The Scottish Government

« Previous | Contents | Next »

Listen

SERIOUS VIOLENT AND SEXUAL OFFENDERS: THE USE OF RISK ASSESSMENT TOOLS IN SCOTLAND

CHAPTER SIX: MULTI-AGENCY RISK ASSESSMENT AND MANAGEMENT

INTRODUCTION

6.1 As the discussion in Chapter Five will have made clear, effective risk management was considered by those who took part in the audit and in the interviews usually to require a co-ordinated, multi-agency approach (see also Maguire et al., 2001). This chapter, therefore, considers the extent to which and ways in which different agencies were involved in the process of risk assessment and in the development and implementation of risk management plans. The chapter concludes by identifying the perceived strengths of a multi-disciplinary approach along with the factors that were perceived to hinder more effective multi-agency work.

MULTI-AGENCY RISK ASSESSMENT

Use of risk assessment tools in multi-agency risk assessment

6.2 Respondents who participated in the audit were asked to indicate how frequently risk assessment tools were employed in a multi-agency context. The resulting data are summarised in Table 6.1.

Table 6.1: Use of tools referred to in the audit in multi-agency risk assessments

6.3 In around half of the returns the tools were reported as always or usually being used as part of a multi-agency risk assessment. In a similar proportion of returns they were said to be only sometimes or never employed in this way. Police officers and prison-based psychologists were more likely than social workers or health professionals to report that they regularly employed tools in a multi-agency context.

6.4 There was also a tendency for specialist tools to be more often employed than generic risk assessment tools as part of a multi-agency risk assessment. In Table 6.2 the various 'specialist' tools have been combined to facilitate comparison. Locally developed police risk assessment procedures were also used regularly as part of a multi-agency risk assessment.

Table 6.2: Frequency of use of different tools in multi-agency risk assessments

Who is involved in multi-agency risk assessment?

6.5 The range of other professionals involved in multi-agency risk assessments of serious violent and sexual offenders is summarised in Table 6.3. The nature of other agencies involved in the risk assessment and management process will clearly differ according to the location of the offender (e.g. in prison, in the community, in a mental health setting). Amongst the sample of respondents in the audit survey (in which social workers were over-represented) the other agencies that were most often involved in multi-agency risk assessments were the police, housing agencies, psychiatrists and 'other mental health professionals'. Staff from education departments were least likely to be involved in multi-agency risk assessments (or decisions flowing from them), presumably because their involvement (like those of other agencies) was on a 'need to know' basis.

Table 6.3: Other agencies involved in multi-agency risk assessments involving the tools referred to in the audit

6.6 'Other' agencies included the Parole Board, tribunal members, staff in secure accommodation and any other agency who might have information relevant to the assessment of offender risk:

"Any agency / professional who has knowledge of the particular person."

"Anyone or any agency who / which has information to assist in the assessment process."

"In cases of high risk sex offenders assessments are shared amongst relevant individuals involved in our multi-disciplinary 'Sex Offender Forum'."

6.7 Social workers did not generally conduct risk assessments jointly with other agencies, though information provided by other agencies often informed these assessments and the resulting risk management plans. As one respondent explained, "Other agencies are used for sources of information, they do not complete the assessment."

6.8 Social workers - including those in specialist projects - shared their risk assessments of sexual offenders with the police. In one area, however, joint visits for the purpose of assessing sex offenders were usually undertaken by social workers and police officers. This arrangement was perceived by the police respondent as more effective (in terms of information sharing) and more efficient (since it avoided duplication of effort).

6.9 Staff in health settings were more likely to adopt a multi-agency approach to risk assessment, with staff from different professional groups assuming responsibility for relevant parts of the assessment (for example, nurses might be asked to examine a patient's medical history or history of admissions). In the State Hospital there was an intention to move towards multi-disciplinary risk assessment though at present different elements of risk assessment were said to be being carried out independently of each other.

6.10 In prisons, psychologists were responsible for undertaking risk assessments. However, as previously noted, Risk Management Groups were being established in long-term prisons to improve the targeting of detailed risk assessments. The Risk Management Groups - consisting of prison managers and officers, social workers and psychologists - would consider the information provided by the collateral file review and decide whether a full psychological risk assessment was required. The risk assessment would then be discussed by the multi-disciplinary team who would have an opportunity to feed into the resulting recommendation.

6.11 The perceived advantages of this arrangement were that it provided psychologists with access to a wider range of information and perspectives and engendered a sense of ownership of risk management plans. The perceived disadvantages were that differences of opinion might arise and that some members of the group might have limited understanding of risk assessment and management, including the tools that are used.

Issues arising from the use of the tools in multi-agency risk assessments

6.12 A number of issues were identified from the audit returns, some of which pertained to the adequacy of individual instruments and some of which concerned their use in a multi-agency context. In general, views about multi-agency risk assessment were mixed, with some respondents indicating that it did not prove problematic. Overall, however, issues were said to have arisen in 30/78 returns in which the tool was reported to be employed as part of a multi-agency risk assessment process, with each of the professional groups indicating that difficulties of this type had been encountered (Table 6.4).

