive care. We believe audit of acute antenatal transfers should be routinely undertaken. Numbers of transfers might be reduced if delivery and the need for neonatal intensive care could be predicted with greater accuracy. The psychological and financial costs of transfer to women and healthcare providers need to be addressed.