Violent behaviour in adolescents: assessment and formulation using a structured risk assessment tool

Summary

Teenagers often present in crisis with risk issues, mainly risk to self but sometimes risk to others. Adolescent violence is commonplace and is not just the remit of adolescent forensic psychiatry. Clinicians may lack confidence assessing risk of violence and can neglect vital areas that are essential to reduce risk. Use of structured violence risk assessments enables the multi-agency professional network to formulate a young person's presentation and their violence in a holistic way and consequently develop targeted risk management plans addressing areas such as supervision, interventions and case management to reduce the risk of future violence. Of the several validated tools developed for young people, the Structured Assessment of Violence Risk – Youth (SAVRY™) is that most used by UK-based forensic adolescent clinicians. This article outlines the epidemiology, causes and purposes of violence among adolescents; discusses types of risk assessment tool; explores and deconstructs the SAVRY; and presents a fictitious risk formulation.

LEARNING OBJECTIVES

After reading this article you will be able to:

• • identify risk and protective factors associated with violence in adolescence

• • consider the role of formulation in assessment of violence in an adolescent

• • understand how the risk assessment can be used to create a risk management plan.

Adolescents often present in crisis to hospital and mental health services (McNicholas 2022). Although clinicians may be well versed in assessing and managing a young person's risk to self, when someone presents with a risk of violence to others, this can induce fear and have a paralysing effect on professionals. Indeed, it might be understood that an important function of the young person's externalising behaviour and attitude is to paralyse those around them, and the clinician's reaction is entirely in keeping with this.

A structured violence risk assessment tool allows the clinician to break down and start to categorise risk that might otherwise be more nebulous in nature. Although reservations have recently been expressed about the validity of existing risk assessment tools developed for use with children and adolescents, the Structured Assessment of Violence Risk – Youth (SAVRY™) and the Youth Level of Service/Case Management Inventory (YLS/CMI) remain the most widely used globally in the absence of any better alternative (Koh 2020; Senior 2021). Understanding the core principles of the SAVRY can equip a non-forensic clinician with a structure to create a holistic care plan to reduce future risk (Lodewijks 2008).

A multidisciplinary approach is advised, as adolescents presenting with violent behaviour can induce complex feelings among clinicians. Countertransference (the professional's reactions and feelings towards the patient) can play a role. The professional's own history and conflicts may be part of this, but it can also involve the patient's feelings towards the professional and what they may be doing to the professional, either consciously or unconsciously. Some patients can provoke anger, fear or disgust in the professional, which may elicit punitive or sadistic responses; others can elicit sympathy and present themselves as innocent victims. Professionals need to be wary of a desire to punish or rescue, of an overly harsh or lenient approach (Morgan 2007).

Professionals need to be alert to their reactions and seek assistance or support, especially if these are recurrent. This may take the form of supervision, case discussion within the multidisciplinary team or seeking a second opinion.

Another factor to consider is any current strain on healthcare resources. Awareness of the imperfections and realities of available resources makes structured risk assessment more valuable, as it allows the clinician to objectively identify the young person's needs, allowing for transparent discussions about available resources and the need to seek funding for any shortfalls in available treatment. Without a mature understanding of the health and social care system, unrealistic expectations and accusations of a failure to cure can lead to scapegoating. This can reflect a blindness to the complexity of the situation and denial of the reality that all teams, clinicians and systems have strengths and weaknesses (Blumenthal 2018).

An exploration of violence

What is violence?

Violence is defined in the SAVRY manual as ‘an act of battery or physical violence that is sufficiently severe to cause injury to another person or persons, regardless of whether the injury actually occurs, […] or a threat made with a weapon in hand’ (Bartel 2002).

Violence in youth

Violent behaviour needs to be understood in context, which includes considering the age of the person. Studies (e.g. Teymoori 2018) show that the highest prevalence of violence or aggression is seen in those aged 2–4 years but that the behaviour is not as serious as violence seen in adolescents or adults. Although children can be very aggressive (hitting, kicking, shoving), this may not lead to injury because of their size. Parents teach their children how to manage aggression, and experiences of poor parenting, such as abuse, neglect, coercive parenting styles, antisocial modelling and minimal boundaries, may lead to an increased risk of violence (Tremblay 2004). It has been estimated that oppositional defiant disorder develops into conduct disorder in approximately 30% of children (Loeber 2000). In approximately 40% of adolescents with conduct disorder, there will be a progression to antisocial personality disorder (Zoccolillo 1992).

