MANAGEMENT

Services for assessment and aftercare following self-harmDavid Owens

Despite the status of non-fatal self-harm as the most significant warning of the danger of suicide, the provision of healthcare ser-vices for those who harm themselves remains in disarray around the UK, and in many other countries, too. For example, the press release that recently heralded England and Wales’s national guidelines for the treatment of those who harm themselves offered the sorry admissions that: ‘170,000 people a year attend emergency departments because they have self-harmed, of those an estimated 80,000 never receive a psychological assessment or follow up even though the risk of committing suicide after self-harming one or more times is 100 times greater than the average risk in the population’ and ‘few people providing care in casualty understand why people self-harm and don’t know how to help them effectively.’1 Thankfully, as we shall see, there is room for some optimism as the recent official guideline and the emergence of useful research evidence pave the way for badly needed improvement in this sorry state of affairs.

Attending hospital

This contribution will deal only with healthcare services that might be available to those who attend the general hospital, but it is important to bear in mind that many people, especially young people, stay away from the emergency department follow-ing self-harm. In a large survey, as few as 1 in 8 young people of school age who had undertaken self-harm reported going to hospital; much of the self-harm in this survey was by self-cutting but, even when the harm was poisoning, only 1 in 4 reported hospital attendance.2 This contribution does not, however, deal with the hospital’s response to children and young people who harm themselves, for whose care there are additional and spe-cific recommendations.1 Little will be said here, either, about episodes in which the primary care team are the first contact. In urban areas at least, most self-harm episodes that lead to contact with any part of the health service are dealt with in the hospital emergency department; in a study in south London, over 90%

David Owens BSc MBChB MRCPsych MD is Senior Lecturer and Honorary

Consultant Psychiatrist at the University of Leeds Institute of Health

Sciences and Public Health Research, and at Leeds Mental Health

Teaching NHS Trust, UK. He has a long-standing research interest in

services for self-harm. He qualified in Leeds and undertook much of

his psychiatric training in Nottingham. He currently spends much of

his time organizing a postgraduate research training programme at

the University of Leeds.

PSYCHIATRY 5:8 271

of self-harm episodes identified through general practice records involved attendance at the emergency department.3

The emergency department

When someone attends hospital after self-harm they generally go to the emergency department where, in England and Wales, there are in place very specific guidelines for their care. The recent guidance comes from the National Institute for Health and Clinical Excellence (NICE) – an independent body that develops guidance on topics commissioned by the Department of Health;1 once NICE has published clinical guidance, health profession-als and the organizations that employ them are expected to take it fully into account when deciding how to treat patients. The NICE guideline on self-harm specifies many aspects of the kind of service that patients should receive.

At triage, the staff of the emergency department are expected to combine assessment of physical and mental state in a respect-ful and understanding way, taking account of any emotional dis-tress. It is expected that doctors and nurses who are not mental health specialists will nevertheless provide sufficient psychosocial assessment to determine mental capacity, the presence of mental illness, and the patient’s willingness to remain for more detailed psychosocial assessment. If there is drug or alcohol intoxication, assessment may be quite unreliable or even impossible to carry out, and waiting for adequate assessment or treatment should be in a safe and supportive environment – supervised by a member of staff if necessary. The guideline clearly states that someone who wishes to leave before he or she has received a psychosocial assessment or treatment should, if their mental capacity is dimin-ished or they have a significant mental illness, be prevented from leaving and referred for urgent mental health assessment.

