A. You should see the child within two hours if he or she has already been triaged (perhaps by NHS Direct or more likely your own practice nurse triage system) and thought to have a 'red' symptoms.
You would assess the child fully, taking a history and examining the child with particular focus on the traffic light table (download a PDF of the table). Your examination would include measuring their temperature with an electronic thermometer in the axilla, taking their pulse, respiratory rate, blood pressure (if available) and capillary refill time. You would record all your findings in the notes.
Children with life-threatening features would need immediate resuscitation and admission. Children with red features should be considered high risk and admitted urgently to a paediatric unit (not A&E). Children with no red features but amber features are at moderate risk of serious illness. The GP must manage this risk appropriately. Some will require admission, others further assessment later, some can be looked after at home with good instructions to the carer or parents on what warning symptoms to look out for and how to act: in effect good safety-netting.
You would assess the child's risk and safety net accordingly. Children with only symptoms in the 'green' category and signs are low risk and can be managed at home.
Q. I never measure a child's temperature. Why should I start now?
A. The review of research carried out by the NICE team showed parents report of fever can be trusted and there is only a weak correlation between height of fever and severity of illness. However it is easy to miss fever and in the very young (under six months) fever is predictive of serious illness. Placing a thermometer under the axilla at the beginning of your assessment will not increase the time it takes to see a child.
Q. I can spot a sick child, why do we need this guidance?
A. I expect you can. Collapsed children are not difficult to spot, and therefore most GPs are probably spotting high-risk 'red' children already. The problem is that the evidence shows we are not managing unwell children as well in the UK as in the rest of Europe. Partly this may be down to failures in secondary care (hence NICE's drive to ensure feverish children are cared for by paediatric specialists) but also there is concern that primary care (in its broadest definition) is missing some children.
We all have stories of children who were seen in the afternoon and collapsed later that evening. Could some of these children have been picked up earlier? There is no direct uncontroversial evidence but collapse is the end stage of a temporal process and the whole premise of admitting children for observation is based upon this.
In addition there are now increasing numbers of professionals working in primary care and the proper assessment and recording of information is vital to allow review by another person later.
Q. Where do the symptoms and signs that make up the traffic light table come from?
A. The clinical group looked at large number of predictive studies that looked at the predictive value of certain symptoms and signs. Many were discounted because they included the use of investigations not available to primary care professionals. The Yale Observation Studies, however, are entirely observation based and make up the basis for most of the symptoms and signs. The group added others where studies existed that confirmed that they were useful. So each one of the symptoms and signs has evidence for being there. This is not an idle list. Vomiting alone does not find its way on the list because it is poorly predictive of severity of illness. Bile-stained vomiting is predictive.
Q. Why does the guidance not support using paracetamol and ibuprofen together?
A. Because there is no evidence that this combination does any good but increase the cost of antipyretics to the NHS and parents. The issue of ‘treating fever' has become more steeped in fashion and tradition than in any evidence.
Treating fever with antipyretics can undoubtedly make a child who is feeling unwell feel better. It will also bring down a fever but it will not stop febrile fits, nor can it be used to filter out seriously ill children from those who will be fine. Parents phone up or consult because they are concerned about fever and we do them a disservice by concentrating on getting rid of the fever at the expense of properly assessing the child and reassuring parents that fever is in itself not harmful. An antipyretic can help a distressed child feel better and that is why we use them, not because they affect outcome in some way.
Q. The guidance in the practice is short on detail, where can I get the complete guidance?
The complete guidance is available on the NICE website.
However if you want a document with the full reasoning behind the guidance including the Delphi consultation and a full database of the 300 papers the clinical group reviewed, contact the National Collaborating Centre for Women's and Children's health (firstname.lastname@example.org). There will be a charge for this.
Dr Cave is a GP in Newbury and a member of the guideline development group for fever in childhood