SEED Sponsored Research
PROTECTING CHILDREN IN SCOTLAND : AN INVESTIGATION OF RISK ASSESSMENT AND INTER-AGENCY COLLABORATION IN THE USE OF CHILD PROTECTION ORDERS
Joe Francis, Janice McGhee and Enid Mordaunt
School Of Social and Political Studies
University Of Edinburgh
Scottish Executive Social Research
Thanks are due to the many people who have supported and been directly involved in this study. We are particularly grateful to all the local authorities that responded to the survey request and in particular to the three authorities who facilitated the authors' arrangements to interview staff. We convey our thanks to all the managers and practitioners who gave their time to take part in the interviews. We would also like to thank the Scottish Children's Reporter Administration for providing the aggregated statistics for the study.
Chapter One: The study context
Chapter Two: Study aims and methodology
Chapter Three: Survey Findings and National Statistics
Chapter Four: Interview Findings
Chapter Five: Conclusions
The Scottish Executive Education Department sponsored the authors to undertake a study of the use of the child protection order in Scotland (sections 57-62 Children (Scotland) Act 1995). The order is based on stringent criteria and allows for the detention of a child in a place of safety where there are urgent concerns about a child's wellbeing and protection. It arose in response to criticism of prior legislation that allowed for undue discretion in the grounds for removal (Scottish Office, 1992). Recent inquiries into the deaths or abuse of young children (e.g. Hammond 2001, Laming 2003, O'Brien, 2003, Scottish Executive 2005a) have indicated that there are ongoing concerns about limited inter-agency communication, information-sharing and collaboration and poor standards in risk assessment. The main aim of the study, therefore, was to examine issues of risk assessment, thresholds of intervention and inter-agency co-operation in the context of the operation of the child protection order.
The study was conducted over a ten month period from January to October 2005. Both quantitative and qualitative research methods were utilised and the research entailed three inter-related stages of data collection: a review of statistical data provided by the Scottish Children's Reporter Administration (SCRA); a national survey of all Scottish local authorities; and interviews in three local authorities with a variety of social work practitioners and managers. Twenty-nine local authorities completed the survey and 25 interviews were conducted with social work staff in three local authorities. The key findings are set out below.
There has been an underlying upward trend in the number of Child Protection Orders (CPO) obtained by Scottish local authorities in the six year period from 1999-2005. Although there has been some variation in the pattern of orders obtained by individual local authorities over this time, the total number of child protection orders increased by 50%, with 558 obtained in 2004 compared to 370 in 1999.
Around 60% of children subject to orders were aged less than five years in each of the six years from 1999-2005. Most orders in this age group relate to children aged less than one year. The gender distribution of children subject to child protection orders is almost evenly divided throughout this period with slightly more boys than girls.
The majority of applications for child protection orders are made by social workers, accounting for 86% to 96% of all orders obtained in the years from 1999 to 2005.
In the six-years from 1999/2000 to 2004/2005 there was an annual average of 21 applications to recall child protection orders and 3 applications to vary directions in the orders (127 applications to recall orders and 18 applications for variation, in total).
Several authorities identified the rise in the number of substance misuse cases they are dealing with, especially drug misuse and associated pre-birth assessments, as a common reason to explain increases in child protection order applications.
Policy and procedures
More than half of the study authorities do not have an agreed written policy on the use of child protection orders. Where such policies do exist, the evidence obtained here suggests that these are based on a multi-agency approach with a range of agencies involved, including police, health, legal services and the child protection committee.
The vast majority of authorities in the study provide procedural guidance on applications for CPOs but this is mainly contained in general child protection guidelines and not issued separately.
The survey findings indicate that policy on the process of applying for child protection orders is diverse across Scottish local authorities. Of the 29 authorities in the study, applications were made by legal services in nine authorities, by social work staff in a further nine and jointly by legal services and social work in 11 authorities.
Having shared policies between relevant agencies and greater clarity about the roles and responsibilities of different professionals is seen as important but there seems also to be scope for continuing improvement in this area.
Almost half of the authorities in the survey reported that they do not employ a specific framework or model to assess risk. Of those who do, the majority use in-house models or frameworks, and only five authorities indicated that they use the Department of Health's Framework for Assessing Children in Need (Department of Health, 2000)
General concern about the adequacy and availability of models of risk assessment was reflected in the three authorities where interviews were conducted. A number of those interviewed indicated that no single model was sufficiently comprehensive to be used in all circumstances. As indicated in one authority, the range of assessment work can include 'pre-birth risk assessments' with vulnerable parents such as those who misuse alcohol or drugs; 'integrated/joint assessments' with other agencies and professionals; 'full assessment reports' for children's hearings.
Several interviewees confirmed that although they were aware of a range of models none were used regularly in an overt manner and they were very conscious that their practice was not based on specific published models or frameworks. A general concern expressed is that practitioners are therefore forced back onto their own experience.
Less than one-third of the authorities in the survey indicated that they issued specific guidance on assessing 'significant harm'.
The study found that defining significant harm in practice is complex. A general view expressed in interviews was that the term 'significant harm' is open to broad interpretation. Only one of the three authorities where interviews were conducted had established a definition of the term and identified factors to be taken into account when deciding whether or not to apply for a child protection order.
There was a level of ambivalence about the need for a written definition. Professional experience and instinct are seen by some to be more valuable than having a written definition. However, it is also recognized that a lack of definition can leave less experienced workers in a difficult position and that individual social workers or individual teams are using their own experience to develop thresholds.
The findings suggest that a broad definition would be useful in child protection guidance but no interviewee was in favour of a regimented or prescriptive definition. Assessing each case individually is seen to be important and the professional judgement of social workers is viewed as a crucial factor in this process.