Audit of emergency department case recordsIt is plainly asking a lot of emergency department staff that they should undertake a preliminary psychosocial assessment that is reasonably thorough, but there has been a clear demonstration in Leicester, UK, of how clinical audit can bring about gratifying improvements. Three years after an initial audit the emergency department staff were found at re-audit to have made substantial progress. Although there were still deficits – especially when it came to asking about substance use, and a basic assessment of the present mental state – the emergency department staff were routinely recording in the case record much of the important information that is required for basic clinical care.4

Psychosocial assessment

Generally speaking, following triage, patients will be dealt with in the main emergency department. For decades, it has been the health service’s official policy that all patients who attend hospi-tal should, before discharge, receive a psychosocial assessment carried out by staff who, regardless of whether they are mental health specialists or general staff in the emergency department, are specifically trained for the task.5,6 Under the new guidance it is similarly expected that everyone who attends because of self-harm should receive a comprehensive assessment of their needs and risk. Unfortunately, until now there has been woeful and widespread failure to meet this basic expectation of care;7–10

© 2006 Elsevier Ltd. All rights reserved.

MANAGEMENT

we must hope that the higher profile for NICE guidelines than for previous requirements will effect a transformation in services.

Assessment of mental health and social needs (and risks)NICE guidance states that everyone’s assessment should include evaluation, and recording in the case records, of the environmen-tal, psychological and motivational factors specific to the act of self-harm, paying attention to current suicidal intent and hope-lessness, as well as providing a full assessment of mental health and social needs. So that there is a thorough assessment of risk, there should also be an identification of the main clinical and demographic features known to be associated with the likelihood of further self-harm or suicide.

So-called ‘risk assessment’ as the guiding principle of psycho-social assessment is a flawed notion. It might be useful were it a reasonably accurate business, but it is not. The predictive values of key questions, or of repetition scales that have been constructed, are low. Put another way, low specificity of the risk factors means that relatively few of the patients who seem to be at high risk go on to repetition over the months that follow,11 while low sensitivity of these scales means that those who seem to be at low risk account for most of the cases of subsequent suicidal behaviour.12,13

Consequently, the appropriate psychosocial assessment of someone who has attended hospital after self-harm will arrive at a formulation that blends the assessment of the two targets – needs and risk. Of course, the patient needs to be mentally fit if the assessor is to make supportable judgements about the nature of the needs and risk; that is, not too drowsy or intoxicated. Often, even for immediate planning of care, the views and corroborative accounts of key informants will be needed. The patient and the person who has undertaken the assessment should read and agree what has been written about needs and arrange for this to be passed to the general practitioner. The main components of assessment of need after self-harm are set out in Table 1.

Service arrangements

The Royal College of Psychiatrists, through a consensus state-ment in the early 1990s and a recent update, has set out in detail some minimum standards for the quality of such a self-harm

The main components of assessment of need after self-harm

• Social situation (including current living arrangements, work

and debt)

• Personal relationships (including recent breakdown of a

significant relationship)

• Recent life events and current difficulties

• Psychiatric history, including any history of previous self-harm

and alcohol or drug use

• Mental state examination

• Enduring psychological characteristics that are known to be

associated with self-harm

• Motivation for the act

Table 1

PSYCHIATRY 5:8 27

assessment service.14,15 Expectations are too many to detail here, but the principles are straightforward and in keeping with the NICE guideline. First, it strongly promotes the idea that a self-harm service planning group should take the lead role; the group would include, for example, senior staff from the emergency department and the mental health service, and representatives of the hospital management, service users and the voluntary sector. The group should undertake the development and running of the hospital’s response to patients who attend because of self-harm.

Well-trained and supervised assessorsIt is generally accepted that there should be arrangements in place at all hospitals for the psychosocial assessment of patients who attend as a result of self-harm. One principle is that the task should be undertaken by someone who has training in such assessment and who is supervised by others who have the time and expertise to supervise properly. The Royal College’s report sets out clinical competencies for staff, at both basic and specialist levels of assessment. In practice, for assessment to be satisfactory it will usually need to be undertaken by specialist mental health staff. Increasingly, this task has been carried out by designated mental health nurses working regularly in this area.

Second, the Royal College’s report offers the planning groups a series of standards for immediate care, psychosocial assessment and aftercare. These stages of the patient’s episode of care are considered according to whether the patient is in the emergency department or has been admitted to a medical (or short-stay) in-patient ward.