The prevalence of violence is high in adolescents compared with that seen in adults. It is reported that 20–25% of adolescent males and 4–10% of adolescent females engage in one or more acts of serious violence annually (Moffitt 2001). Non- violent criminality (e.g. shoplifting, vandalism) is more common than violence in both adolescents and adults but adolescence is the period when violence towards others is most likely.

The highest risk of initiation of serious violent behaviour is seen in those 15–16 years of age, with ongoing high rates of participation in violence at 16 and 17 years of age and a drop as the young person becomes an adult. Longitudinal studies show that 80% of those who are violent during adolescence will terminate violence by age 21 (Moffitt 1993).

Early onset of violence increases the likelihood of future violence and criminality. Mroczkowski (2021) has described two main patterns of development of violence: late onset and early onset. Late-onset violence begins after puberty, accounting for 70% (s.d. = 15%) of serious violent offenders, with 2% of those going on to violent careers lasting more than 2 years. It is associated with poor social support, association with peers with conduct disorder and involvement with gangs. Early-onset violence begins before puberty; although accounting for only 30% (s.d. = 15%) of serious violent offenders, a much greater proportion (13%) are thought to go on to violent careers lasting longer than 2 years (Ash 2019).

Violent behaviour in adolescents and adults show differences in diagnoses, epidemiology, behaviour patterns and treatment (Tremblay 2004). As mentioned above, violence is much more common in adolescents and accounts for a higher proportion of all deaths in that age group, and many violent careers are adolescent limited. However, in all cases, the first episode of serious violence rarely occurs in adulthood. It usually occurs in adolescence and sometimes childhood. (Tremblay 2004). Psychotic disorder plays a much lesser role in adolescents who are violent than in adults. Adolescents are more likely to enact violence as part of a group (Ash 2019). Violent behaviour is seen more in males than females (Odgers 2005). Intimate partner violence during adolescence has been found to be a significant global public and social health problem (Desmarais 2012). However, currently available violence risk assessment tools appear to have limitations in evaluating this type of violence and there have been calls for a specialised risk assessment tool for adolescent intimate partner violence to be developed (Shaffer 2022).

Causes of violence

Violence is contextual, with individuals making decisions to commit a type of violence against a specific victim at a specific time. People will and often do decide not to commit violence. However, the decision-making itself may be unconscious (Morgan 2007), especially if related to previous trauma, or very rapid if related to conditions that exacerbate impulsivity.

Numerous research studies have concluded that many factors and their interplay lead to an increased risk of violent behaviour in children and adolescents (AACAP 2017). These factors include:

‘Previous aggressive or violent behaviour, being the victim of physical abuse and/or sexual abuse, exposure to violence in the home and/or community, being the victim of bullying, genetic (family heredity) factors, use of drugs and/or alcohol, combination of stressful family socioeconomic factors (poverty, severe deprivation, marital breakup, single parenting, unemployment, loss of support from extended family), brain damage from head injury’ (AACAP 2017).

Violence as communication

Fundamental to any assessment is an understanding of the meaning of the violence. Violent behaviour can be meaningful and a form of communication for those struggling to communicate their emotional or physical needs. The meaning or drive to communicate may be unconscious. The violent act could be viewed as behaviour that has replaced thinking. The violent person's mode of relating with others is characterised by action, not thought. Where therapeutic interventions are possible, the task of therapy is to insert thought between impulse and action (Morgan 2007).

Risk assessment in mental healthcare

What is risk assessment?

Risk assessment is the process of understanding a risk, to limit its potential negative impact. An understanding of the interaction of a number of factors is required, and this can lead to the assumption that risk assessment is uncertain, complex, inferential and contextual (Hart 2016).

• • Uncertain: A risk is a hazard or threat that is incompletely understood and therefore that can be forecast only with uncertainty.

• • Complex: Risk incorporates notions of the nature, severity, frequency, imminence, and likelihood of harm – not just the probability.

• • Inferential: Risk does not exist physically, but rather reflects the perception of a potential or possible future.

• • Contextual: Risk exists in and is dependent on a specific situation or social-physical environment.

