Article Text

Download PDFPDF

Collaborative risk assessment in secure and forensic mental health settings in the UK
  1. Sarah Markham
  1. Department of Biostatistics and Health Informatics, King's College London, London, UK
  1. Correspondence to Dr Sarah Markham; sarah.markham{at}


Collaborative risk assessment and management have been recommended in health policy for over a decade. We consider the nature and need for collaborative risk assessment and management between patients and clinicians in secure and forensic mental health settings in the context of shared decision making and personalised care in the UK. We examine the extent to which policy and recent initiatives have influenced the embedding of such practice in services through consideration of the evidence provided by research and the Commissioning for Quality and Innovation framework, and conclude that there is a need for further improvement.

  • forensic psychiatry
  • mental health services
  • risk
  • risk assessment
  • risk factors

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Patient-centred care has been defined as the provision of care that is respectful of, and responsive to, individual patient preferences, needs and values, and ensuring that patient values guide all clinical decisions.1 It has become the focus of policy documents and mission statements, including the National Health Service (NHS) Long Term Plan1. The Recovery Model is predominant in mental health service policy as is recognition of the importance of person-centred practice and the positive impact it can have on outcomes for patients.2 Recovery-oriented mental health policy and practice aim to enhance the agency of the individual, prioritising self-determination, strengths-based practice and collaborative working.2 The NHS Long Term Plan commits to making personalised care ‘business as usual’ for 2.5 million people over a period of 5 years through a range of personalised care packages.1 The Universal Personalised Care model sets out a comprehensive model of personalised care with shared decision making (SDM) as one of its six key parts.1 The practice of SDM is usually associated with models of practice used in medical and surgical healthcare settings and not mental health settings in which patients are detained under the Mental Health Act (MHA).

The extent to which SDM can be implemented especially with patients in secure and forensic mental health settings is questionable due to issues such as patients’ insight, mental capacity and risk. In the context of forensic mental health services, the concept of risk refers to the harm individuals with a mental disorder (including neurodevelopmental disorders) pose, or have posed, to others, where that risk is usually related to their mental disorder.3 Lack of insight about illness and the consequences of refusal to take psychotropic medication may lead to violence and other forms of risk. These concerns are essentially unique to psychiatry and are not generally seen in other illnesses. Nevertheless, under the umbrella of the Recovery Model, collaborative risk assessment and management have been recommended in health policy for over a decade and may be a means to optimise SDM in such settings. However, there is limited research or information regarding the embedding of such practices in secure and forensic inpatient services. In the following sections of this paper, I will set out the thesis that collaborative risk assessment and management practices in forensic settings may be suboptimal and do not sufficiently implement SDM and mental health recovery principles.

Shared decision making

According to policy and other directives, SDM is to become paradigmatic in healthcare, including mental health services. In theory, the ethos of medicine has shifted from paternalistic, clinician-directed care towards a great emphasis on patient engagement, autonomy and choice, in which the clinician shares information and gives advice, thereby supporting patients to make informed decisions.4 SDM is a process by which clinicians and patients make decisions together using the best available evidence about the likely benefits and harms of each option, and where patients are supported to arrive at clinically informed preferences.4 SDM can be framed as moral praxis in which commitment from both parties, both clinicians and patients is crucial, with the caveat that in psychiatry this may be dependent on the nature and severity of patients' clinical symptoms.

It has yet to be seen whether person-centred care will lead to actual improvement in the provision of healthcare. There is evidence that SDM is not reliably implemented and SDM training often unavailable.5 For instance, it has been observed that patient involvement in care planning is rarely implemented despite policies advocating it.6

Shared decision making in the context of mental health and risk management

The final report of the Independent Review of the MHA recommends that ‘shared decision-making between clinicians and patients should be used to develop care and treatment plans and all treatment decisions as far as is practicable’ and that ‘statutory advance choice documents should be created that enable people to make a range of choices and statements about their inpatient care and treatment’.7 In the context of individuals suffering from cognitive disabilities, including those caused by mental disorder, the SDM process is a both a rational knowledge-based process and a relational one, with the balance between allowing the patient autonomy and imposing restriction determined by the vulnerability and capacity of the patient.