Facilities for good careThird, the Royal College’s report proposes some basic standards for the facilities that should be available. These include the provision of privacy for patients, security for staff and the rapid availability of specialist mental health staff to attend the emergency department when urgently needed. The report repeatedly sets standards for written records of care: what should be included and how quickly it should be passed on to the patient’s general practitioner.

Another important provision to be found in the NICE guide-line is the need for the hospital to make available on occasions an overnight stay for patients who cannot be adequately assessed – perhaps because of intoxication or because key informants cannot be located quickly.

Interventions

It is widely thought that there is no clear evidence about what are the best interventions for people who attend hospital as a consequence of self-harm. This absence of a simple remedy might be no surprise considering how diverse the determinants of self-harming behav-iour are. The NICE guideline puts this heterogeneity rather well: just as cough may be due to smoking, common cold, tuberculosis, or lung cancer – each pointing towards a different course of action – so it is with self-harm;1 and presumably there are interventions that are more or less suitable in different circumstances.

Thin evidence for effective interventionThe trouble is that we have little evidence about effective-ness even where there have been investigations of treatments for relatively homogeneous subgroups of patients. There have

2 © 2006 Elsevier Ltd. All rights reserved.

MANAGEMENT

been three basic weaknesses in the research base. First, the few clinical trials undertaken have generally been too small to be useful, and they have too often been targeted at selected sub-groups of patients. Second, most of the trials carried out have offered disappointing findings.1,16 Third, although we need inter-ventions that will reduce suicide rates, suicide is an infrequent outcome after self-harm, and a trial to show whether an interven-tion would reduce suicides would need many thousands of parti-cipants; no such trial has been carried out (nor is one likely) and, instead, non-fatal repetition has been the principle outcome.

Simple alterations to the service structureThere have been two randomized trials in Bristol, UK, to deter-mine whether people who have attended hospital because of self-harm benefit from being given a printed card that sets out various offers of help, should the person go on to feel like further self-harm – always provided that he or she has not already self-harmed again on this later occasion. The first of these ‘Green Card’ studies suggested advantage to the issue of the card but the larger second study contradicted the earlier findings. In short, both studies suggested that the issue of a card might help patients attending for the first time (although neither study showed a statistically significant reduction in repetition); conversely, the second study showed a significant worsening of repetition among patients receiving the green card when their current self-harm was already a repeat episode.17,18

Recently, a large sample of people who attended the emergency department in Newcastle, Australia, because of self-poisoning were randomly allocated to receive (or not) a series of postcards sent by the emergency department over the following year.19 The postcard was couched in an informal way, saying that the staff at the hospital’s toxicology department hoped things were going well, and that the recipient would be welcome to drop them a note. Although there was no significant reduction in the propor-tion of people who repeated, there were fewer repeat episodes – mainly because of a smaller number of further poisonings among a few women in the postcard-receiving arm of the study. It would be speculative to predict a big impact from an intervention of this nature in highly urbanized communities in the UK, where we know that a substantial proportion of those who self-harm change address frequently without making arrangements for forwarding mail or notifying the change.

Brief psychological treatments

Overviews of the evidence from clinical trials: evidence about psychological treatments was painstakingly brought together a few years ago in a Cochrane review and again for the recent NICE guideline.1,20 In neither case did the review point to clear indica-tions for practice. Each found that most of the evidence about psychosocial interventions arose from trials of problem-oriented, particularly problem-solving, treatments. Disappointingly, both sets of reviewers resorted to an appeal for more research – calling for larger trials to determine whether repeat episodes might be reduced by psychosocial interventions. Between them, the two systematic reviews identified 7 randomized controlled trials of a brief psychological or psychosocial therapy in adult patients; each review was able to amalgamate results in a meta-analysis for 5 of the trials, and each found a pooled value which favoured the

PSYCHIATRY 5:8 27

problem-oriented therapy over standard care, but these reduc-tions did not amount to a statistically significant effect – either for non-fatal or fatal repetition.

