Feature

A question of delivery

Caesarean section rates have been rising for years – but nobody is sure why. Alison Macfarlane looks at the figures

The steep rise in caesarean section rates in England during the 1990s, following an apparent levelling off in the 1980s (figure 1), prompted the National Sentinel Caesarean Section Audit.¹ While caesarean rates for Nordic countries remained at about the 10-15 per cent recommended by the World Health Organisation, rates for England and Scotland continued to rise to North American levels, reaching 20 per cent by 2000.^{2 3} No national statistics are available on trends in Wales or Northern Ireland.

The audit included a literature review and a massive two-phase data collection exercise. In the first phase the study team, based at the Royal College of Obstetricians and Gynaecologists, asked maternity units to send them data on all births between May and July 2000 in England and Wales, and between December 2000 and February 2001 in Northern Ireland, the Channel Islands and the Isle of Man. Scotland, which did its own audit in 1994/95, did not take part this time.⁴

Additional information was collected about all births by caesarean section and each delivery suite was asked to keep a diary of its activity over two weeks. Because of the poor quality of computerised information systems in many maternity units — and their complete absence in about a third – the diary had to be kept on paper. Each unit was also asked about labour ward staffing, activity and anaesthesia.

In the second phase, 40 maternity units were selected at random and questionnaires asking about attitudes to caesarean section were sent to women delivering there and to consultant obstetricians.

The report of the audit published at the end of October 2001 contained analyses by country and region only. No data were published for individual maternity units — despite the fact that the Department of Health routinely publishes caesarean section rates for individual trusts. As figure 2 shows, there was considerable variation between caesarean rates for countries and the non-standard set of English regions used. There was also variation in the use of forceps and the extent to which these are being superseded by the ventouse or vacuum extractor. Units were asked to use their own local definitions to distinguish between elective (planned) and emergency caesareans.

There is considerable disagreement about how to make this distinction so data were also collected, using a new definition, about the ‘grade of urgency’:

1: immediate threat to life of woman or fetus;

2: maternal or fetal compromise;

3: no maternal or fetal compromise but needs early delivery;

4: delivery times to suit woman and staff.

As figure 3 shows, this attempt to use a more consistent definition showed considerable regional variation in caesarean rates.

Among the factors which prompted the audit was the view of many obstetricians that the rising caesarean rate is fuelled by requests for elective caesareans from women who have no clinical need for one. The current emphasis on consumer choice in the NHS has led to much debate over how clinicians should respond to this – but the audit suggests that such requests may not play a major role. Only seven per cent of caesareans were reported by clinicians as undertaken at the woman’s request, and these included caesareans for which there were also clinical indications. The survey of obstetricians suggested that only three per cent of caesareans were requested by women in the absence of medical justification, and only about half of such requests were granted.

The authors are sparing in their overall conclusions but, given the volume of data collected, it is not surprising that further analyses are under way.

What impact will the audit have? We may not know unless maternity information systems are improved. Otherwise there will be inadequate data to monitor caesarean section and related trends routinely. In Scotland, which has good national statistics, the caesarean section rate has continued to rise since the mid-1990s when its audit was done. Clearly data and guidelines are not enough.