Protocol for the assessment of self-harm in young people: initial audit and training implications

AIMS AND METHODS This paper describes the development and initial audit of a protocol for 
the assessment of young people up to the age of 18 years who presented to the 
accident and emergency department (A&E) with self-harm. A key part of the 
project was education and training. RESULTS Regular training of senior house officers (SHOs) in A&E may have 
contributed to an increase in young people being admitted to a bed for proper 
assessment (as per the protocol), but psychosocial assessments undertaken by 
SHOs in A&E were still only partial, and there was no apparent use of the 
mental state examination. CLINICAL IMPLICATIONS Child and adolescent mental health services have an important role to play 
in liaising with local A&E departments in training of junior staff in 
psychosocial assessment and the use of the mental state examination. This is 
especially relevant in the light of the new training requirements of the 
foundation years.


RESULTS
Regular training of senior house officers (SHOs) in A&E may have contributed to an increase in young people being admitted to a bed for proper assessment (as per the protocol), but psychosocial assessments undertaken by SHOs in A&E were still only partial, and there was no apparent use of the mental state examination.

CLINICAL IMPLICATIONS
Child and adolescent mental health services have an important role to play in liaising with local A&E departments in training of junior staff in psychosocial assessment and the use of the mental state examination. This is especially relevant in the light of the new training requirements of the foundation years.
Self-harm in young people is a significant public health problem and the guidelines produced in 2004 (National Institute for Clinical Excellence, 2004), in the same month that this audit was planned, and therefore not taken into account in this work, highlight the importance of psychosocial assessment, although nowhere has it been specified who should be undertaking this.
In 2002, approximately 20 young people per month were presenting to the accident and emergency (A&E) department of Derriford Hospital, Plymouth with selfharm. Excluded from these figures are children under 5 years of age, young people 18 years or older and those who had ingested alcohol (approximately 5 young people per month), drugs and other substances by accident.
Prior to 2002, a protocol for the assessment of selfharm in young people had existed but was not easily available and had not been revised for some time. It was unclear whether trainees knew of its existence. Staff in the child and adolescent mental health service (CAMHS) would often hear anecdotally about young people who were discharged from the A&E department after taking significant overdoses. There were also frequent difficulties in the A&E department over who was responsible for assessments.
An initial survey undertaken by a senior house officer (SHO) attached to the CAMHS at the end of 2002 indicated that a third of young people presenting with self-poisoning were admitted to the paediatric ward, while two-thirds were discharged straight from the A&E department following psychosocial assessments of varying quality. Two further surveys were undertaken during this time, or started shortly afterwards, and concerned the nature of the psychosocial assessments undertaken by paediatricians on the wards and what happened to young people aged 16-18 years. It emerged that trainees in paediatrics were undertaking good psychosocial assessments but confused the mental state examination with the assessment of consciousness using the Glasgow Coma Scale (Teasdale & Jennett, 1974). Moreover, paediatric trainees who had been trained outside the UK had no experience of psychiatry as medical students. The proportion of young people aged 16-18 years being admitted to a ward had increased slightly from the previous survey (15 out of 30). A third of patients (10 out of 30) in this age group are presenting at the weekend, when the duty rota for routine assessment by CAMHS was not in operation.

Development of the protocol
A multidisciplinary, multi-agency group was therefore convened by S.H.-D. comprising A&E consultants, the liaison health visitor (also a child protection advisor), a social services team leader and a consultant in paediatrics. The working party met on four occasions between July 2003 and April 2004, during which time successive drafts of the protocol were amended as consensus was reached.

Features of the protocol
The protocol incorporates national guidelines from the Royal College of Psychiatrists (1998) and is informed by guidelines from a number of other CAMHS (e.g. Chichester, Bath and Swindon). It specifies what is required as part of the psychosocial assessment, and teaching, audit and regular review are essential elements.
Important aspects of the protocol are: . the recommendation that all young people up to 18 are admitted to a ward and assessed the next working day by CAMHS, regardless of the type or quantity of overdose taken . the importance of the psychosocial assessment (as specified by the College (Royal College of Psychiatrists,1994) and highlighted by the National Institute for Clinical Excellence (2004) . a recommendation that suicide intent be estimated using the PATHOS scale (Kingsbury,1996), which is simple to administer . an outline of who to contact if there are child protection concerns . a summary of legal matters pertaining to emergency treatment, such as consent issues and use of the Mental Health Act 1983 . a system of triage undertaken by doctors in paediatrics and A&E at weekends (Friday and Saturday) when young people are refusing to stay until Monday. Thus, the young person, on the basis of the psychosocial assessment and therefore risk, can be: . discharged home (low risk) . or discharged home with request to CAMHS for follow-up (low risk but ongoing difficulties) . or the SHO in psychiatry is called, who can then if necessary liase with the consultant child psychiatrist on call (high risk).
Although the protocol deals mainly with young people presenting with self-poisoning, suggestions are also made concerning the further management of those young people presenting with alcohol intoxication and self-mutilation.