Psychodynamic interpersonal therapy: the meta-analysis in the NICE review did not include a recent trial, carried out in Manchester, of a form of therapy which aims to change people’s ability to solve problems in interpersonal relationships;21 it was excluded because of difficulties with extracting its data in the form needed for pooling with other studies. The trial provided, nonetheless, a positive and encouraging finding: that psycho-dynamic interpersonal therapy led to a statistically significant reduction in repetition when compared to the effects of usual care. Participants were consecutive patients who had attended a university hospital’s emergency department due to self-poisoning, although half of those approached declined to take part. The principal outcome of the trial was repetition of self-harm; repeat episodes were defined as all self-reported acts of self-harm – not just those that led to hospital attendance. Six months after entering the trial, only 9% of those who had under-gone the new treatment had repeated, compared with 28% of those who had received the usual care. The therapy was carried out in just 4 sessions, delivered in the patient’s home; its main features are summarized in Table 2.

Cognitive therapy for prevention of suicide attempts: since then there has been a further trial, carried out by Beck’s group in Philadelphia, USA, which has shown clear benefit to 10-session cognitive therapy when compared with usual care.22 It is an important feature of the study that a great many patients who attended the hospital because of self-harm were excluded; the trial’s participants were only those in whom the research team judged the initial self-harm to be a suicide attempt. The trial demonstrated a significantly lower rate of repetition and a lower proportion of patients repeating among those receiving the cognitive therapy; the treated group were about half as likely to make a suicide attempt during the 18 months of follow-up. The principles of the therapy are set out in Table 3.

Key aspects of psychodynamic interpersonal therapy for self-harm

• A focus on the self-harm episode

• Use of ‘here-and-now’ and ‘focus-on-feelings’ strategies

• Identification of interpersonal difficulties and problems

• Identification of maladaptive patterns in relationships,

disputes and dilemmas

• Development of a strong working alliance with the therapist

• Testing of problem solutions in therapy

• Testing of problem solutions in relationships outside

of therapy

• Repeated assessment of mood and suicide risk

• Closing therapy with a ‘Goodbye letter’ – summarizing the

therapy, with strategies and goals for future management

of interpersonal problems

• Close liaison with the GP

Table 2

3 © 2006 Elsevier Ltd. All rights reserved.

MANAGEMENT

Benefits other than reductions in repetitionAlthough research has been slow to confirm that psychologi-cal treatments are effective in reducing repetition, most of the randomized trials referred to above nevertheless showed clear benefits in other important ways: particularly in mood, hopeless-ness and problems experienced. When put together, the body of evidence now suggests substantial benefits from problem-solving or other cognitive forms of treatment. But it is gratifying that, at last, well-designed trials have shown a clear reduction in repeat-ing behaviour. When the Cochrane review of treatments after self-harm is updated shortly, perhaps there will be a summary that will set out, more clearly than has been possible up until now, how the health service should proceed in developing and implementing practicable and feasible treatments for patients who have harmed themselves. ◆

ReFeRenceS

1 National Institute for Clinical Excellence. The short-term physical and

psychological management and secondary prevention of self-harm

in primary and secondary care. London: NICE, 2004. Available at:

http://www.nice.org.uk/page.aspx?o=213665

2 Hawton K, Rodham K, Evans E, Weatherall R. Deliberate self harm

in adolescents: self report survey in schools in England. BMJ 2002;

325: 1207–11.

3 Crawford M J, Wessely S. Does initial management affect the rate of

repetition of deliberate self harm? Cohort study. BMJ 1998; 317: 985.

4 Dennis M, Evans A, Wakefield P, Chakrabarti S. The psychosocial

assessment of deliberate self harm: using clinical audit to improve

the quality of the service. Emerg Med J 2001; 18: 448–50.