The establishing of a mutually respectful and trusting relationship is an essential part of SDM; however, whereas the clinician can consistently maintain a mutually respectful and trusting relationship, a psychiatric patient may only be able to reciprocate when their illness is under control. Once certain symptoms (especially suspiciousness and paranoia) develop, the patient’s ability to trust their clinician may be compromised.

Collaborative risk assessment and management

In collaborative risk assessment and management, the patient is actively involved in the actual formulation of the risk they pose and associated management strategies.8 According to the Department of Health (2007), ‘Risk assessment and management need to become more open, more transparent with service users and staff working collaboratively together. This is particularly important in forensic and high-risk settings, where recovery is just as important a principle as it is in any other part of the mental health service’.7 The NHS Long Term Plan describes a fundamental shift towards recognising that people who use health services can also help solve problems and take control.1 From a human rights perspective, it is important that secure and forensic patients are included in this.

The Department of Health has recommended that the risk assessment and management process should be explained to patients as soon as possible and this could take the form of psychoeducation.8 Psychoeducation might include defining risk, risk assessment (including the various types of risk assessment) and risk management (including the difference between risk and protective factors) and explaining how they relate to care and treatment planning.8 Psychoeducation should also involve strengthening an individual’s coping skills, communication and problem-solving abilities in the context of risk, safety and beneficence.8

Rationale for collaborative risk assessment and management

There is a clear rationale for integrating an individual’s self-assessment of risk into clinical risk assessment. The patient has self‐awareness and knowledge to inform the risk assessment and is ultimately the arbiter of whether or not they enact risk. The patient has unique access to their own internal mental states which in itself confers an advantage in terms of risk estimate and prediction.9 Furthermore, in assessing their own risk levels and associated intervention needs, patients are empowered to engage and take responsibility for their own actions. This process in itself is a motivator for change.10 Furthermore, psychological research has provided evidence that an individual will make a greater commitment to progress if there is a strong sense of engagement and responsibility.11

Risk assessment is a major task of mental health professionals. However, it is still only one of several considerations mental health professionals face in clinical practice. Of equal clinical importance are the evaluation of mental disorder, need for treatment, treatment adherence, risk of victimisation and self-neglect,12 hence the importance of the implementation of SDM being optimised in mental health settings, including risk assessment and management.

Positive risk management acknowledges ‘risk can never be completely eliminated’; and that risk management plans will contain decisions associated with risk.8 It is suggested that individuals should be supported to take reasonable risks, taking into autonomy, well-being and choice.8 It has also been suggested that when an individual’s strengths are acknowledged, in addition to their vulnerabilities, and strategies to address problems are built around their adaptive capabilities, the individual will feel more able to cope, resulting in more effective risk management.13

The concept of risk and a perceived need for risk assessment and management dominate the culture and clinical practice in secure and forensic mental health hospitals. The implications for patients include the nature and extent of the qualification of their human rights: leave, length of stay and discharge planning. It is therefore crucial that patients are involved collaboratively in risk assessment and safety planning, not merely in terms of their own treatment and care, but also in the wider context of development and revision of service security and other risk-based policies and practice.

Risk assessment processes which do not adopt a collaborative approach position the patient as a passive recipient of judgment and management strategies. Disempowerment of the patient in the risk assessment and management process may result in a failure to foster a sense of responsibility and positive agency, thereby potentially acting to enhance rather than mitigate risk. Patients need to be fully educated about the risk assessment process and the tools used; otherwise, they may feel bewildered and epistemically excluded from the process. Such feelings of exclusion and disempowerment may foster confrontational and other attitudes non-conducive to the recovery and risk mitigation process. Excluding the patient creates a situation in which they are not present at discussions about their risk and are therefore shielded from hearing their multidisciplinary team’s views. However, at care programme approach (CPA) meetings or tribunals, when decisions about the patient’s care pathway are made on the basis of these risk assessments, the patient is present and hears for the first time the team’s views of their risk, which may be at odds with their own perceptions. This sudden revelation encourages mistrust, which fosters lack of cooperation, possible rejection and denial of risk issues, and creates an atmosphere in which the patient’s cognitive distortions regarding their risk can be strengthened.14