Teaching
Teaching was provided to SHOs in A&E by S.H.-D. from April 2003 on a 6-monthly basis. This not only provided SHOs with a basic knowledge about the CAMHS, but developed into a forum for disseminating the protocol and for discussing the difficulties associated with the assessment of self-harm in young people. Attitudes are notoriously negative towards individuals who present with self-harm, especially if they are 'repeaters', or are seen to be wasting doctors' time, or if they don't appear to be overtly distressed. It was gratifying that SHOs could often see their responses as unhelpful after a group discussion, and found it useful to appreciate that young people presenting in crises can often improve over time.

Aims
The audit of the protocol in its final draft (March 2004) examined adherence to the protocol by measuring the following four standards in the A&E department: . that all children under the age of 16 following an overdose should be admitted to a paediatric bed . that young people between the ages of 16 and 18 should be admitted to the A&E observation ward overnight . that a psychosocial history be undertaken (according to the guidance issued by the Royal College of Psychiatrists in1994) and also now part of the NICE guidelines (National Institute for Clinical Excellence, 2004). Areas to be covered are: level of consciousness, triggers to the overdose, degree of intent, past psychiatric history, drug and alcohol history and a mental state examination . that there is a clear treatment plan in the notes.

Method and results
We inspected 20 consecutive casualty cards of young people presenting with self-harm during the month of September 2004. Children under 5, young people who were already 18, and those who had ingested alcohol, drugs and other substances by accident were all excluded. All 20 assessments had been undertaken by SHOs in A&E. For children under 16 years, 7 out of 10 (70%) were admitted to a ward and 2 (20%) discharged home (in 1 the outcome was unknown). For young people between 16 and 18 years, 6 out of 10 (60%) were admitted and 3 (30%) discharged (again in 1 the outcome was unknown).
Doctors in the A&E department performed Glasgow Coma Scale assessments in 19 (95%), asked about triggers to the overdose in 12 (60%), degree of intent in 11 (55%), past psychiatric history in 17 (85%), and substance misuse in 10 (50%). In only one instance was a mental state examination performed and this was because that patient had also been seen by a psychiatric SHO on request. A treatment plan was recorded for 11 young people (55%).

Discussion
Clearly, there are limitations to this work, the most obvious being that the number of case notes examined was very small. Other criticisms apply to the imperfect way the protocol was disseminated up to the time of the audit; it had not been fully published in the A&E department handbook by then, and it is also highly likely that not every SHO was able to attend the single teaching session offered every 6 months owing to shift patterns of working. These points must therefore be borne in mind when considering the results. Nevertheless, in this snapshot of practice in September 2004, results of this exercise raise two main questions. Is the protocol being followed? The answer is 'to an extent', if we can take Huline-Dickens & Adiele Protocol for the assessment of self-harm in young people education & training figures for admission to a bed as an indication. From the initial survey in 2002, approximately two-thirds of young people were being discharged directly from the A&E department, and now that proportion is 35%. Who should be undertaking the psychosocial assessments? Although it had been agreed that as part of the protocol junior doctors were to undertake these, and training was provided (notwithstanding the limitations above), it was clear that although A&E doctors felt comfortable with some aspects of the psychosocial assessment, such as the use of the Glasgow Coma Scale, no mental state examinations were performed in this small sample. Much has been done since this audit was undertaken to improve practice further. A front sheet has been designed for use in the A&E department, which will outline the psychosocial assessment and include prompts for the mental state examination, which, it is hoped, will enhance recording of information and also prevent duplication of work. The protocol has also been incorporated in the A&E department handbook. Plans also exist for it to be made available on the Plymouth Hospitals NHS Trust's intranet.
What are the possible reasons for the lack of mental state examinations in this study? Perhaps it was assumed that these would be undertaken by another assessor later (for example by CAMHS staff the next day), or perhaps a mental state examination was not considered relevant to adolescents, or perhaps SHOs feel unskilled in dealing with young people in complex psychosocial situations. (We also recall the earlier point that doctors from overseas may not have had any psychiatric training.) It has been suggested by some that mental state examinations should only be undertaken by CAMHS staff who are used to doing these routinely in young people. However, it is the authors' view that the mental state examination is an important assessment tool and its use by doctors of any specialty should be encouraged. Furthermore, performing a mental state assessment is now a core competency that all graduating medical students and foundation trainees should be able to do, as required by the General Medical Council and Postgraduate Medical Education and Training Board (Foundation Programme Committee, 2005).
Finally, since this audit was planned before the introduction of the NICE guidelines (July 2004) and the curriculum for the foundation years (implemented in August 2005), and also as further improvements to the procedure have been made, it will need to be repeated with these in mind.

Conclusions
It is hoped that this project has provided useful training experiences in audit activity for junior doctors attached to the CAMHS department. The results of this small audit show that the protocol is largely being followed, and as a result we hope that patient care is improving. This project has also raised the interesting question of the use of the mental state examination by non-psychiatric SHOs. Child and adolescent mental health services have an important role to play in liaising with their A&E colleagues concerning the group of vulnerable young people who self-harm. Teaching and training of SHOs is an important aspect of this relationship.

Declaration of interest
None.