Table 6.4: Whether issues had arisen in the use of the tool in a multi-agency context

Table 6.5: Types of issues that have arisen in multi-agency risk assessment

6.13 The types of issues identified are summarised in Table 6.5. Respondents most often believed that difficulties had been created by the use of different tools by different professional groups. For example, as one social work respondent commented:

"The [specialist project] have assessed sex offenders as medium risk when the case holder using RA1-4 assesses high risk. This can be confusing for the courts."

6.14 Respondents also indicated that there had sometimes arisen lack of agreement as to how to interpret the results. As one police officer observed:

"Social work apply Thornton Risk Matrix 2000 for production of their assessment although all interviews are joint police/social work. Occasionally there is 'disagreement' on the assessed risk. All cases go before the Joint Scrutiny Board who recommend/confirm grading to be applied as well as management processes in respect of subject." (original emphasis)

6.15 In some cases respondents suggested that there had been a lack of clarity as to how the results should inform risk management strategies. As one mental health respondent explained, different agencies sometimes placed:

"…different emphasis on importance of mental illness in contributing to risk assessment and management."

6.16 In addition to these shared concerns, some of the different professional groups who participated in the audit raised particular issues. Some social work respondents expressed some concern that if different tools were being used by different professional groups the same offender may be identified as presenting different levels of risk. One social work respondent also suggested that problems had arisen with respect to disclosure, particularly in the context of joint police/social work risk assessments of sexual offenders.

6.17 Police officers' concerns centred on the fact that different risk assessment procedures were adopted by different forces and by different local authorities, creating particular difficulties when offenders moved from one part of the country to another. Multi-agency risk assessments were also thought by some police respondents to be hampered by the fact that police risk assessment procedures had not been validated and relied heavily upon professional judgement: structured tools such as Matrix 2000 or the Tay Project Assessment Tool would be welcomed but were not yet widely used.

6.18 In practice, although they adopted different approaches to risk assessment, police and social workers' assessments were said generally to be in accordance. Thus while the use of common tools might, in principle, reduce the potential for disagreement, this did not appear to be a salient concern. One of the challenges in developing a common approach to risk assessment would, it was suggested, be how the tool could be incorporated in a wider framework that met the specific information needs of the two organisations.

6.19 The benefit of different agencies approaching risk assessment differently was thought by respondents in different professional groups to be that it enabled a wider range of information to be brought into a multi-disciplinary forum. However, instances were cited where different conclusions about risk had been reached by different professionals, usually because there had been no communication between them prior to the preparation of risk assessment reports.

6.20 Discrepancies were also reported occasionally to have arisen when patients in the State Hospital hired independent psychologists to provide risk assessments because they were unhappy with the risk assessment carried out at Carstairs. Independent psychologists tended, it was suggested, to be less objective in their scoring and interpretation of risk assessment tools and to be inclined to produce more favourable risk assessment reports.

A consistent approach to risk assessment?

6.21 In their attempt to bring uniformity and a greater degree of structure to their risk assessments, the Scottish police forces were considering the adoption of the Tay Project's risk assessment tool. In interview the police respondents expressed a clear preference for this tool over Matrix 2000 (which has been adopted by police forces in England and Wales). The former they believed was more comprehensive and was easier to use, though the fact that it had not yet been validated was recognised as a disadvantage. Social work respondents were also open to the possibility of adopting the TAYPREP-30, particularly if it was adopted nationally by the police, since this would allow for a more consistent approach to risk assessment by different agencies.

6.22 Some respondents believed that there would be value in having a common approach to risk assessment across Scotland, though others were less convinced that this was necessary so long as whatever tools were being used provided the information required. As one social work respondent observed:

"It doesn't particularly bother me what risk assessment it is as long as they've used something which encapsulates risk of re-offending, risk of harm, criminogenic needs, that sort of stuff."

6.23 Other social work respondents questioned whether it would be possible to have a standardised approach to risk assessment while different professional groups operated with different theoretical and value bases. Another stressed that using the same tool was not enough if the different agencies did not communicate adequately with each other. There was, however, some support for the development of a more standardised approach to the production of reports, with a consistent emphasis upon identifying risk factors and how different agencies might have a role in managing various risks.

MULTI-AGENCY RISK MANAGEMENT

6.24 Social workers reported being involved with other agencies in risk management. These included the police, psychologists, psychiatrists and any other agencies that had a relevant input to make. Police officers would similarly involve other agencies in the management of offender risk, for example by informing education or housing departments if they believed that a particular offender's behaviour suggested that they might pose a risk.

6.25 In some parts of the country more formalised arrangements had been instituted for liaison between different agencies. In Edinburgh, for example, designated sex offender liaison officers - SOLOs - had been appointed by the police, social work and housing. This arrangement was said by the police to have greatly facilitated the exchange of information to manage sex offender risk.

6.26 Psychologists in health settings were also keen to encourage multi-agency involvement in risk management since they reviewed this as essential for the effective management of risk. Within Carstairs, for example, risk management plans were reported to be shared between all relevant disciplines and information forwarded in the event of a patient being transferred to another institution.

