THE HISTORY OF CONFIDENTIAL ENQUIRIES

The current system of confidential enquiries started in 1952, just four years after the inception of the NHS. Before that, maternal deaths were reported to the Ministry of Health on an ad hoc basis.¹ The first report of CEMD covered 1952–54, capturing 77% of maternal deaths during that period. From 1985–87 onwards, a single report was published triennially for the whole of the United Kingdom.² The purposes of the CEMD were to assess the main causes of maternal deaths and, through the identification of avoidable causes, to reduce maternal morbidity and mortality by recommending improvements in clinical care and service provision; it also indicated directions for future research and audit.

CESDI was established in 1992, after the Department of Health directed that the 14 regions of England should undertake perinatal mortality surveys. CESDI’s remit was to improve understanding of the causes of death in late fetal life and infancy—that is, from 20 weeks post-conception to one year after birth. Its aim was to reduce mortality by identifying suboptimal patterns of practice and service provision related to those deaths and to make recommendations for improvement. Combining CEMD and CESDI, CEMACH retains this regional organisation and these overall aims.

There are two other national confidential enquiries: the National Confidential Enquiry into Perioperative Deaths (NCEPOD) and the Confidential Enquiry into Suicides and Homicides (CISH). All the confidential enquiries now fall under the umbrella of the National Institute of Clinical Excellence (NICE).

All confidential enquiry reports describe the conclusions of audits of their work by those involved in that field. The reports have been authoritative and have influenced clinical practice. Their frequency of publication is determined by the rate of occurrence of the events described. In the United Kingdom there are an estimated 957 500 pregnancies a year* and between 300 and 400 maternal deaths a year (giving a maternal mortality rate of 11.4 deaths per 100 000 maternities over the most recent period reported, 1997–99; see box 1):³ reports on maternal deaths are published every three years. The scale of the work undertaken by CESDI was quite different. About 10 000 deaths occur annually between 20 weeks gestation and 1 year of age in England, Wales, and Northern Ireland (644 940 live births and stillbirths in 1999; perinatal mortality rate 7.9 per 1000 live births) and reports are published annually. It is of course not possible to have an enquiry about all these deaths and CESDI has always had a rolling programme, identifying specific criteria for detailed confidential enquiry. Table 1 summarises some of CESDI’s previous work programmes.

Confidential enquiries are driven by a desire to improve care and a great many people are involved.² CEMACH has a central office with a permanent staff of six and there are managers in each of the regions of England and Wales, who are responsible for individual enquiries. Each enquiry panel for a death that fulfils pre-set criteria comprises clinicians from relevant specialties, who are independent of the hospital where the patient died and who are unaware of the clinicians concerned with the patient’s care—one of the regional managers’ tasks is to ensure that the case notes are anonymised. The results of these enquiries are then collated.

CESDI’s early projects often lacked controls and denominator information, which limited their interpretation and generalisability.³ However, the study of Sudden Unexpected Deaths in Infants,^4,5 the Antepartum Term Stillbirth Study,⁶ and the study on babies at 27 and 28 weeks gestation⁶ have included information about controls. Other CESDI projects have used a focus group methodology (for example, reports on shoulder dystocia,⁶ ruptured uterus,⁶ planned home delivery,⁶ anaesthetic complications and delays,⁸ breech presentation,⁸ and the onset of labour (all undertaken in 1994–95) and stillbirths (undertaken in 1996–97)⁹).

SOME FINDINGS OF CEMACH’S PREDECESSORS

The CEMD reports showed a fall in maternal deaths due to abortion from 153 in 1952–54 to one in 1994–99, attributable to the legalisation of abortion. Between the same periods, maternal deaths from haemorrhage fell from 188 to nine because of oxytocic injections, ultrasound diagnosis of placenta praevia, and improved intensive care. Death due to thromboembolism, which remains an important cause of maternal mortality, fell from 138 to 46 over this same 40 year period.¹⁰ After 1993, there was improved case ascertainment through linking data with that provided by the Office for National Statistics. Table 2 summarises major causes of maternal deaths and their rates.¹¹

In 1994–95, 52% of 873 deaths related to intrapartum events were found to have received suboptimal care and the intrapartum related mortality rate was 0.95 per 1000 live and stillbirths;¹² by 1999, the intrapartum related mortality rate had decreased to 0.62.⁹ There is, however, no room for complacency. The 1999 CESDI report warned about a failure to recognise that fetal growth failure was a precursor of many stillbirths. It was necessary to repeat recommendations that had been made in 1997 and 1998: there is a need to improve the quality of maternity records to enable clear identification of risk factors and appropriate management plans during the ante- and intra-partum periods.⁹

Several CESDI reports have commented on the use of fetal monitoring. The 1997 CESDI report showed a failure of use and/or interpretation of the cardiotocograph (CTG) in more than half of deaths that occurred intrapartum.¹² A focus group on the ruptured uterus reported that 26% of comments on substandard care related to fetal monitoring.¹³ The following year, a “4 kg and over” survey also found deficiencies in the interpretation of CTGs.¹⁴ CESDI consequently recommended that every hospital offering intrapartum care should train and update staff regularly in the use of CTGs, and standards were developed:^15–17

• Basic provision of electronic fetal monitoring (EFM) facilities should be 2–4 machines per 1000 deliveries.