Types of violence risk assessment

There are three methods for assessing risk (Hales 2022). The first is unstructured clinical assessment, where the clinician gathers information they think relevant, combines it and processes it to come to a conclusion. The second is standardised or actuarial assessment, which involves collecting a standardised set of data about an individual, which is used to classify the person into one of several groups. Each group is considered to present a different level of violence risk. The third is structured clinical assessment, which is an attempt to combine the best aspects of clinical and standardised (or actuarial) approaches to risk assessment in forensic populations. It requires clinicians to gather specified information (based on known associations with violence risk) and consider other information about the individual before finally drawing conclusions about their level of risk.

The advantages and disadvantages of the three methods are outlined in (Table 1). Relative or conditional risk judgements are more useful than absolute or probabilistic ones (Hales 2022). An example of a relative or conditional risk judgement may be that a young person poses a low risk of future violence with adequate parental supervision, engagement in education and having treatment for attention-deficit hyperactivity disorder (ADHD) but a moderate to high risk without these elements of the treatment plan being enacted.

TABLE 1 Risk assessment methods: advantages and disadvantages

Source: Hales & Lewis (2022).

Risk and protective factors

Both risk and protective factors tend to be personal characteristics or environmental conditions that predict onset, continuity or escalation of risk of violence. An accumulation of risk factors is more important than individual factors. Risk factors are usually present in clusters, for example physical abuse, poor parental management, parental criminality and peer delinquency.

Limitations on undertaking risk assessments with young people

Some clinicians have reservations about the practice of undertaking risk assessments of young people at all, because of the possible impact of stigma. They argue that an adverse assessment can result in a young person being stigmatised as ‘dangerous’ for the rest of their life.

To reduce the risk of a young person becoming stuck in forensic services, most community-based forensic child and adolescent mental health services seek to share the individual's care with the local child and adolescent mental health team (Hindley 2017).

This collaborative working allows for the acknowledgment that many of the young people referred to adolescent forensic services have committed serious crimes involving high levels of violence or risk to others but are often themselves victims of abuse or violence; they have mental and emotional health difficulties and they need the support of community health services to maintain them in the community.

For this reason, it is appropriate for the mental health professionals dealing with them to work together both to assess their future risk to others and to develop plans to manage that risk. Although not all risk can be eliminated, it is safer to identify risk and make a risk management plan rather than be unaware of it.

Assessing violence risk with adolescents compared with adults

When assessing violence risk in adolescents rather than in adults (Hales 2022), it is essential to be cautious when interpreting risk assessments. First, adolescence is a time of rapid developmental change. This process of change can also extend to the dynamic risk factors exhibited by a young person. Such factors may increase or decrease in significance over the period of adolescence. As a result, the level of risk that a young person presents to others may also increase or decrease. Second, much of the research supporting the predictive validity of violence risk assessment in adolescents is based on follow-up data of less than 3 years. For adults, data exist for longer follow-up periods. As a result of these points, it is essential to note that estimated violence risk should be re-evaluated:

• • after a period of, at most, 2 years, or

• • following significant social, environmental, familial, sexual, affective, physical or psychological change.

Factors to consider when conducting risk assessments

Diverse populations

When undertaking a structured assessment of violence risk, it is important to acknowledge that, despite the benefits over unstructured clinical risk assessment, various forms of bias can continue to affect accurate assessment. Bias can involve individual professionals, organisational systems and the specific tools used. This becomes particularly relevant when assessing young people from minority ethnic groups (Perrault 2017) and those with neurodiverse profiles.

The body of research on this topic is growing, and evidence remains mixed as to predictive validity and fairness in the most commonly used tools (McGowan 2011). It appears that the SAVRY (the structure of which is outlined later in this article) does have predictive validity across different ethnic groups (Chu 2016), but given the vast overrepresentation of some groups in youth justice services, it is vital that risk assessors are mindful of good practice in this area as the evidence base grows.

A strength of the SAVRY tool is the inclusion of dynamic contextual risk factors and protective factors, which are generally thought to improve fairness of risk appraisal for diverse groups. Some researchers highlight that the ratings of some factors (e.g. commitment to school, negative attitudes, community risk) serve to unjustly increase the violence risk scores in those from certain ethnic or neurodiverse groups (Muir 2023). They argue that there is a lack of acknowledgement of underlying systemic disadvantage and discrimination suffered by those young people. These underlying societal factors may represent the ‘cause of the cause’ of higher ratings in these ‘individual’ risk factor items. Irrespective of any impact this has on predictive validity, it is vital that an appreciation of systemic inequalities feeds into culturally sensitive risk formulations which follow on from rating.