5 Ministry of Health. Attempted suicide. HM(61): 94. London: Ministry

of Health, 1961.

Key aspects of cognitive therapy to prevent suicide attempts

• Identification of proximal thoughts, images and core beliefs

that were activated prior to the suicide attempt

• Development of cognitive and behavioural strategies that

might address these thoughts, images and beliefs

• Help to develop adaptive ways of coping with stressors

• Addressing vulnerability factors, such as hopelessness, poor

problem-solving, impaired impulse control, treatment non-

compliance and social isolation

• Developing and undertaking a relapse prevention task

(near the end of therapy)

• Extra sessions if the participant failed to respond adaptively

to the relapse prevention task

Source: Brown et al., 2005.22

Table 3

PSYCHIATRY 5:8 2

6 Department of Health and Social Security. The management of

deliberate self harm. Health Notice HN(84): 25. London: DHSS,

1984.

7 Kapur N, House A, Creed F et al. Management of deliberate self

poisoning in adults in four teaching hospitals: descriptive study.

BMJ 1998; 316: 831–2.

8 Head L, Campbell-Hewson G L, O’Keane V. No harm done?

Psychological assessment in the A&E department of patients who

deliberately harm themselves. J R Coll Physicians Lond 1999; 33:

51–5.

9 Hickey L, Hawton K, Fagg J et al. Deliberate self-harm patients who

leave the accident and emergency department without a psychiatric

assessment. A neglected population at risk of suicide. J Psychosom

Res 2001; 50: 87–93.

10 Bennewith O, Gunnell D, Peters T et al. Variations in the hospital

management of self harm in adults in England: observational study.

BMJ 2004; 328: 1108–9.

11 Owens D, Dennis M, Read S, Davis N. Outcome of deliberate

self-poisoning: an examination of risk factors for repetition.

Br J Psychiatry 1994; 165: 797–801.

12 Kreitman N, Foster J. The construction and selection of predictive

scales, with special reference to parasuicide. Br J Psychiatry 1991;

159: 185–92.

13 Kapur N, Cooper J, Rodway C et al. Predicting the risk of repetition

after self-harm: cohort study. BMJ 2005; 330: 394–5.

14 Royal College of Psychiatrists. The general hospital management

of adult deliberate self-harm: a consensus statement on standards

for service provision (council report number 32). London: Royal

College of Psychiatrists, 1994.

15 Royal College of Psychiatrists. Assessment following self-harm

in adults (council report number 122). London: Royal College of

Psychiatrists, 2004.

16 Deliberate self-harm. Effective Health Care 1998; 4(6). Available at:

http://www.york.ac.uk/inst/crd/ehcb.htm

17 Morgan H G, Jones E M, Owen J H. Secondary prevention of

non-fatal deliberate self-harm. The green card study. Br J Psychiatry

1993; 163: 111–12.

18 Evans M O, Morgan H G, Hayward A et al. Crisis telephone

consultation for deliberate self-harm patients: effects on repetition.

Br J Psychiatry 1999; 175: 23–7.

19 Carter G L, Clover K, Whyte I M et al. Postcards from the EDge

project: randomised controlled trial of an intervention using

postcards to reduce repetition of hospital treated deliberate self

poisoning. BMJ 2005; 331: 805–7.

20 Hawton K, Townsend E, Arensman E et al. Psychosocial and

pharmacological treatments for deliberate self harm. Cochrane

Database Syst Rev 1999; 4.

21 Guthrie E, Kapur N, Mackway-Jones K et al. Randomised controlled

trial of brief psychological intervention after deliberate self

poisoning. BMJ 2001; 323: 135–8.

22 Brown G K, Ten Have T, Henriques G R et al. Cognitive therapy for

the prevention of suicide attempts: a randomized controlled trial.

JAMA 2005; 294: 623–4.

74 © 2006 Elsevier Ltd. All rights reserved.