Notes of all previous attendances in Sheffield can be checked

EDITOR—Sidebotham and Pearce's audit of child protection procedures in an accident and emergency department should raise general awareness that children in any such departments need special consideration.1 It should also raise specific awareness that non-accidental injury often presents to accident and emergency departments.

It is interesting that the nurses seem to have been given the job of scoring the risk factors for abuse in this study and that the doctors were given the job of auditing nurse practice. All accident and emergency staff should be informed and aware of the possibility of non-accidental injury, although the checking system would be carried out more efficiently if it was the responsibility of one clearly identified individual, in this case the initial assessment nurse.

The accident and emergency department at Sheffield Children's Hospital has a system of amalgamating a child's notes so that every previous attendance to the department can be checked at each new attendance. In accident and emergency one does not have the benefit that general practitioners have of having known the child and family over a period. Every clue to the child's health and lifestyle is needed to put the current attendance in context.

There has long been an alerting system not only for children officially on the child protection register but also for families who have caused concern to other health professionals such as paediatricians, health visitors, and general practitioners. Family problems such as alcoholism and substance abuse have an impact on children and their care. With the modern practice of working with parents and registering as few children as possible for as short a time as possible, there are many children and families who need extra assessment, care, and support when children are not officially registered.

I believe that there are even broader and more subtle considerations when trying to define a child at risk. The paediatric influence in accident and emergency departments would be best served by increasing the number of accident and emergency consultants with a special interest in paediatrics or at least by having a paediatrician linked closely to each department.

General practitioners need training in child protection

EDITOR—We support Sidebotham and Pearce when they recommend improvements in child protection procedures in accident and emergency departments by means of clear protocols, regular staff training, and increased levels of communication.1 Though appreciable numbers of children attend accident and emergency departments, general practitioners are now the first point of contact for most child health problems. They are therefore extremely likely to regularly encounter children who are at risk of abuse. General practitioners could undertake a substantial role in the overall child protection process in view of their increased commitment to health surveillance programmes. Experience to date, however, indicates that they have a relatively peripheral involvement in this area of their work, particularly with respect to attendance at child protection case conferences and participation in training events.2

We recognise that child protection issues present general practitioners with specific challenges. Many fear that the unique relationships that they hold with families may be compromised when they invoke child protection procedures, while others express anxiety about the medicolegal consequences of sharing confidential information with social services.3 The Department of Health's publication Clarification of Arrangements between the NHS and Other Agencies, while recognising these difficulties, is clear that general practitioners are required to place the safety and welfare of children above all other considerations and to participate fully in child protection procedures.4

We undertook five focus group discussions among 38 members of primary healthcare teams in the West Midlands in 1996 to identify themes relevant to child protection issues as they present in a primary care setting. We found that none of the 16 general practitioners who participated had received sufficient training in child protection. All showed impaired awareness of both their own role in the child protection process and local referral procedures. A similar lack of awareness of child protection procedures among general practitioners has been found elsewhere.5 Not surprisingly, child protection work was a source of anxiety and uncertainty to them. In addition to practice based multidisciplinary training in child protection, participants requested simple, A4-size flow diagrams that summarised local referral procedures.

Implementation of practice based clinical guidelines that reflect the policies of local area child protection committees would be a step forward. The specific training needs of general practitioners will need to be addressed by area child protection committees and those involved in postgraduate education. However, general practitioners must recognise the unique role they could have in this area and show further commitment to the protection of children from abuse and neglect.

Checks on children in Southend have to be via a social worker

EDITOR—Sidebotham and Pearce rightly emphasised the important role of accident and emergency departments in the recognition of non-accidental injury.1 The true incidence of non-accidental injury in paediatric attenders at accident and emergency departments is unknown, but few accident and emergency departments will identify more than a small proportion of cases. We should at least be able to offer adequate protection to those already on the child protection register. The authors state that they checked the register but do not make clear how they were able to do this. My department too used to check a child's name and date of birth against the register, initially on a simple rotary file and more recently in a secure file on the hospital computer. Recently, however, the child protection authorities have refused to allow us copies of the register, insisting that any access to it has to be via a duty social worker. This entails telephone calls and lengthy delays and is clearly impossible for all of the 30-40 paediatric attendances each day in my department.

In these days of sophisticated and well secured electronic data it should be possible for all accident and emergency departments to check children rapidly and accurately against the local register. The Department of Health should seek ways of extending this to national registers to prevent children falling through the net by attending departments in neighbouring districts. If suspicion of non-accidental injury is limited to those with risk factors or with overtly suspicious injuries we will continue to overlook substantial numbers of abused children.