The relevance and efficacy of care planning and risk assessment and management may be optimised through the development of mutual epistemic trust and understanding between patients and clinicians. Barriers may include patients’ priorities differing from those of their clinicians. Patients may place a greater emphasis on their need to regain their personal freedom whereas a clinician may place a higher premium on first achieving certain clinical outcomes. Patients may enter forensic psychiatric services with a low sense of self-esteem and a confused sense of their place in society. They may believe their offending behaviour was justified because of life or society dealing them with a bad hand. This belief may lead to the conviction that the disabling effect of their mental ill health means they have little chance of advance by conventional, law-abiding means. Collaborative working with regard to risk gives patients the opportunity to confront such maladaptive attitudes with their clinicians in a more informed and empowered manner and thereby facilitate the development of greater insight and more adaptive perceptions regarding the risk they may present and how to manage this.

The development by mentally disordered offenders of their own risk narratives may enhance self-awareness and insight and enable them to make better sense of their life experiences. They may also help them to begin to reposition themselves in society. According to Felton and Stickley, ‘The opportunity for new narratives and interpretations are important in terms of risk, particularly in relation to historical incidents of harm caused by or to an individual.15 The reinterpretation of events provides scope for people to come to terms and move on in accordance with the principles of recovery. Engaging with people’s own interpretations of these also enables professionals to consider the meaning of such events in the context of peoples current circumstances.15

Risk assessment and management in policy and practice

In the case of secure and forensic mental healthcare, the prediction of risk assessment is still problematic. The presumed link between mental disorder and violence has been the driving force behind mental health law and policy for centuries. Yet for three decades, research has shown that clinicians’ unaided assessments of ‘dangerousness’ are barely better than chance.16 Hence, those patients who are a risk to others form a special category in treatment planning, including SDM. At times, the patient may not accept the clinician’s or care team's views of his or her risk. In that case, the patient’s trust in the clinician and treatment will decrease; cognitive distortions may develop leading to problems in recompliance. Hence, in a high secure ward, portions of the risk assessment may need to be kept confidential especially in cases where the patient needs to be monitored for in the long term to prevent harm to others.

Another issue is the contested value of risk assessment tools such as the HCR 2017. The increasing use of risk assessment tools within the mental health sector has been driven by the imperative within services and those who oversee them provide more consistent and defensible measures of likelihood of future risk. The adoption of these instruments can also be explained by research findings, which indicate that they are better at identifying levels of risk than practicing clinicians.17 Proponents of risk tools often claim that they have been validated, seemingly ignoring the fact that validity research necessitates an ongoing process as contexts, populations, evaluation criteria and the salience of risk factors change.18 The variability in the quality and context-dependent appropriateness of these artefacts, in conjunction with the degree of objectivity and subjectivity with which they are applied at the individual patient level is likely to heavily influence the outcome of using collaborative risk assessment in forensic patients. This is an important factor to consider when comparing studies within this subject area, and in choosing which risk assessment methods to use.

Clinical professionals may not always recognise what constitutes authentic patient engagement, mistaking unquestioning receptivity and submission to their professional opinion as indicative of patient insight and compliance. In reality, such seemingly meek and respectful effect may be indicative of little more than fear and a deep sense that their own opinion, as the mentally disordered offender will displease their clinicians if it conflicts with the professional point of view. What someone may interpret as a good example of patient engagement may be little more than a facade of submission and compliance. However, given clinicians are ultimately responsible for patients' behaviour such as suicide, violence and homicide and are answerable to the tribunal, they will usually maintain an appropriate degree of scepticism and proceed with due caution, but with the patients’ welfare in their mind.