6.27 Another example of multi-agency risk management was described by the Tay Project, where a nurse co-ran the groups for sex offenders with learning disabilities. There were also plans in place for the sex offender registration officers from the local police force to share premises with the project to facilitate the sharing of information and their attendance at reviews.

6.28 Prison personnel perceived the absence of a national framework for risk assessment by social workers as undermining the potential for effective risk management of prisoners once they returned to the community. A prison-based psychologist suggested that " if psychology is the lead agency in risk assessment, I think social work is the lead agency in risk management". The potential for a 'seamless' transition from prison to the community was also said by prison respondents to be undermined by the fact that similar programmes of intervention were not in place in the two settings, with the result that there was little continuity of provision when a prisoner was released.

6.29 Within prisons, the nature and extent of multi-agency risk management between prison psychologists and other professionals appeared to vary across establishments. At one level, a copy of the risk management plan would be placed in the prisoner's warrant file and therefore be available to any other professionals who wished to access it ²⁶. At another level, other disciplines would be brought into a case if particular issues or difficulties arose. At a third level, one psychologists reported regularly involving prison and community-based social workers in discussions about the management of offenders' risk and would have welcomed the opportunity for this to occur on a more formalised basis.

6.30 Overall, therefore, different professional groups recognised the value of multi-agency working and expressed willingness in principle to engage in it. As one prison psychologist explained:

"If we were able to develop good written multi-disciplinary risk management plans that are short, medium and long term so that there are plans for within the establishment, plans for when someone leaves and that the targets are set for each professional within the multi-agency group and those plans are reviewed as the prisoner progresses through the system - I think that would be a good way forward in terms of bringing in more disciplines into the risk management process."

6.31 At a practical level, however, there was acknowledged to be some way to go before different agencies adopted a common approach to risk assessment and management.

Perceived advantages and disadvantages of multi-agency risk management

6.32 Multi-agency risk management was seen as having a number of advantages. For example, it enabled the approach adopted to be tailored to the nature of the risks posed by the offender and it also served to spread responsibility for managing the offender's behaviour across a number of agencies rather than it being the sole responsibility of one. As one social work respondent commented:

" I think it's a rather brave person who would say that our agency can do it all… So I think there's a bit of support."

6.33 A respondent from a specialist project suggested that the sharing of information among different professionals could provide a forum for obtaining feedback about how effective various elements of the programme had been and could also help communicate a message to offenders that their behaviour was being closely scrutinised. A similar point was made by a police respondent as follows:

"Most high risk offenders are well aware that we do sit down and discuss them with everybody… Particularly your older offenders are well aware that police and social work did not work well together in the past or got on that well. They are now under the impression, quite rightly, that we now work closely. They realise we sit down and talk about them."

6.34 In principle, therefore, respondents viewed multi-agency involvement in risk management as something to be encouraged. However, some disadvantages, which might undermine an effective partnership approach to the management of serious violent and sexual offenders, were also identified. They can be summarised as:

• Some agencies having to assume responsibility for a disproportionate amount of the work.
• If sensitive information is going to be shared with other agencies on a need to know basis, they need to be trusted to deal with it appropriately and not to over-react.
• The practical difficulties involved in getting all the relevant parties together, especially when a large number of agencies might be involved ²⁷.
• Lack of agreement about the risk presented and resources required in a particular case, possibly as a result of different agencies placing differing emphases on issues.
• Obtaining information from health services as a result of issues around patient confidentiality, though in some areas local protocols have been developed.
• While the sharing of information between social workers and the police was generally good, social workers occasionally omitted to pass information on.
• Lack of agreement as to who should contribute what in terms of funding of risk management strategies.
• Different agencies having different priorities.
• Other agencies' lack of knowledge and understanding of issues relating to sex offenders (e.g. community care workers who were said often to have little awareness of the issues presented by sexual offenders with learning disabilities).

6.35 Multi-disciplinary working required mutual trust and could in some circumstances result in the blurring of professional boundaries. As one respondent, who worked in a multi-disciplinary team in a health setting explained, "it is sometimes shocking when a person from another discipline volunteers to or is asked to take on a piece of work that is another discipline's".

SUMMARY

6.36 There was a general recognition, among those who participated in the research, of the value of a multi-agency approach to risk assessment and, in particular, to risk management, though the extent to which this occurred was somewhat variable. Opinions were divided as to whether it was feasible or desirable to implement a common approach to risk assessment across different disciplinary groups, though a greater degree of consistency would be welcomed.

6.37Multi-agency risk management operated in different ways and at different levels within various settings. The strengths of multi-agency approaches to risk management included the increased ability to tailor responses to the risks presented by an individual and the avoidance of a single agency being held accountable for any adverse outcomes. Effective multi-agency approaches could, however, be undermined by issues such as the failure to share relevant information and the varied approaches adopted by different agencies towards risk assessments and their interpretation.

« Previous | Contents | Next »