Why Children Die: a commentary on the CEMACH report

Vignette 1

A teenager took a potentially lethal overdose. When the overdose was discovered she was brought to A&E and was seen by a senior house officer (SHO). She gave an honest history in relation to the type and amount of drug ingested. The doctor did not check that the dose involved was potentially lethal. She was sent home without arrangements for follow-up. Her condition deteriorated over the next two days. When she re-presented to A&E, there was a critical failure to recognize the severity of her symptoms and a consequent significant delay in medical management. She collapsed while waiting in the A&E and started to convulse. She died later in intensive care.

The drug in question here could have been any one of a number of potentially lethal medicines, but let us imagine that it was paracetamol because this is still the most easily available analgesic in most households; and it is especially important because, unlike some other drugs, there is a specific antidote that can be given to prevent severe toxicity and death. One important public health measure that was taken in England in 1998 was to limit by law the pack size of paracetamol that can be purchased at any one time; this was associated with a marked reduction in the incidence of lethal paracetamol overdose, ² but the problem has not disappeared.

Senior house officers in an A & E department may have had no clinical paediatric training since being medical students, and need to have good support and supervision as they gain much needed experience with children in this setting. So the first issue is that of clinical support from someone more senior, since this could have helped the SHO to realize the importance of the clinical history they had been given, to check whether the amount of the drug was trivial or not, and to decide whether the patient needed active management rather than being sent home. Guidance on what constitutes a potentially serious overdose, including paracetamol, is available in the British National Formulary and is very easily found using Internet search engines, so this is in no sense obscure or arcane knowledge.

Sending home a teenager who has self-harmed, without any arrangements for follow-up, contravenes the relevant National Institute for Health and Clinical Excellence (NICE) guidance on self-harm. ³ It is unrealistic to expect junior medical staff to be familiar with the details of every pronouncement from NICE, but it is reasonable to expect that those in a supervisory capacity would be aware of this framework and that local policies would have incorporated the guidance. After all, SHOs are there to learn about these things; it is up to the system within which they function to ensure that their experience is not gained at the expense of the patient.

Not surprisingly, the story does not end there. When the teenager comes back to the same A&E a couple of days later, the severity of her illness is not appreciated so she waits in the general waiting area until she collapses with a seizure. Now it might well be that the two days of delay had irrevocably cast the die in terms of a lethal outcome, so that the effects of the overdose had progressed to the point of irreversibility, and even with intensive care, she would not survive. The important question that emerges from this part of the tale is whether the arrangements for triage were adequate, and whether the failure to recognize the seriousness of the situation was an issue of education and training, or of supervision.

Vignette 2

A 10-year-old boy at known risk of high blood pressure for medical reasons presented to A&E with clear symptoms and signs of hypertensive encephalopathy. The initial assessment of risk was wrong and the case-notes were misplaced for one hour. Nurses failed to measure his blood pressure and did not recognize the severity of the encephalopathy. He started to convulse when attempts were eventually made to control his blood pressure. His cranial CT scan was misinterpreted as normal by a neurosurgical registrar. He died despite emergency transfer for surgical intervention.

Here, there was a clear failure of the A&E triage system, compounded by failures of several professionals to do the right things. Like the previous scenario, there were issues of supervision and decision-making by inappropriately junior staff. Significantly, specialist advice was not sought and adult-trained personnel tried to manage a situation that was clearly outside their competence, illustrating that sometimes ‘you don't know what you don't know’. It was a consistent theme that some children were dying preventably after inappropriate management that, had there been input from personnel with paediatric training (nursing or medical), there was a high probability that matters might have turned out differently.

Vignette 3

A young boy, old and tall enough to travel without a booster seat, was a back-seat passenger in a car which was involved in a head-on collision with another car. He was not wearing a seatbelt at the time and died from multiple injuries. The other three occupants of the car had seatbelts on and were not injured. There was no suggestion of poor driving conditions, visibility or other road hazards. Neither driver had been drinking.

This case demonstrates that merely having laws about the restraint of passengers in cars, including children, is not by itself sufficient to prevent unnecessary injury and death. In the report, this vignette came under the heading of public information, but in reality the situation is much more complex than this. Knowledge of the law and of the efficacy of seatbelts is commonplace; attitudes and behaviour are a different matter. Seatbelts will only be worn if the driver, or other occupants of the car, ensures that children are in the habit of using them. This is not just in the interests of the otherwise unrestrained passenger: even a little child can become a lethal missile that can kill an adult or another child in the vehicle if a collision occurs.