Similarly, some have argued that those risk factors involving adverse childhood events (Webb 2022) (e.g. early caregiver disruption) erroneously cause higher risk ratings in those with developmental disorders, as these populations are known to experience higher rates of traumatic events. It does not necessarily follow that these traumatic events lead to higher risk of violence in these groups, and further research is needed in this area.

Risk assessment tools are rarely well validated in female youth populations, so it is important to acknowledge aspects of intersectionality – young women from minority ethnic backgrounds and neurodiverse young women are likely to be the populations in which risk assessment tools are least well validated.

Neurodiversity should be taken into account in any risk assessment. ADHD is considered and rated in the SAVRY. Although empathy is also considered, other cognitive processes of autism spectrum disorder (ASD) are not. Therefore, an additional tool to support understanding of risk in those with ASD is needed. The Framework for Assessment of Risk and Protective Factors in Autism Spectrum (FARAS) (Al-Attar 2018) has been developed to help clinicians in considering the different needs and risks in people with ASD. This is an adjunctive tool to use alongside the SAVRY to support the development of the formulation. It has not yet been officially validated, although in our experience, clinicians find it useful.

Attachment, trauma and loss

The development of affect regulation, impulse control, empathy and the capacity to reflect or mentalise requires normal secure attachment (Yakeley 2018). Many violent individuals experience early trauma and loss and violence may be understood as a failure in adequate caregiving or ‘faulty attachments’. A defensive response to feelings of shame and humiliation, which have their roots in disorders of attachment, may result in violence (Yakeley 2018).

Practical tips for the risk assessment and formulation

Box 1 outlines the risk assessment process and Box 2 suggest questions to be considered during a risk assessment interview. This section summarises tips for clinicians conducting a risk assessment interview and preparing a formulation.

• • Listening is vital to understanding what is going on. It seems simple but it can be very hard to listen to someone in distress or who has done something that provokes negative feelings. The young person may be quiet and withdrawn or you may feel helpless or find the situation upsetting to think about. You may need to build rapport. Ensure that you have the energy and the support needed to be able to listen. Make time to listen properly with minimal distraction and no interruptions. Let the young person see that you take their problems and worries seriously; do not blame them or belittle them (Graham 2005). Strive to draw SAVRY evidence from broad sources, particularly from professionals such as mentors and youth workers, who may not traditionally be involved in risk assessments but often have good knowledge of the young person's attitudes and experiences.

• • Create an atmosphere where those contributing information feel able to give their opinions on evidence and predominant risk narratives, or to present contrary evidence.

• • Use risk assessment aids in relevant populations, for instance the FARAS tool in young people with autism.

• • Work to challenge assumptions about the young person, their offending and their capacities both to offend and to engage in risk management strategies (avoiding either over- or underestimating risk).

• • Examine the reasons behind high rating factors, including systemic factors such as the reasons for permanent school exclusion.

• • Adapt the language in risk formulations and risk management plans to minimise cultural insensitivity; for example ‘community disorganisation’ might be better described as ‘community disadvantage’.

• • Emphasise proactive evidence-gathering for protective factors, as such evidence is often less full.

• • The risk formulation is key to communicating balanced and culturally nuanced expressions of risk, which will in turn maximise the utility and real-world applicability of risk management plans.

• • A risk assessment should always lead to a risk management plan to monitor, mitigate and reduce future risks.

BOX 1 Steps to risk assessment

• • Gather information

• • Identify risk factors

• • Develop scenarios of different risks in different settings

• • Plan management strategies

• • Document judgements.

BOX 2 Interviewing on issues of violence

Consider each of the following when conducing a risk assessment interview

• • Did the aggression result in injury?

• • Was a weapon used?

• • In what context/setting did the violence occur?

• • What is the young person's perception of triggers

• • What was the young person thinking at the time of the incidents?

• • Had drugs or alcohol been consumed?

• • Have drugs/alcohol precipitated other incidents?

• • Psychotic symptoms? Medication?

• • Who was the victim? Relationship with the victim?

• • Purpose/meaning of violence?

• • Young person's and assessor's views regarding patterns to violence

• • Cues to violence? Has the young person ever refrained from violence?

• • What responses would the young person suggest to prevent future violence?

