The value of confidential enquiries.South Africa's first Saving Mothers report was published in 1999, and the first triennial report appeared in 2003. It showed that in 1999-2001 there were 2 445 maternal deaths, of which 1 462 were Direct. In 2002 - 2004 the total number of deaths rose to 3 296. In the next triennium it was 3 959, and now the fifth report, covering 2008 2010, includes 4 867 deaths, 2 252 of them Direct. The institutional maternal mortality rate (MMR) has increased to 176/100 000 births, and the pattern of causes in 2008 - 2010 was largely unchanged from the previous triennium. This is not good news.
In ancient times the reaction to bad news was to kill the messenger, and this instinct still survives. Today when a medical survey brings us unpalatable facts our first reaction is to question its accuracy. Then, if the facts cannot be disproved, we are often tempted to change the way we report them. Some countries, including the UK and many states in the USA, have either discontinued their confidential enquiries or tinkered with them in the hope that by changing the audit system they will speed up improvements in the health system. These efforts are well intended, but they are aimed at the wrong target.
Any report or scientific study can do nothing by itself. It is a tool to be used by those who have a specific end in view. When it comes to maternal mortality, we all share the same overall aim - to make pregnancy safer--but the task for each of us is to focus on what we can do as individuals. Every confidential enquiry report contains a
wide (perhaps too wide) range of recommendations, and all too often the reader nods in agreement with those directed at other people and skips past those that apply nearer home. Civil servants wait for action from clinicians, while clinicians expect leadership from politicians and managers. It is the job of the report to explain to each group what they need to do, to help them work together, and--perhaps hardest of all--to motivate them.
Back to basics
I have been involved with the UK Confidential Enquiry into Maternal Death (CEMD) for many years and more recently have worked with the World Health Organization (WHO) in countries with a wide range of MMRs and a variety of health systems. Helping to establish confidential enquiries in countries with such differing needs meant having to answer difficult questions from many, often sceptical, doctors, midwives and lay people. This process helped me clarify the basic principles that apply in all countries, but which up till then I had taken for granted. All those discussions had one thing in common: the more that people understood about confidential enquiries, the more convinced they became about their value. Perhaps it would be helpful to summarise those basic principles.
What is a confidential enquiry for?
The purpose of confidential enquiries is to save lives. They are not part of a blame game. They might be more popular if they were, as people like to identify scapegoats and punish them, feeling that this will somehow make things better. In reality 'naming and shaming' achieves little or nothing, but it is hard to change this attitude. Carers often blame themselves when a death occurs, and they become defensive. Confidentiality is essential if they are to give the enquiry accurate information and perhaps even suggest how another similar tragedy could be avoided.
Exactly how does a national enquiry save lives? The answer is that when you look at similar cases across a large area, themes emerge that are not obvious in local or hospital-based enquiries. You learn generalisable lessons about how life-threatening complications can be managed better, and sometimes you see how they could be prevented altogether. As well as teasing out lessons for individual professionals and teams, you also see how systems--local, regional and national--can be improved. You begin to recognise risk factors and early warning signs that become obvious only when you have reviewed a large number of cases. It is like gaining a lifetime's clinical experience in a few short weeks.
Once you have done this, the challenge for the national committee is how to get these lessons across to the right people. Publication of the report is the first essential step. The next step is to get people to read it. Some professional groups--particularly obstetricians, midwives and anaesthetists--do so as a matter of course, but in other medical specialties, such as emergency medicine or general practice, it is less of a priority. Politicians rarely have time to read it at all, though when they do the results can be dramatic.
Closing the loop
The value of confidential enquiries is well recognised by the WHO, whose 'Beyond the Numbers' project (1) emphasises that tackling rising MMRs needs expert analysis of cases as well as number-crunching. The process of analysis is usefully summarised by the WHO's diagram of the audit cycle--an endless round of five steps, repeated over a (usually) 3-year period. The steps are:
1. Identify cases
2. Collect information
3. Analyse the information
4. Formulate recommendations
5. Implement the recommendations.
The fifth step is arguably the most important, but each of them deserves some comment.
1. Identify cases
Confidential enquiries try to identify every single maternal death and generally achieve what public health experts call 'a high rate of ascertainment'. This has political disadvantages. In most countries the MMR is an underestimate because it is based on official death registrations, which may not take pregnancy into account. The immediate effect of introducing a confidential enquiry, therefore,
[TEXT NOT REPRODUCIBLE IN ASCII.]
is an apparent rise in the country's MMR. In the UK, the MMR according to the Registrar General is about half that according to the CEMD, with the result that the UK compares badly with the (often unfeasibly low) MMRs published by some European countries.
In other words, a confidential enquiry tells the truth about maternal deaths. Truth is rarely popular, but I believe denial is worse. In the long run, it is bad for politicians. Front-line staff become disillusioned and demotivated if they feel that problems are being ignored, or that official figures cannot be believed. Even when a country's MMR fails to improve, staff gain some reassurance from knowing that those in charge are acknowledging reality.
It is sometimes suggested that a confidential enquiry should be selective, limiting itself to the most urgent problems. This implies that we know what these problems are. There is a case to be made for selectivity, but not at the first step of the audit cycle. Even when an enquiry sets out to be comprehensive, deaths among disadvantaged women--the poor, the abused and migrants--tend to be the ones that are missed. In the UK we could have guessed that the MMR is higher among socially excluded women than among the professional classes, but it took the CEMD to show us just how huge the gap is. Without the CEMD we would also have been unaware that in the UK black women are at much higher risk than other ethnicities, or that among all women domestic violence is now responsible for more maternal deaths than obstetric haemorrhage.
