Letters

Home versus hospital delivery

EDITOR,--R S Settatree's analysis purporting to show that delivery in hospital is safer than delivery at home is flawed.¹ The 388 deaths among normally formed infants weighing 2500 g or more on which this analysis was based were the subject of confidential inquiries in 1993. Such inquiries into stillbirths and deaths in infancy are initiated when relevant deaths have been identified through a voluntary rapid reporting system. In 1993 there was no case by case cross checking between statutory registration of deaths and the rapid reporting system, but a comparison with the total numbers of deaths indicated considerable underreporting to the rapid reporting system.² Furthermore, the annual report for 1993 noted that the results of 45 confidential inquiries into deaths in this category were submitted too late to be included in the analyses.³

Not only were the numerators of the ratios quoted for home and hospital deliveries incomplete but the adjustments made to the denominators to take account of unplanned births at home are also questionable. It was suggested that estimates of the proportion of births at home that are unplanned range from 10% to 60%, although references were not cited to support this. Settatree's calculations shown are based on an estimate of 26%, which is almost certainly too low. A national survey of all births at home in 1979 found that a third were unplanned.⁴ A more recent survey of births in the former Northern Regional Health Authority found that 40% of births at home had not been planned to occur there.⁵

There is a further error in the calculations in that the numerator data include deaths in England, Wales, and Northern Ireland while the denominator data are for only England and Wales.

In discussing intrapartum deaths at home the report of the inquiry for 1993 stated, "A consistent theme with the small number of home deliveries which resulted in the death of the baby was insufficient surveillance and monitoring of the labour, and a lack of experience in the community management of problems such as shoulder dystocia and breech presentation. These points were equally evident in the cases of hospital delivery."³ Its authors also strove to examine the deaths "without drawing potentially erroneous conclusions from the limited sample which incorporated both planned and unplanned home deliveries."² Other commentators would be wise to display similar caution.

Lecturer, health services research Department of Social Medicine, University of Bristol, Bristol BS8 2PR

Rona Campbell 

1. Dowswell T, Thornton JG, Hewison J, Lilford RJL, Raisler J, Macfarlane A, et al. Should there be a trial of home versus hospital delivery in the United Kingdom? BMJ 1996;312:753-7. (23 March.)
2. Confidential Inquiry into Stillbirths and Deaths in Infancy. Annual report for 1 January-31 December 1993. Part II. London: HMSO, 1995.
3. Confidential Inquiry into Stillbirths and Deaths in Infancy. Annual report for 1 January-31 December 1993. Part I. London: HMSO, 1995.
4. Campbell R, Macdonald Davies I, Macfarlane AJ, Beral V. Home births in England and Wales; perinatal mortality according to intended place of delivery. BMJ 1984;289:721-4.
5. Northern and Yorkshire Regional Health Authority. Report of the Northern regional home births survey 1993. Newcastle upon Tyne: Regional Maternity Survey, 1994.