  (Borum 2000)

The Structured Assessment of Violence Risk in Youth (SAVRY)

The SAVRY™ is a violence risk assessment tool used with adolescents in the UK, particularly in medium secure psychiatric units and forensic child and adolescent mental health services. Clinicians require training in its use.

The SAVRY has 24 items in three risk domains (historical risk factors, social/contextual risk factors and individual/clinical risk factors). These items are the culmination of research and the professional literature on adolescent development, including that on violence and aggression in youth. As well as the 24 risk factors, the SAVRY also includes six ‘protective factor’ items that are rated as either present or absent. This is an innovative feature. The value of considering protective factors has been increasingly recognised (O'Shea 2015) in violence risk assessment and management in all age groups.

The SAVRY is validated for use with male and female adolescents between 12 and 18 years of age. Once training is completed, a clinician can use this assessment tool, along with the SAVRY manual (Bartel 2002), to help in structuring an assessment. A good structure ensures that the important factors will not be missed and can be emphasised when formulating a final professional judgement about a youth's level of risk and producing a risk management plan.

Not all risk factors are given the same weight, but all are considered and balanced using a rating structure laid out in the manual.

The SAVRY's historical risk factors (Box 3) relate to past behaviour or experiences and they have been shown to be associated with violence risk recidivism in juveniles. They are generally static and not subject to change. The social and contextual risk factors (Box 4) consider the influence of interpersonal relationships (i.e. with peers and family) and connection to social institutions and the environment. The individual/clinical risk factors (Box 5) focus on the young person's attitudes and key aspects of their psychological and behavioural functioning.

BOX 3 The historical risk factors included in the SAVRY

• • History of violence

• • History of nonviolent offending

• • Early initiation of violence

• • Past supervision/intervention failures

• • History of self-harm or suicide attempts

• • Exposure to violence in the home

• • Childhood history of maltreatment

• • Parental/caregiver criminality

• • Early caregiver disruption

• • Poor school achievement.

  (Borum 2000)

BOX 4 The social and contextual risk factors included in the SAVRY

• • Peer delinquency

• • Peer rejection

• • Stress and poor coping

• • Poor parental management

• • Lack of personal/social support

• • Community disorganisation.

  (Borum 2000)

BOX 5 The individual/clinical risk factors included in the SAVRY

• • Negative attitudes

• • Risk-taking/impulsivity

• • Substance-use difficulties

• • Anger management problems

• • Low empathy/remorse

• • Attention-deficit hyperactivity difficulties

• • Poor compliance [adherence to treatment/interventions]

• • Low interest/commitment to school.

  (Borum 2000)

Violence is not inevitable, even in the face of one or more risk factors:

‘Just as there are risk factors that increase the likelihood of violence, there are individual and contextual protective factors that can reduce the negative impact of (i.e. buffer) a risk factor or otherwise act to diminish the probability of a violent outcome. Protective factors may mitigate the appraisal of risk or, in some circumstances, may be integrated in treatment or intervention planning to enhance or facilitate risk reduction efforts’ (Bartel 2002).

Box 6 shows the protective factors chosen for inclusion in the SAVRY. These are consistent with the conceptualisation of protective factors by Rutter (1987).

BOX 6 The protective factors included in the SAVRY

• • Prosocial involvement

• • Strong social support

• • Strong attachments and bonds

• • Positive attitude toward intervention and authority

• • Strong commitment to school

• • Resilient personality traits.

  (Borum 2000)

The SAVRY lends itself to intervention and risk management, and attention paid to dynamic changeable risk factors in young people's lives.

Risk formulation

The aim of the formulation is to explain the individual's problems and symptoms, with the caveat that it is always subject to empirical testing via positive risk-taking and revision as new information becomes available (Sturmey 2011). Formulation provides a map to guide treatment (Eells 2007); its primary purpose is to help the clinician develop and implement a treatment plan that will lead to a successful outcome, allowing anticipation of future events. For example, an adolescent exhibiting violent behaviour when not supervised at night and intoxicated while mixing with a local gang would probably benefit from parental support to allow for appropriate supervision, an intervention targeting substance misuse and opportunities for prosocial activities with an alternative peer group.

Taking a holistic approach is essential; often mental health is only one part of the jigsaw. If the young person and family can contribute to the formulation and risk management plan this engagement will allow for a better understanding of the origins, development and maintenance of the problem. Diagnosis, although the basis of evidence-based psychiatry, is not by itself a sufficient guide to treatment selection in these complex cases. A holistic formulation enables a multi-agency care plan to be developed, tailoring the treatment to address the individual's circumstances.

