Self-harm itself is not a medical diagnosis but is a term used to describe an act, for which there are many different meanings and functions for individuals. There is a range of disorders associated with self-harm and interventions should address underlying conditions that may perpetuate the behaviour. For some individuals, motivations behind self-harm may be multiple and can change over time. Treatment options should be considered in view of the individual’s specific needs, their vulnerability and protective characteristics and with clinicians seeking a full picture of the reasons behind the self-harm act. These may include social circumstances, the immediate environment the individual will be returning to, physical health, interpersonal relationships and coping styles and underlying psychiatric conditions, including alcohol and drug misuse. Interventions may aim to deal with one or more of these contributing factors but should take into account the complexity behind self-harm behaviour. Throughout this chapter we use the term ‘self-harm’ to describe intentional acts of self-poisoning or self-injury irrespective of motivation [1], that is, the acts may or may not involve suicidal intent. We will consider the importance of acting early to support people who self-harm, summarise the current recommendations for treatment and review the evidence that the guidance is based on.

It is estimated that there are around 220,000 presentations to Accident and Emergency (A&E) departments in England (equivalent to emergency rooms or departments elsewhere) each year [2] and recent estimates of annual rates indicate 300–500 individuals per 100,000 in England will attend hospital having self-harmed [3], rising to 1400 per 100,000 amongst adolescent females [4]. People who present to hospital with self-harm may represent the visible ‘tip of the iceberg’ of the problem of self-harm in the general population. Many people will not present to services following self-harm and figures based on self-reporting suggest that around 1 in 40 adults have self-harmed in their lifetime [5]. Self-harm attendances to general hospitals are not routinely audited but both short- and long-term monitoring has been set up in a number of hospitals [3, 6] allowing the investigation of trends and providing an overview of how self-harm is managed in hospitals in England. The majority – up to 80% – of hospital presentations are acts of self-poisoning, around 15% self-cutting and the remainder involve other methods, predominantly more violent methods such as attempted hanging and jumping from a height or a combination of these methods [3, 7]. People who present to hospital having self-harmed are not always offered a psychosocial assessment by a mental health specialist [6], yet hospital attendance is an opportunity for services to intervene [8], with general hospitals existing as an important base for secondary prevention of self-harm. Services for self-harm patients attending hospital vary, with some providing mental health liaison teams situated within the A&E departments available 24 hours a day, and others with more limited access to mental health specialists [6]. Hospitals have an important role not only in treating the self-harm episode itself but also in identifying the need for follow-up care to address existing problems that may not have otherwise reached the attention of services, such as depression and alcohol misuse.

All acts of self-harm serve as a warning sign for an increased risk of suicide, but those who repeat self-harm are at further elevated risk [9]. When we compare the outcomes of people who repeat to those who do not, we see that those who have attended hospital more than once tend to have poorer outcomes. They have more than twice the risk of suicide than people with a single episode, with this increase more pronounced amongst females where the risk is threefold [9]. In terms of the possible psychological process, repeat self-harm may operate on a trajectory with multiple episodes increasing individuals’ tolerance to the negative elements of the experience whilst also resulting in cognitive sensitisation [10, 11].

There are various ways in which interventions for self-harm have been evaluated for effectiveness. Randomised controlled trials (RCTs) have been conducted on a range of interventions, including, brief psychological and behavioural therapies, postal and telephone ‘contact-based’ interventions and pharmacological therapy [17, 18]. These tend to be conducted in groups of patients with particular characteristics that make them suitable for the specific treatment. It is possible, however, to pool results from trials with similar characteristics to attempt to estimate efficacy of similar treatments, so-called meta-analysis. We can also use information collected from self-harm attendances to hospitals to detect patterns in attendances and clinical management. Repetition of self-harm points to ongoing distress and unresolved problems. Incidence of repeat self-harm, therefore, is often used as an outcome to measure efficacy of interventions. However, reasons behind self-harm can be numerous and complex, and information gleaned from qualitative discussions with service users and clinicians provides valuable insight into the mechanisms through which interventions have an effect and can uncover more subtle changes in behaviour following interventions.