According to Langan and Lindow, most practitioners agree with the principle of involving patients more in their recovery process, but few are acting to implement it.19 The same authors in a study in 2008 explored how mental health professionals assessed patient’s risk to others and patients’ knowledge of, and involvement in, risk assessment. The study concluded that there is a research and practice gap about effective means of engaging service users in risk assessment.1 20

Very little literature is available concerning patients’ views and experiences of risk assessment and management. One exception is an observational study that found that patients ‘attempted to understand the system of assessment and sought to affect and reduce their risk status by engaging in overt, compliant behaviours’.21 This is an important consideration with regard to psychoeducation in the context of risk management and warrants further research. It is crucial that with regard to the forensic mental health patient cohort that clinical teams understand, are aware of, and are able to mitigate the risk of patients using their knowledge and understanding of risk assessment to manipulate the process with a view to the risk they pose being underestimated. This remains a regrettably under-researched area.

In another study, it was reported that some professionals lacked confidence or experience in openly discussing risk with patients.22 This indicated a need for more robust training in this area for clinicians. There is also concern that collaborative risk assessment and management might be perceived by clinicians as a threat to their professional hegemony.12

Decisions regarding risk are complex, discretionary professional judgements. It is important for the patients to understand the decision frame, that is, the values, assumptions and contextual pressures that shape the clinician’s decision-making. Multiple barriers to optimising collaborative working between clinicians and patients can present. It is likely that a range of contextual pressures impinge on risk assessment and management decisions, including demand for patient flow, which may overrule individual patient considerations about security needs (ie, a decision to transfer a patient to a ward with fewer security restrictions even when the patient is not ready for it).12

Clinicians may also be influenced by the current climate of policy and public expectation in which they practice and the expectations of service managers and regulators. Qualitative studies of decision making in forensic psychiatry services are scarce but the complexity of forensic psychiatry practice merits more use of such methodology.23 Studies have noted a ‘blame culture’ leading to the adoption of defensive practice and increased paperwork. Concern about independent Inquiries has affected practice and outlooks, including the willingness of clinicians to share decision making with colleagues and patients.24

Examples of good practice in risk assessment and management

Examples of good practice in collaborative risk assessment do exist.13 Consider the following two hypothetical approaches to collaborative risk assessment and management: one good based on practice in East London NHS Foundation Trust and one less so.14

A mental health foundation trust sets up a 20-week risk group for mentally disordered offenders in their low secure services. The initial stages in the programme are designed to support patients to develop an understanding of what was meant by risk. This is followed by education regarding the HCR-20 risk assessment tool together with help and encouragement as to how they might apply the tool to themselves and develop their own risk management plans. Role-play is used to allow patients the opportunity to present their risks and management plans to a mock mental health review tribunal, thereby improving their ability to articulate their knowledge and insight into their personal risk.

Another mental health foundation trust invites patients to attend a risk clinic 2 weeks prior to their 6 monthly CPA meeting. The purpose of the risk clinic is to complete the ‘Collaborative Risk Assessment’ section of three documents (one locally developed risk assessment form, the HCR-20 Violence Risk Assessment Scheme and a care plan regarding risk and safety planning). The patient’s contribution to their risk assessment is recorded within the ‘Collaborative Risk Assessment’ section of the documents, separate to staff’s assessment of their risk. There is no evidence of any training for patients regarding risk assessment or management and no integration of service user’s views with those of the multidisciplinary team.

Recovery model and collaborative risk assessment and management

The recovery perspective elevates the role of the patient from being a follower to one where they are able to lead, change and direct their own care. Collaborative risk assessment and management fit well within the Recovery Model. The role of the clinicians within this model is to support the process of recovery by empowering the patient to broaden the scope of his autonomy and sharing the responsibility for decisions about care and treatment. However, the reality of provision and implementation of the Recovery Model does not meet the officially recommended standards and, as a consequence, the quality of therapeutic relationships and collaborative working with regard to risk can be compromised.23 Reasons for this include staff resistance to change,6 lack of resources, unavailability of services and the perceived expectation that services should manage risk and provide social control.23

Nevertheless, evidence of the value of collaborative care in evaluating risk in secure and forensic settings does exist. A literature review exploring models of patients' involvement in risk assessment and the impact on outcomes in forensic mental healthcare found indications that patient involvement in assessing risk is feasible when correlated with staff ratings and that there is encouraging evidence of the predictive validity of self-rated risk alongside staff-rated risk assessment.25 This evidence endorses the relevance and need for further research into this area and importantly highlights how shared care is not necessarily detrimental to patient and public outcomes.