Formulation and risk management planning: a fictitious example

James is a 14-year-old White male who was referred for violence risk assessment after being suspended from school because of a violent altercation. The index incident occurred when James and two friends were truanting from school. They came across a boy of a similar age and began to taunt and threaten him. James then punched the boy. A friend tried to stop the fight, but James became angrier and attempted to hit his friend. The friends attempted to restrain James, but he ran away. The victim's mother informed the police, who attended the school the next day, but James was not charged or arrested. However, he was suspended from school because of concerns about the safety of students and teachers. The purpose of the assessment is to decide whether James can safely return to school.

School records indicated that James had physically threatened students on two occasions and had hit a teacher when she attempted to stop him walking out of a class. He had been suspended for fighting in junior school. James said that he had been in numerous fights over the past year and enjoys the ‘rush’, feeling he is very good at fighting and enjoying the feeling of power when others are fearful of him. As regards the recent fight, he felt that the other boy ‘had it coming’ and he is angry that everyone seems to be ‘making a meal out of things’.

James changes friends frequently, falling out over minor things. He mixes with other boys who truant, and they often roam the neighbourhood vandalising property and stealing things. He is not involved in sport or community activities and frequently smokes cannabis.

James is poor academically; he does not complete homework and has a 34% attendance record. There is little parental supervision, his father is not in contact with the family and there is violence in the home from an older brother. James has not had any contact with mental health services but has experienced self-limiting paranoia following ingestion of cannabis.

In approaching this case as a health practitioner, it is worth considering the need for a multi-agency approach to manage James's risk. The SAVRY is helpful in considering risk from both an individual and a systemic perspective. James appears to be using violence to threaten and ‘control’ others, even the friends he was truanting with – the violence appears ‘instrumental’ in nature, but there may be emotional drivers that might be ameliorated with an appropriate health or social intervention. A comprehensive objective risk assessment might help focus services on James, especially an assessment that will probably be shared with the school and professional network. James appears to be following a criminogenic pathway, with little parental supervision and reports of violence in the home – the family might benefit from the support of social services, and we must consider James's vulnerability and any safeguarding issues, as these might be driving his current trajectory.

To understand risk, the SAVRY helps to add detail in certain important domains, such as James's history of violence (static factors) and current life/health situation (dynamic factors). These elements add to the picture, but more importantly allow us to start to structure a risk management plan that might allow his safe return to school, or conversely might illustrate that the risk of return is too high. A critical part of the SAVRY is our consideration of protective factors, and in the case of James it may be that he would benefit from a number of interventions that could mitigate risk and allow a safe return to school. These might include: a named person at school he can confide in, with regular daily check-ins; use of a ‘timeline’ at check-ins, to see any potentially unsettling life events; engagement in prosocial group activities in and after school, such as music or sports; support with his academic work; and a collaborative school–pupil ‘contract’ regarding James's conduct should he be keen to return to school. The school might be advised to engage in regular multi-agency meetings with James's wider professional network, such as health and social care services, and might wish to instigate an Education, Health and Care plan to that end. James might benefit from engagement with a drugs worker and, if possible, a community-based youth mentor. The school should contact the police if other violence occurs.

Reflection

Violent behaviour is common among adolescents and most outgrow it. Some are brought to the attention of child and adolescent and forensic mental health services for assessment and treatment planning. As mentioned above, there is a concern that an adverse assessment can label a young person as ‘dangerous’ for life. James, the subject of our formulation, was not arrested as he did not meet the threshold for police intervention, but perhaps we might reflect here on our societal reluctance to ‘criminalise’ young people. We might consider the fact that many useful community resources are scarce, but potentially available, should James be engaging with a Youth Offending Service – an argument to counter our moral anxiety in ‘criminalising’ him. Indeed, James appears to be criminalising himself, and as a professional network psychiatrists require the appropriate resources to manage his risk in the community – this may necessitate involvement of the Youth Justice system and therefore the police must be considered if his violence should continue or escalate, however unsavoury that might feel. Future articles need to look at the development of structured professional judgement tools (SAPROF-YV, EARL, START-AV) and their use with the SAVRY and delve more deeply into the relationship between the various disorders and violence.