Best Practice in Managing Risk is a recent Department of Health publication (2007) directed at mental health professionals. It provides a framework for risk assessment underpinned with principles of good practice for all mental health settings. This paper looks at potential issues that may influence the successful implementation of this framework across services. The authors have considerable experience in training mental health professionals to use risk assessment schemes and in implementation of risk management strategies within mental health and criminal justice settings.

Best Practice in Managing Risk (Department of Health, 2007) details 16 best practice points for effective clinical risk assessment and management within mental health services. The guidelines emphasise evidence-based practice and collaboration with the service user as crucial in the process of decision-making. They recommend structured clinical judgement approach to risk assessment, multidisciplinary working, well thought-out and well-imparted training, and clear procedures on how, and to whom, risk must be communicated. Both organisations and individual practitioners can gauge their current practice against this framework. The plan is piloted by the Care Standards Improvement Partnership in a small number of mental health services across the UK. Many services may fall far short of the recommendations contained within this document and major changes may be required to risk management procedures and the cultures within which they operate.

The authors have direct experience in training mental health practitioners to use clinical risk assessment tools, including those with a structured clinical judgement approach. The schemes include the Historical, Clinical, Risk Management-20 (HCR-20; Reference Webster, Douglas and EavesWebster et al, 1997) and the Short-Term Assessment of Risk and Treatability (START; Reference Webster, Martin and BrinkWebster et al, 2004). The HCR-20 is a research-based clinical practice guide which aids the construction of risk management plans. It consists of 20 items (that are well described in clinical practice and the published literature as important in the systematic evaluation of violence risk). There is an increasing amount of validation evidence from the UK and internationally that shows that HCR-20 scores are related to violence in different samples of individuals with mental and personality disorders in civil, forensic and criminal justice settings (Reference Gray, Snowden and MaccullochGray et al, 2004; Reference Doyle and DolanDoyle & Dolan, 2006). The START is based on similar principles to the HCR-20 by indexing change via 20 dynamic variables which may relate to certain short-term, often overlapping risks. The scheme was initially developed with forensic in-patient and out-patient services in mind, but it can also be applied in a wide range of general mental health settings, especially when used by multidisciplinary teams to guide clinical interventions and assess changes in risks over time.

This article provides a framework to aid organisations in developing local risk management strategies and advises on overcoming specific implementation barriers.

Uptake of risk assessment courses

Between 2004 and 2007, the authors delivered a total of 17 2-day basic risk assessment workshops organised by the Department of Forensic Mental Health Science, Institute of Psychiatry, and attended by 417 delegates. Workshops included training on the use of the HCR-20 and related schemes. They were attended by psychiatrists, psychologists, nurses and social workers. For the first 3 years, those mainly working in forensic mental health and criminal justice settings attended, but more recently practitioners from civil settings have also participated. The effectiveness of isolated workshops was evaluated by testing knowledge acquisition and the achievement of interrater reliability when scoring specific case exercises with the various structured clinical judgement guides. Although workshops have an important role in risk assessment training, their role is limited and a more comprehensive training strategy is required. The strategy should include the following:

1. • Providing all mental health professionals with introductory ‘awareness’ modules on clinical risk assessment and management. These modules should include information on the strengths and limitations of various risk assessment tools, and written and visual materials on risk management, for example the HCR-20 companion guide (Reference Douglas, Webster and HartDouglas et al, 2001).

2. • Specific training to use schemes such as the HCR-20.

3. • The use of locally relevant clinical case material, ideally supported by video interviews.

4. • A system of consultation and ‘booster events’ to resolve administration and implementation issues.

5. • Concurrent implementation of evidence-based quality assurance systems.

Workshops are likely to be most effective when delivered ‘in-house’ to multidisciplinary teams supported by a continuing implementation strategy.

Implementing a three-tier risk management approach in a large independent provider of secure services

In 2004, Priory Secure and Step-Down Services decided to implement a risk management strategy supported by structured clinical judgement approaches in all of its secure hospitals. The treatment settings involved were medium secure, low secure, intensive care and step-down wards caring for male and female patients typically detained under the Mental Health Act 1983. The main objective was to generate risk assessments that were evidence based, transparent, regularly updated, collaborative and action orientated. The authors (Q.H. and A.C.) supported a team of professionals at each hospital in implementing the HCR-20 across all settings. Training and implementation was introduced on a ward-by-ward basis. The focus of the training was to provide practical skills in structured clinical judgement and completion of practice case studies. Additional training was provided on the evaluation of mental disorders, including the construct of psychopathy. Each clinical team could decide how best to implement the tool once the training was completed, but the implementation had to meet the criteria set out in the hospital protocols for risk management. Phasing in the training and implementation of the risk assessment devices resulted in wards developing a core of experience and providing advice and support for subsequent wards going through the process. Regular monthly feedback from each ward allowed to quickly address problems and to adapt training and implementation as necessary.

