Intended for healthcare professionals


Investigating infant deaths

BMJ 2005; 330 doi: (Published 27 January 2005) Cite this as: BMJ 2005;330:206
  1. M P Ward Platt, clinical director (m.p.ward-platt{at}
  1. Regional Maternity Survey Office, Newcastle upon Tyne NE2 4AA

The protocol suggested by the Kennedy report is good, but will it work?

There is now another “Kennedy report” for paediatricians to consider.1 This time the chair was Baroness Helena Kennedy QC, the working group was set up by the royal colleges of Paediatrics and Child Health and of Pathology, and the subject was the investigation of sudden unexpected deaths in infants (SUDI). The report recommends a systematic and evidence based protocol for the history, examination, investigation, autopsy, death scene investigation, and subsequent multiprofessional meeting in relation to each death.1 It also recommends that this should be compulsory, although it doesn't say how that might be enforced. But will it have the desired effect?

The background is several recent high profile cases in the United Kingdom of mothers accused of killing their infants: the quashing of the convictions of Sally Clarke and Angela Cannings; the acquittal of Trupti Patel; and cases such as that of Maxine Robinson, who originally protested her innocence of the deaths of the two children she was convicted of murdering but who this year admitted their murders, together with the murder of her first child.2 The death of this infant, who died aged 9 months, was originally labelled as a cot death. These cases highlight the widespread problem of the inadequate investigation of infant deaths. Improvements should work both ways: a greater chance of avoiding criminal proceedings for innocent parents (the majority) but also a higher chance of identifying homicide. Whether the criminal justice system is an appropriate place to deal with infant murder is debatable, but the need for a more coherent and evidence based approach to investigating infant deaths is hard to dispute.

Much of the UK evidence base for an improved and comprehensive approach has come from the large scale case-control study of sudden unexpected deaths in infancy conducted by the Confidential Enquiry into Stillbirths and Deaths in Infancy (CESDI) in the early 1990s.3 Helena Kennedy's recommendations bear a close resemblance to those in the CESDI-SUDI report and are explicitly based on the practices that are routine in Bristol and the south west region.

So will the protocol recommended by Baroness Kennedy make a difference? Here we are on uncertain ground. The mere existence of a protocol is not a guarantee that it will be followed, however much it might be “compulsory”; the history of protocolology is one of worthy aspirations that largely fail to change practice in the real, messy world.4 5 This issue includes data from Sussex that indicate that the messy world has once again triumphed (p 227) 6. Even in the management of sudden unexpected deaths in infants, trying to change practice needs a whole lot more than a protocol, however much the protocol has been agreed among the different agencies and disciplines.

Paediatricians will view some of the recommendations as aspirational rather than immediately practical. How many paediatricians will feel able to commit to making home visits, which may last several hours, at very short notice to bereaved families, or participate in death scene investigations? Given the potential for child protection issues to arise, and the current reluctance of paediatricians to take on child protection work, is it realistic to expect paediatricians in general to be enthusiastic about this task? Taking a gold standard, research supported system in a part of the country served by internationally renowned enthusiasts and trying to replicate it in hectic paediatric services elsewhere may prove overoptimistic.

Yet much of the protocol is not about home visits or the detailed role of paediatricians. It is about the comprehensive gathering of information on first contact with health services, and the systematic recording of the data. It is about ensuring that pathologists and coroners recognise that making a positive diagnosis and using appropriate investigations and histology is at least as important as making a statement about the likelihood of foul play on macroscopic examination alone. It is about convening a multidisciplinary meeting to pool all the information and determine the medical view of the cause of death. We should not lose sight of the potential of this protocol for improving the investigation of unexpected infant deaths by getting too focused on the role of paediatricians in the immediate aftermath of the death.

What might drive improvements in the management of sudden unexpected infant deaths if protocols by themselves are not enough? Audit could be a powerful tool, especially if the anonymised and aggregated results were published. Just as the SUDI study was performed under the auspices of CESDI, perhaps its successor, the confidential enquiry into maternal and child health (CEMACH), should undertake this function in the future. Confidential inquiry would be an appropriate method to examine the processes by which each death was investigated, rating actual practice against the standards set out in Baroness Kennedy's recommendations. Feedback could be relatively swift; could embrace coroners, police, social services, and health services; and could operate within the areas served by each confidential inquiry office. Under these circumstances, each death would yield learning points, could engage paediatricians, and should contribute to our evolving understanding of sudden infant deaths.

Papers p 227


  • Competing interests The author is a member of the CESDI-SUDI study group.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
View Abstract

Log in

Log in through your institution


* For online subscription