2. Collect information
Part of the reason why a confidential enquiry is good at identifying deaths is that clinicians know about the enquiry and can contact it directly. Over time they learn that 'confidential' means what it says, though this message needs constant reinforcement in cultures where managers are wary of 'whistleblowers' or in regions (like the former USSR) with a tradition of punitive official enquiries. The growth of civil litigation (everywhere, especially in the USA) has also encouraged staff to be guarded in what they write, and this is becoming a real impediment to the efforts of confidential enquires to improve care.
Other methods of gathering information have fundamental flaws. Academic research projects are narrowly focused and in developing countries may be carried out by an overseas team whose prime concern is the research agenda. Local or hospital-based enquiries often lack objectivity: we have seen this in recent years in the UK, where the CEMD now receives the results of internal enquiries. Usually (though not always) they are deeply disappointing, being either needlessly self-critical or--more often--failing to recognise remediable factors that may have contributed to the death.
3. Analyse the results
Confidential enquiries are a professional self-audit performed by clinicians. This used to mean obstetricians only, but now includes midwives and all medical specialties involved in caring for sick pregnant women. All those who review the cases are in active practice and understand the difficulties faced by staff on the spot, as well as knowing what ought to happen in an ideal health system. The information they receive is anonymised, mainly to reassure those who have written the reports that confidentiality will be respected. Discussions among the various specialists are enlightening, as each learns about the challenges faced by his or her colleagues. The committee includes representatives from various types of practice and different regions, reflecting the fact that rural and inner-city circumstances are very different. For every member, reading the cases may provoke both sorrow and anger, but their resulting recommendations must be constructive.
4. Recommendations for action
The recommendations are directed towards all levels of the service, from individuals in community clinics to politicians in parliament. Many committee members, as well as their clinical work, have other roles such as directing a service or working with a national professional body. It helps if they know how things are done in the corridors of power. The original purpose the reports was to help clinicians learn from their colleagues' experience (sometimes through description of individual cases), but doctors can do only so much unless they work in a well-functioning service. Therefore some recommendations are aimed at managers, professional organisations and government ministries. There is an art in making all these recommendations--clinical and managerial--focused and achievable.
There is also an art in ensuring that they reach the appropriate people. After a report has become established the 3-year periods seem to pass very quickly, and doctors and politicians think they have heard it all before. The media can help to increase the impact of a report, even without the names and pictures of the dead women, and in the UK in the 1990s we tried shock tactics by renaming our reports 'Why Mothers Die'. This attracted some attention, but the effect was transient. The most effective way of drawing attention to recommendations is through personal contact with the right people.
5. Implement recommendations
Implementation, the most important of the five steps, is outside the control of the confidential enquiry, but this does not mean it must be left entirely in the hands of fate. The enquiry stands a better chance than most bodies of ensuring that its reports are not filed and forgotten. Highlighting specific cases can have a powerful emotional impact: as someone said in another context, 'One death is a tragedy: a hundred thousand deaths is a statistic.' The recommendations come from a group with professional credibility and are rarely challenged by clinical or academic colleagues. In fact, many scientific papers cite a confidential enquiry report prominently among their references, suggesting that at least the recommendations for research are being heeded, even though this may not have an immediate effect on the MMR.
I believe the key to having recommendations implemented is liaison with other bodies. There are several examples in the UK. The country has a network of Supervisors of Midwives, who work hard to ensure that the midwives in their area know about new recommendations. Our Obstetric Anaesthetists' Association, encouraged by anaesthetist members of the committee, pays meticulous attention to lessons in the report. The National Litigation Authority's criteria for grading hospitals' insurance premiums have been based heavily on CEMD recommendations.
[TEXT NOT REPRODUCIBLE IN ASCII.]
The links between the CEMD and the Royal College of Obstetricians and Gynaecologists (RCOG) are crucial. Since the 1990s the RCOG has published guidelines for clinical practice, and many of these take account of CEMD recommendations. Indeed, the latest RCOG guideline on thromboprophylaxis was called for by the CEMD, and the resulting dramatic fall in deaths from thrombo-embolism is a tribute to the co-operation between the two bodies--not to mention the staff who implemented the guideline.
The most difficult group to liaise with are politicians. Their tenure in each post tends to be relatively short, the health portfolio is a large one, and the time-lag between reform and results is long. Trust between doctors and politicians is often lacking, and their understanding of one another's problems is poor. And yet maternal mortality, perhaps more than most medical problems, requires action at a political level. It does not need any basic research, the investment required is relatively small, and the confidential enquiry is already there to monitor the results and give credit where it is due.
A final opinion
During my 20-odd years with the CEMD, the UK's leading cause of direct deaths, by some distance, was thrombo-embolism. Our reaction was to prioritise this issue and encourage the RCOG to issue guidelines based on the risk factors we had identified. This worked, though it took some time. (2) I feel that although confidential enquiries should remain comprehensive they sometimes need to call for a focused effort, and I believe that in South Africa the focus should be on those 2 252 Direct deaths. The leading cause is haemorrhage. We know how to treat that. Delivering the treatment is complex and challenging, but a fall in the leading cause of Direct deaths will be a boost to national morale and can surely be achieved. It would be the best demonstration of the value of confidential enquiries, and more importantly, it would save lives.
(1.) World Health Organization. www.who.int/making_pregnancy_safer/documents/9241591838 (accessed 30 March 2012).
(2.) Lewis G, ed. Saving Mother's Lives: Reviewing Maternal Deaths to Make Motherhood Safer: 2006-2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118 Suppl 1:1-203.
James Drife, MD, FRCOG, FRCPE, FRCSE, FCOG (SA), FSRH Emeritus Professor of Obstetrics and Gynaecology, University of Leeds, UK