• A guideline on the use of EFM should be available in every unit.

• Continuous EFM should be used in selected high risk pregnancies.

• If EFM is used, fetal blood sampling should be available.

• In situations of suspected fetal compromise, umbilical cord pH should be measured at delivery.

One of CESDI’s last projects was “Project 27/28—An enquiry into the quality of care and its effect on the survival of babies born at 27–28 weeks”.⁷ This major piece of work aimed to identify patterns of practice and service provision that were considered to have contributed to the deaths of babies born at 27 and 28 weeks gestation born between 1 September 1998 and 31 August 2000. It collected denominator data, set out standards, and used a case-control study approach to make recommendations for the better care of these very vulnerable infants. During the study period the neonatal mortality rate in this group was 12%. The Project made 16 recommendations for standards at a national and commissioning level and 64 at Trust level. These included recommendations about:

• Avoiding super-ovulation and multiple pregnancies when assisted reproductive techniques are used.

• Ensuring that units have a system to identify mothers who are at risk of preterm delivery at booking and during antenatal care and that these mothers should receive appropriate consultant care.

• A specialist high risk team to manage the labour and delivery of the baby of 28 weeks and less.

• Appropriate guidelines, which should include communication issues between professionals, the management of infection in the mother and baby, and thermal, respiratory, and cardiovascular support of the baby.

• Managed clinical networks and appropriate staffing levels.

Both the CEMD and CESDI have drawn attention to public health messages. In 1996, CESDI promoted key messages from the SUDI (Sudden Unexpected Deaths in Infancy) studies (box 2).¹⁹ In 1998, the CEMD gave clear advice on the correct use of seat belts by pregnant women (box 3).¹⁰

A review of trends in reproductive epidemiology by the CEMD has shown a steady increase in the proportion of deliveries by caesarean section: 3% in the 1950s, 10% in the early 1980s, 15% in 1994–95, and over 18% in 1997–99.¹⁸ This increase has implications both for the health of mothers and for the healthy survival of their children. The most recent CEMD publication highlighted 20-fold increased mortality rate among the most disadvantaged.³ These issues, which cut across the interests of different groups of patients, are areas that CEMACH will be able to address.

THE FUTURE

The CEMD and CESDI were previously managed by the Department of Health. In 1996, the management of CESDI was taken over by a consortium of royal colleges. Now NICE provides most of the funding. However, the profession still runs CEMACH and its management board comprises members nominated by the Royal Colleges of Paediatrics and Child Health (RCPCH), Obstetrics and Gynaecology (RCOG), Pathology (RCPath), Midwives (RCM), Anaesthetics (RCA), and the Faculty of Public Health.

CEMACH will continue to review the causes of deaths related to pregnancy and those occurring during infancy, but will also include children up to the age of 16 years. Furthermore, as the change in title implies, the new enquiry will be concerned with morbidity as well as mortality.

There will be three sections, each headed by a National Advisory Group, which will advise on the content of the programme of work. One section will continue the work of the CEMD, considering the problems of pregnancy. The next edition of the triennial report, covering the period 2000–02 is being prepared and data are being collated for 2003–05 report. The last CEMD report identified suicide as a leading cause of maternal death—there were 28 such deaths, compared with 35 due to thrombosis and thromboembolism and 15 due to pregnancy induced hypertension. This is the sort of issue where work undertaken by the new Confidential Enquiry might be linked with that of an existing one, CISH.

The second section of CEMACH will take on the work of CESDI: the present programme considers the problems of the diabetic mother and her infant, using a case-control approach. Diabetes is the most common pre-existing medical disorder complicating pregnancy in the United Kingdom, affecting approximately four per 1000 pregnancies. The St Vincent Declaration²⁰ stated that every diabetic pregnancy should have a near normal outcome, but in the UK, where the background perinatal mortality rate is 7.9/1000, the perinatal mortality rate of pregnancies complicated by maternal diabetes varies between 36.1 and 42.8 per 1000.^21–23 A survey by CEMACH has found that facilities available to pregnant diabetic women appear to be variable throughout the UK. CEMACH has set standards, and appointed panel chairs; the confidential enquiries are about to start.

The third of CEMACH’s sections will consider the problems of childhood. This latter task is potentially broad and challenging. The precise area has not yet been decided, but it is likely to cross professional boundaries, perhaps involving social services and education. The methodology will continue to be based on confidential enquiries around individual cases and it will be case-controlled. Whatever the precise problems of childhood that are to be considered, CEMACH is determined to make a difference to the lives of children. Many of the CESDI and CEMD reports are available on the CEMACH website (www.cemach.org.uk).