Current national guidance in England [19] recommends that a full psychosocial needs assessment and mental state examination should follow each hospital presentation involving self-harm and highlights the importance of emergency department and local mental health services to jointly plan liaison psychiatric services available 24 hours a day. The assessment should be carried out by professionals, who may include psychiatric nurses, social workers and psychiatrists, who are trained to carry out assessments specifically for people who have self-harmed and where adequate supervision is provided. People who self-harm repeatedly may not experience the same motivation behind each act. An assessment at each episode, therefore, increases the likelihood that appropriate help will be offered. Research using observational data from hospitals where self-harm attendances are monitored has demonstrated that receiving psychosocial assessment upon presentation to hospital can improve outcomes for certain groups of patients [20–22]. Whilst these findings are not generated from controlled conditions, as is the case in RCTs, they are informative because they report real-world, population-level scenarios. Patients who discharged themselves from hospital before any assessment could be carried out had three times the rate of repetition compared to those who were assessed [22]. Recent estimates are that around six in ten patients who attend hospital with self-harm receive a psychosocial assessment by a mental health specialist [6, 20], although this proportion varies widely by hospital [6]. Studies have found that specialist psychosocial assessment is associated with reduced risk of repetition in centres where higher proportions of patients are assessed [21]. This suggests that when mental health specialists in general hospitals assess more people, presumably by including more people deemed lower risk, the benefits in terms of the reduced number of repeat attendances are greater.

Of course data from observational studies cannot prove causation but these findings indicate that for some individuals the assessment itself may be beneficial. This effect may operate through a number of mechanisms, and has been explored by interviewing service users about their experiences. Service users placed value on psychosocial assessments conducted in a manner that promoted hope and acceptance. Positive experiences of assessment had the potential to promote future therapeutic engagement with services. However, assessments where service users felt staff were judgemental or did not instil hope for change, or where arrangements for follow-up care did not materialise, could disengage service users and counteract feelings of hope [23]. Interestingly, a recent RCT of ‘therapeutic assessment’, consisting of a 30-minute intervention based on cognitive analytic therapy alongside standard psychosocial assessment, found that adolescents receiving the intervention had higher levels of attendance at follow-up appointments than those receiving standard assessment alone [24].

The current evidence for what might be effective in treating people who self-harm has recently been systematically reviewed and incorporated into guidance published in the United Kingdom by the National Institute for Health and Clinical Excellence (NICE) [17]. There is no single treatment that is known to be especially effective at treating self-harm, though recommendations have emerged nonetheless. Most interventions aim to identify and address the underlying problems that lead an individual to self-harm. Whilst there is no single treatment, there are overarching principles of care that have been set out to guide health care professionals on the best approach to take when working with people who self-harm. Central to working effectively with people who self-harm is building a trusting, supportive and engaging therapeutic relationship and aiming to maintain continuity of relationships. Interventions should involve working collaboratively with service users, involving them fully in decisions relating to their care and encouraging the development of autonomy and independence. Because of the stigma associated with self-harm [25] and the negative experiences of services reported by some service users [17, 23], health care professionals should ensure services are delivered in a nonjudgemental and sensitive manner. When service users were given more responsibility for their own care alongside a genuinely compassionate relationship with professionals [26], they were more likely to view services positively. Service users’ accounts of recovery from self-harm and its associated conditions can offer valuable insight into what works. Common threads running through people’s accounts of stopping self-harm include accessing services that fit with their needs, establishing therapeutic relationships (either with individual health professionals, statutory or non-statutory services, local community groups or online support networks), gaining an understanding of the reasons they self-harmed, addressing psychological symptoms, building autonomy and being able to use alternative coping strategies [17, 27, 28]. Recovery often correlated with increased levels of engagement with life pursuits and increased quality of interpersonal ties and use of social support [2].