The 2014/2015 Commissioning for Quality and Innovation (CQUIN) commissioning framework for collaborative risk assessment sought to ensure a collaborative approach to risk assessment and management throughout secure and forensic hospitals by making it a part of their contract with the NHS26. Providers were required to deliver an education package jointly to staff and service users around risk assessment.26 The subsequent CQUIN framework for 2015–2016 sought to further develop and increase patient involvement in risk assessment.27 The rationale for training patients and the staff was that without knowledge of the key concept of risk and comprehension of the measures used to assess it, patients were at a disadvantage and risked becoming disempowered in the risk assessment process.8

A paucity of research into nature and extent to which this CQUIN framework has become embedded in the implementation of secure services means any effects of the CQUIN framework for risk assessment and management practices and outcomes for patients remain indeterminate. There is no evidence to indicate that this has led to improvements in clinical decision making regarding risk and patient pathways. It has been asserted that staff immersed in risk management paradigms may appear to acknowledge the virtue of collaborative working with patients, but their integrity may be in question.28 Furthermore, enduring systems of governance and authority may co-opt patient involvement, collaboration and advocacy as a means to pacify resistance.29


One of the key limitations of this paper is the lack of relevant research studies regarding collaborative risk assessment and management in secure and forensic settings. This is of special concern regarding the impact of the 2014–2016 CQUIN framework regarding risk assessment and management education on staff and the embedding of collaborative risk assessment and management in secure and forensic settings.


It has been evidenced that the implementation of the Recovery Model has fallen short of officially recommended standards and, as a consequence, the quality of therapeutic relationships and collaborative working with regard to risk can be compromised. However, evidence of the value of collaborative care in evaluating risk in secure and forensic settings does exist.

A complex range of clinical and contextual pressures may impact decision-making with regard to risk and the implementation of collaborative risk assessment and management in secure and forensic mental health settings. Research is required to elucidate the values, beliefs and professional insights underpinning decisions with regard to risk assessment, management and collaborative working with patients in order to improve understanding, inform clinical training and enhance the care, treatment and recovery outcomes of forensic mental health patients. This may best be achieved in association with academic clinical researchers currently practicing in this area.

According to the Independent Review of the MHA, meaningful embedding of collaborative risk assessment and management in secure and forensic mental health settings may require a systematic shift away from an anxious and rigid culture of risk aversion to a more fluid and responsive culture with increased emphasis on relational safety and epistemic regard for patient self-insight and testimony. More research is required to address the issues, barriers and complexities associated with SDM in secure and forensic settings. This would not merely inform good practice, but given the possible tensions between two key components of SDM, namely the patient’s human rights and the need for robust risk assessment and management to avoid recidivism, it may be essential if clinical practice is to be optimised. It is to be hoped that collaborative risk assessment in the context of mutual relational and epistemic trust within the therapeutic alliance could be a means of achieving such an end.


Dr Sarah Markham graduated from the University of Cambridge with a master's degree in Mathematics in 1996, and from the University of Durham with a PhD in Hypercomplex Hyperbolic Geometry in 2003. She currently is a Visiting Researcher in the Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King's College London. She is also active in raising public and patient awareness of the clinical trial process and the nature and experience of participation in research. In 2015 she designed and built a beta version online clinical trial recruitment portal for the NIHR Maudsley Biomedical Research Centre. Her main research interests include risk related practices in secure and forensic psychiatric services, the quality of practice in the First Tier Tribunals for mental health and the development and application of digital technologies to deliberation in Health Technology Assessment.

Embedded Image


  • Twitter @DrSMarkham

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.