Article Text


How reliable are SIDS rates? The importance of a standardised, multiprofessional approach to “diagnosis”
  1. P J Fleming,
  2. P S Blair
  1. Institute of Child Life and Health, University of Bristol, UK
  1. Correspondence to:
    Prof. P Fleming
    Institute of Child Life and Health, UBHT Education Centre, Upper Maudlin St, Bristol BS2 8AE, UK;

Statistics from

Commentary on the paper by Sheehan et al (see page1082)

As a “diagnosis”, sudden infant death syndrome (SIDS) is unique in that the definition is reached by exclusion: by failing to show an adequate cause of death. The “diagnosis” of SIDS is inevitably subjective and imprecise as it depends on the knowledge, skill, and thoroughness of the professionals who obtain the clinical history, conduct the evaluation of the circumstances of the death, and perform the postmortem examination, as well as the synthesis and interpretation of this information.

Even if each of these component parts of the investigation have been conducted to a specified standard, the collection of this information is not the end point but should be the starting point for a multidisciplinary review process in which the professionals involved should jointly examine the evidence, highlight concerns, and decide on the cause of death. This multidisciplinary approach has been used effectively in the field of SIDS research,1 has been shown to work in clinical practice,2 and has recently been recommended by the Kennedy Report3 as the routine national approach in the UK. Once the postmortem results are available, usually 8–12 weeks after death, the paediatrician, pathologist, GP, health visitor, senior investigating police officer, and where appropriate, social worker, should meet to discuss the case.

Sheehan and her colleagues4 have unambiguously shown how published SIDS rates become a mockery if a clear distinction is not made between contributory factors and what constitutes a sufficient explanation of death. Unfortunately the hypothetical exercise they show here has already become a reality in England and Wales. From 1996 to 2003, Office for National Statistic (ONS) figures suggest the rate for unascertained deaths has risen more than fivefold to 0.2 deaths per 1000 live births in a period when the SIDS rate has dropped by 0.35 deaths per 1000 live births. The potential of a diagnostic transfer has not been ruled out by the ONS.5,6 A recent survey of over 60 UK pathologists currently conducting SIDS postmortem examinations by Limerick and Bacon7 suggests a growing reluctance to diagnose SIDS when the death occurs while sharing the parental bed or when a suspicious history is presented such as parental drug and alcohol abuse, previous unexplained illness, or age and time of death “not typical” of SIDS. Clearly the decision on what constitutes a sufficient explanation of death is not an easy one to make and there is a need, in some cases, to highlight salient features that may be contributory or even fully explain a death. However this decision is not one the pathologist should be asked to make alone or without the required guidance. Most infants who bed-share or live with parents who have a problem with alcohol or drugs are still alive the next morning, while “typical” characteristics of SIDS are changing, as rates fall.1 General practitioners, health visitors, paediatricians, and child protection teams with broad experience of normal childcare practices in their communities are less likely to attribute death or injury to normal variants in patterns of childcare. A multidisciplinary team is needed to diagnose SIDS and requires a standardised classification system that highlights concerns but gives guidelines on what constitutes a full explanation of death.

In Avon we have been developing such a classification system over the past 20 years.1,2,8,9,10 It is based on a simple grid system, one axis representing the three components of the definition of SIDS (the history, the death scene examination, and the pathology), the other axis representing an incremental scale towards the explanation of death, with gradation between the completely inexplicable sudden death (classification Ia) and the fully explained (III). This system allows for notable factors to be identified (classification Ib); for instance, the fact an infant was put down prone to sleep or co-slept in the parental bed. This system also allows for factors that possibly contributed to the death (IIa), such as sleeping prone while heavily wrapped or co-sleeping under an adult duvet, and factors that probably contributed to the death (IIb) such as sleeping prone while heavily wrapped with an infection or co-sleeping under an adult duvet with an adult who had consumed a lot of alcohol. The overall classification of death is taken as the highest classification given among the three investigative components of SIDS. Note this is not a diagnostic system; all of these deaths are still SIDS as none of these factors on their own are sufficient to fully explain the death. Nor is this an attempt to redefine SIDS; all we are doing is trying to put into words our thought processes on the shade of grey, to quantify our degree of uncertainty between the inexplicable and explained. The presence of a finding that on its own gives a complete and sufficient explanation of the death (for example, meningococcal septicaemia) takes the diagnosis out of the field of SIDS, but there may still be important contributory factors to the death that have implications for future care of the family and the quality of care in the community.2,9,10

A multidisciplinary case discussion meeting is more than just reaching a consensus of opinion about the death. The meeting is for sharing information and future care planning for the family. It is important that there is an explicit discussion of the possibility of neglect or abuse as a contributory factor to the infant’s death, but equally the quality of the medical and social care given should also be discussed at the meeting, identifying any shortcomings and appropriate measures to improve future care.

The international acceptance of SIDS as an entity in1969 was an admission, perhaps a courageous one, that we were ignorant as to the cause of such deaths but aware of the profound effect on our infant mortality rates. By careful study we have substantially reduced the numbers, yet still do not fully understand why these infants die. Perhaps we need to be more courageous in our admission of ignorance, or at least distinguish between probable contribution and explanation when assigning the cause of death.

Commentary on the paper by Sheehan et al (see page1082)

View Abstract


  • Competing interests: Both authors were principal investigators in the CESDI SUDI studies,1 and have been involved in the development and publication of the Avon multiprofessional approach to the investigation of SUDI.2 PJF was a member of the Kennedy Committee.3

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.