A multiagency protocol for responding to sudden unexpected death in infancy: descriptive study

A working party set up by the Royal Colleges of Pathologists and of Paediatrics and Child Health has recommended introducing a national multiagency protocol for the management and investigation of sudden and unexpected deaths in infancy.¹

In 1999, a protocol embodying many of the features now recommended for the national protocol was introduced in Sussex, England.²

This report on how such a protocol works in practice and its findings have implications for the implementation of a national protocol.

Participants, methods, and results

I collected anonymised data from coroners' records on postmortem investigations and cause of death of all infants who were reported to have died suddenly and unexpectedly in Sussex (total population 1 500 000, 150 000 aged under 15) in the three years 2000-2.

I used postal questionnaire and semistructured interview to get information about the working of the protocol from professionals in the seven relevant disciplines and from parents (subject to their general practitioners' consent).

I identified 29 infants, aged 3 days to 8 months; adequate records for analysis were available on all but one. Eight of the 29 deaths were attributed to a specific natural cause, 16 to sudden and unexpected death in infancy or to sudden infant death syndrome, one to unintentional overlay (suffocation), and two to unnatural causes; two were classified as unascertained (table).

Table 1.

Compliance with protocol for responding to sudden unexpected death in infancy

Protocol requirement	Total cases	Cases complying
Baby transported to hospital (not direct to mortuary)	28	26
Pathology samples taken in emergency department	28	10
Joint home visit by paediatrician and police officer	28	1
Early interagency strategy discussion	28	28
Postmortem examination by paediatric pathologist (or general pathologist with a special interest)	28	27
Postmortem examination included:	28	27
Skeletal survey	28	25
Toxicology	28	10
Frozen section of liver (metabolic)	28	19
Cultures	28	24
Police attendance minimised	28	*
Diagnosis of SIDS/SUDI after early metabolic investigations	15	5
Diagnosis of SIDS/SUDI after abnormal toxicology excluded	15	6

SIDS=sudden infant death syndrome; SUDI=sudden and unexpected death in infancy.

^*Data incomplete, but no more than five officers present.

Implementation of the protocol varied considerably. The ambulance service had not implemented it. In accordance with the protocol, coroners or their officers sometimes refused permission for pathology samples to be taken immediately but could not always readily be contacted out of hours. Interagency discussions were held in all cases, but relevant professionals were not always invited to contribute. Joint home visits by police and paediatricians were generally not initiated. Most paediatricians had concerns about being available at short notice, and some were unwilling to visit the home. Despite the guidelines, some also expressed uncertainty about their role. The number of police involved tended to be disproportionate and some parental feedback on police involvement was negative. Although postmortem examinations were done in all but one case by pathologists with training, or a special interest, in paediatrics, the transport of bodies to a specialist centre for autopsy contributed to delays of three or more days in investigations in 13 cases. Delays can affect results.³ Diagnoses of sudden infant death syndrome sudden and unexpected deaths in infancy were made without consistent investigations.

Comment

The implementation of a multiagency protocol for managing sudden and unexpected deaths in childhood had serious deficiencies. These may have arisen from a lack of overall leadership and responsibility, failure to anticipate its implications, and non-compliance by some individuals. This occurred even though it had been drawn up in consultation with all the relevant disciplines

What is already known on this topic

A uniform system for the care and investigation of sudden unexpected death in infancy is recommended in a national protocol

What this study adds

Implementation of this multiagency protocol had serious deficiencies; strong and clear lines of responsibility within and between the agencies are needed and was approved by coroners, area child protection committees, and senior managers.

Although numbers were small, reflecting the rarity of these deaths, and though responses were incomplete, more than half the parents and professionals contacted responded, and coroners provided access to all relevant data. Parents provided valuable feedback on local practice; this will be useful in future evaluations.

Effective implementation of a national protocol will need strong lines of intra-agency and interagency accountability and may need statutory backing.