The case also demonstrates that the agents whose lack of awareness, skills, knowledge or appropriate behaviour contributes to a death are not just professionals – they can also be families. The factors that contribute to the failure to wear a seatbelt may be rooted in the way a family functions, or ways in which peers or adults provide role models for children: not wearing a seatbelt may have been a habit, or have been regarded as more macho than choosing to wear one, or have been what the child's friends did.

Vignette 4

A teenage girl in an apparently stable long-term foster placement committed suicide. The act appeared to have been precipitated by a missed visit with the natural mother with whom she had not lived for several years. The review panel considered that the girl's behavioural difficulties were symptomatic of her previous psychological trauma that included sexual abuse.

That suicide in children appears to be much more common than the official statistics reflect was a very important finding. Not all suicides are preventable, but in each individual case it is often possible to identify opportunities for managing vulnerable children differently that have been foregone. It is unlikely that the only factor in this child's suicide was the failure of her mother to visit, even though it seemed to precipitate the event. Children such as this girl are inherently emotionally vulnerable and often have a number of psychological difficulties, but tend to be reluctant to engage with conventional services. They are often suspicious of general practices, where they often feel patronized by staff, and they tend to be concerned about confidentiality (sometimes with good reason). Children such as her need to have access to services they can trust and in which they can feel comfortable.

Vignette 5

A 17-year-old boy hanged himself at home. He had a history of suicidal thoughts which began years previously. He was admitted to hospital with an alcohol overdose aged 13 years and reviewed afterwards by a child psychiatrist and family therapist. The psychiatrist wrote to the family for follow-up arrangements but they did not reply and no further action was taken. He had annual reviews by the GP for his asthma in the four years leading up to his death. It was recorded that he was using his inhaler too frequently and always running out. There was no mention of his psychological state or care in the GP records.

In contrast to the previous case, this child maintained contact with his general practice, regularly attending because of his asthma. His asthma control appeared to be poor, and though it is common for poor control to reflect psychological rather than physical issues, the focus was on the asthma rather than his mental state. Thus a possible opportunity for ascertainment and intervention was lost. Now, it would be ridiculous to suggest that just because someone is in touch with services prior to committing suicide, those services should have necessarily have been able to spot a patient's depression and do something about it. Furthermore, any risk assessment is only as good as the day, or even the hour, at which it is done. But as practitioners we should reflect on the importance of a holistic approach, and remember that just because someone has asthma, it does not mean that it is either their only, or even their most important problem.

Vignette 6

A 16-year-old girl and three other members of her family were killed in an arson attack committed by her ex-boyfriend – a 31-year-old who had been released early from prison for killing his previous girlfriend by attacking her with a mallet. He was in breach of his licence conditions at the time that he committed the arson attack and the review panel were unable to discern what steps, if any, had been made to supervise the licence.

The agency singled out for criticism by the review panel in this case was the probation service, because of the issue of supervising a violent offender on release from prison. But several other issues can be discerned. First, although we don't know whether or not the man was or had been mentally ill, there is still poor recognition among adult mental health and forensic services of the knock-on consequences of the behaviour of patients or clients on children: sometimes the patient's own children, and sometimes other children. In this case, there is also the possibility that the older boyfriend may have been having an illegal and abusive under-age sexual relationship with the girl who was 16 years old at the time of her death, but may have been 15 when the relationship began, which brings up another dimension altogether. Finally, it is in the nature of the case that we can't be certain (and neither might the review panel) that the failure was necessarily or entirely that of supervising the convicted man: for example, it may have been a failure of witness protection.

Vignette 7

The child to which this letter referred died later of an acute asthma attack.

‘Dear [General Practitioner],

This case does not prove that attendance at a hospital clinic prevents deaths from acute severe asthma. What we uncover here is a particularly insidious, institutionalized means of disadvantaging children. It is not the children who fail to come to the clinics at hospital; it is their parents who may be unable to bring them, for various reasons. Indeed failure to attend is one of the markers for particularly vulnerable children and families. In the wake of the launch of ‘Why Children Die’ we learnt from professionals that commissioners often imposed these policies on Trusts. Policies designed for managing failures of attendance among autonomous adults should have no place in services for children, and the failure to discuss, consult on or think through the unintended consequences of such rules is an indictment of the commissioning process.