Once established, START and a risk communication protocol (a concise communication and patient zoning system based on the findings from the structured risk assessment; additional information available from Q.H.) were implemented in a similar staged manner across all sites. The START is a concise clinical guide for the dynamic assessment of a variety of often overlapping harmful behaviours (e.g. violence, self-harm, suicide, self-neglect, unauthorised absence, substance misuse, victimisation), founded on structured clinical judgement principles. The scheme provides a method for systematic assessment of a service user's strengths and vulnerabilities, and encourages multidisciplinary collaboration in the construction of clear and workable care plans.

Each ward held a three-tier approach to risk management that addressed concerns on a shift-by-shift basis using the risk communication protocol, regular START reviews and other specific structured clinical judgement tools such as the HCR-20. This system encouraged teams to complete more sophisticated updating of HCR-20 assessments that could be reviewed at ward rounds and at care programme approach meetings. Each hospital established a local risk management committee.

Audit

An audit strategy, initially focused on quantitative aspects of implementation, then progressing to qualitative issues, helped routine decision-making. Existing audit tools evaluate whether risk assessment schemes are used appropriately in day-to-day practice, and whether risk management plans reach acceptable standards of quality. They also help assess whether such plans are effective in improving individuals’ mental health and decreasing the risks associated with their management in various psychiatric settings. Parallel audits and research projects have monitored the nature and frequency of adverse incidents and quality of documentation in the clinical notes.

Avoiding pitfalls

The successful implementation of structured risk management programmes is enhanced if:

1. • key personnel have a good understanding of approaches to implementation of evidence-based strategies (e.g. Reference Hyde, Falls and MorrisHyde et al, 2003). This requires study outside of traditional behavioural sciences literature

2. • there is close support from and regular communication with managers who hold clinical and financial responsibility

3. • the limited shelf-life of ‘product champions’ and good will is appreciated and individuals with key roles in clinical risk management research and training have their responsibilities recognised in their job descriptions

4. • staff are consulted early and regularly to understand reasons for resistance, should it occur. From our experience, staff resistance is often unrelated to specific concerns about the clinical utility of structured clinical judgement

5. • professionals are responsible for completing individual risk assessments. This should be shared among disciplines, as should the paperwork

6. • there is collaboration with service users and their families

7. • mental health services are aware that there can never be a definitive guide to risk assessment and they allocate funds for the evaluation of risk management strategies. Ideally, this would encourage the routine collection of follow-up data. Only through establishing administrative research projects within specific organisations is it possible to find out the extent to which clinical judgements and predictions about specific risks are fulfilled. More importantly, such projects need to demonstrate the effectiveness of structured clinical judgement schemes in managing risk, for example by demonstrating that clinical teams that use the HCR-20 are better able to identify factors associated with violence in individuals and that they have intervened to lessen these factors.

Networks

The increasing pressure to approve trainers to teach structured clinical judgement has led to discussion about what qualities they need. That they should have a considerable degree of technical and scientific knowledge in the field is obvious. Yet, potential trainers should also have considerable worldly experience in the assessment of service users with structured clinical judgement schemes and other approaches. For this reason, we have developed an Advanced Programme in Structured Clinical Judgement at the Institute of Psychiatry, London, which encourages participants to think beyond the mere mechanical application of risk assessment tools. Themes stressed in the programme include relevant aspects of research methodology, teaching skill development, law and ethics, and understanding the limitations of one's practice. Over 50 senior professionals have attended the course and many have retained close contact with the authors, forming a support network of trainers.

Future directions

Clinicians who receive well-organised but limited training in evidence-based risk assessment improve self-confidence and are able to better articulate the rationale for their risk assessment and risk management plans (Reference McNiel, Chamberlain and WeaverMcNiel et al, 2008). We are proposing a more sophisticated training strategy. Research is required to understand whether, and how, clinical teams can deal more effectively with these kinds of tasks as compared with individual practitioners. Our experience in recent years has shown that, in the UK, training in structured clinical judgement approaches has been too narrowly applied to clinicians working in forensic environments. Implementation of structured risk assessment and management frameworks to in-patient and community general adult services should now be prioritised and carefully evaluated. The Best Practice in Managing Risk document provides a timely invitation to all mental health services.

Declaration of interest

None.