Provision of intensive care for children

A geographically integrated service may now be achieved

Paediatric intensive care is a low volume, high cost specialty which depends on the training and skill of staff and availability of specialist equipment. Critically ill children have a changing range of illness and pathophysiology from early infancy to adolescence which is different from that of critically ill adults. In Britain paediatric intensive care has developed in an ad hoc and fragmented way. Now, however, after two decades of effort, Britain may be moving towards a more integrated service.

The Paediatric Intensive Care Society and the British Paediatric Association voiced concerns about paediatric intensive care in the early 1980s. The Paediatric Intensive Care Society has defined standards for paediatric intensive care,1 and these have been an important reference source that has informed many of the subsequent developments. This document defines two levels of paediatric intensive care. Level 2 refers to stable intubated children or unstable non-intubated children with airway problems requiring continuous nursing supervision, and level 3 to children who require complex therapeutic and nursing procedures in addition. (Level 1 is high dependency care.)

In 1993 a multidisciplinary working party on paediatric intensive care highlighted the fragmented configuration of paediatric intensive care provision.2 Its report was based on a retrospective survey of 12 882 children identified as having received intensive care in 1991: 29% were cared for in children's wards, 20% in adult intensive care units, and only 51% in paediatric intensive care units. Of the 2627 children cared for in adult units, 23% were aged under 1 year and almost 5% were aged under 1 month. In adult units fewer than 2% of nurses had a children's nursing qualification. Only 36% of paediatric intensive care units provided a transport service for retrieving critically ill children. All units reported refusal rates of up to 16% of their annual admission rate. The working party expressed particular concern about facilities where medical and nursing staff had not had specific training and where the staffing levels were too low for managing critically ill children.

In 1995 a child called Nicholas Geldard died in a paediatric intensive care unit after a spontaneous cerebral haemorrhage. Before reaching the unit he had been moved from the admitting hospital to another hospital for computed tomography and only then to an intensive care unit (in another region) for management. After publication of the resulting inquiry,3 the secretary of state asked for a report on the development of paediatric intensive care services,4 and the Department of Health set up a national coordinating group in June 1996 to develop a policy framework. Paediatric Intensive Care: A Framework for the Future, published in July 1997,5 sets out a strategy for developing and integrating the service for critically ill children within a geographical area, over a three year time scale. Implementation has been devolved to the eight English regions. The Scottish, Welsh, and Northern Ireland offices are also considering the document.

The framework report confirmed the picture of fragmented services. Ten of the 29 paediatric intensive care units had 3 beds or fewer and only six had 8 or more. The report cited studies from Britain and abroad which supported the average figure of 1.2 intensive care admissions per 1000 children per year. It considered configuration of the service in terms of flexibility of provision for acute illness and support for tertiary services, the latter often representing half of all admissions to paediatric intensive care units.

Several studies support the view that the most important element of the paediatric intensive care service is the skills and experience of the medical and nursing staff and that therefore the service should be centralised. A study comparing illness adjusted mortality for children living in the Trent region, where paediatric intensive care provision is fragmented among 19 centres, with that in the two paediatric intensive care units in Victoria, Australia—which has similar size of child population and similar rate of admission to paediatric intensive care—showed both an excess mortality and a greater length of stay in Trent.6 Studies from the United States7 and the Netherlands8 showed lower mortality in specialist tertiary paediatric intensive care units.

The framework document's standards also say what facilities each hospital in a geographical area must provide as its part of an integrated service. The service is to be configured around three types of hospitals: district general hospitals, lead centres for paediatric intensive care, and single specialty hospitals. In a few parts of England a fourth type, a major acute district general hospital, has been designated (generally where the lead centre is some distance away). These have a large throughput of adult intensive care patients and already manage an appreciable number of critically ill children, and they may continue to do so up to level 2 intensity of care if they meet the standards. Otherwise district general hospitals must have medical and nursing staff on site who can resuscitate and stabilise critically ill children and separate child orientated facilities to provide support until the arrival of the lead centre's transport team. Single specialty hospitals caring for children must develop joint protocols with the lead centre and also meet specified standards. Within each region a lead centre, or at most two, must be designated, serving a population of at least 500 000 children.

Lead centres should be based in hospitals with a full range of tertiary paediatric services, run a 24 hour transport service for the region, and have sufficient throughput to maintain staff expertise and act as educational and training centres. Lead centres will be responsible for data collection, audit, and developing joint protocols with the other hospitals within the region. Future consultants will have had training in centres approved by the Intercollegiate Committee on Training in Paediatric Intensive Care Medicine. By July this year children who need intensive care should no longer be cared for in general children's wards, and within another two years children should no longer be cared for in centres which do not meet the standards.

Quality of paediatric intensive care includes effectiveness and appropriateness of treatment within a child and family orientated environment. There is no validated paediatric scoring system for severity of illness in the United Kingdom and no information about long term outcome. We urgently need such studies so that further reorganisation of the paediatric intensive care service is informed by research and audit.