Thresholds for intervention
There is no clear consensus about whether thresholds for intervention have been raised as a consequence of the introduction of child protection orders. However, the findings do suggest there is a consensus that the application process has become more rigorous since the introduction of the CPO. Interviewees reported that there is now more scrutiny, with sheriffs requiring better evidence based applications than a Justice of the Peace did under the old system. Most thought that it was easier to be granted a place of safety order under the previous system than to obtain a child protection order.
Notwithstanding this view, refusal of a child protection order is apparently rare and only one interviewee could recall this taking place. Two main explanations are suggested: first, applications are subject to a process of joint or collegiate decision-making, with debate and internal scrutiny, before an application for an order is made and then they are scrutinised by a legally trained professional (the Sheriff). Second, they are now more firmly rooted in an evidence based approach.
Despite concerns identified in the Scottish Executive audit and review (2002), interviewees accept the need for sound justification of their application for a CPO. Overall, they believe that sheriffs are appropriately rigorous in their questioning and they do not feel they are being cross-examined in this process. However, valued as it is, a more rigorous system was also seen to lead to other problems, especially in relation to time-scales in applications and potential delay in protecting a child. There appear to be some issues about the immediacy/emergency requirements in the application process as it can take some time for all the relevant material to be gathered together and placements arranged.
The minimum intervention principle appears to be having a strong influence on contemporary practice. Where it was felt that the threshold for removing children from their parents had been raised, interviewees identified a number of factors including: greater understanding that more harm can sometimes be done to children by removing them than by leaving them at home (even with some degree of risk); greater emphasis on supporting the family to avoid the removal of the child; a general view that alternatives to safeguard the child, such as voluntary admission to public care (section 25 Children (Scotland) Act 1995) enlisting the support of other relatives or neighbours to accommodate the child, or the deployment of additional resources in the home, would be preferred to a child protection order.
Alongside this, however, there is a view that the level of risk being managed by social work staff in working with a family may be higher now than it was before the implementation of the Children (Scotland) Act 1995.
The findings suggest that the nature and quality of joint working arrangements varies considerably between different agencies and professionals in different areas. A number of obstacles will have to be overcome in order to generate more effective inter-agency practice. However, the principle of joint working, both at ground-level and in strategic terms, is generally accepted by social work staff and there is a climate of openness to improving joint working arrangements.
Both survey and interview responses suggest that there are currently particularly close working relationships between social work services and the police and between social workers and health visitors. There is a sense that the reality of close working relationships leads to improved levels of understanding about differences in roles and responsibilities. Joint training appears to facilitate this process. Besides good relationships with police and health visitors, almost all of the survey authorities also reported a positive degree of liaison between social work and legal services.
Agencies less directly involved in child protection work are seen as having less understanding about social work roles and responsibilities. The interview findings suggest that education is viewed as the agency where there are most likely to be concerns around trust and confidence in working relationships and where participation and ownership issues can arise in child protection work. Even in the context of an integrated children's services department, concern was expressed that the divide between social work and education reflects deep-seated differences in professional values, culture, language and attitudes.
Several of those interviewed implied that a significant difficulty in effective joint working in child protection work is the lack of corporate ownership and shared responsibility for child protection practice. The view that social workers were usually left to carry responsibility for child protection issues was a recurring theme throughout the interviews. This also seemed to contribute to a concern expressed by some that aspects of social work are being "hived off" to other agencies while social workers are left to cope with child protection. There is a view that such failure to 'own' responsibility can result in poor practice within agencies that have a key role in this work.
Both structural and cultural barriers to improving joint working arrangements are indicated, including: different geographical boundaries; different organisational arrangements; and different priorities, cultures and languages. Additionally, some barriers to effective inter-agency working appear to be rooted in low levels of trust and confidence in the way different professionals and agencies engage in child protection work.
For many in the study, effective collaboration is based upon the quality of individual working relationships that workers build between themselves. It was suggested that closer working relationships may not be brought about solely by structural changes but that fundamental shifts in attitudes will be required too. There are likely to be resource implications in creating the context to allow this to happen. Some interviewees suggested that improvements in staffing resources are necessary to facilitate the development of closer working relationships between a range of agencies and professionals.
Concerns were raised that other professionals may not be adequately trained for child protection practice and about the lack of shared professional standards. These concerns were not shared by all, however, and one respondent questioned whether social work staff can be too possessive in relation to this field of work and hold unreasonable fears that the involvement of other professional or agencies can result in difficulties.
Interviewees' comments suggest that social workers regard information sharing as an important aspect of child protection work but some frustration was expressed about the consistency of practice in this area, with variability between different agencies.
The value of information sharing was seen as particularly important in relation to risk assessment and in providing a consistent view at the time of applying for a child protection order. Overwhelmingly, the survey responses most frequently identified police and health visitors as the main providers of information (93%, n=29), followed by schools and paediatricians (62%, n=29). Drug advice services were notably involved in providing information in a quarter of the survey authorities (24%, n=29). However, less than one-fifth of authorities (17%, n=29) saw general practitioners as routine providers of information.
Some of the responses indicate that professionals, particularly in the health sector, have shifted ground on this issue and are now more willing to share information in child protection cases than they would have done in the past. More commonly though, interviewees reported that they experience difficulty in obtaining information where the question of professional confidentiality arises. The main problem appears to be the differing perspectives professional hold about their responsibilities towards the adults in child protection cases and in determining the balance between parents' rights and children's rights.
Developing joint training and raising awareness of other professionals' roles and responsibilities are viewed as vital elements in promoting effective inter-agency collaboration. However, following a surge of in-house training at the time of the implementation of the Children (Scotland) Act 1995, interviewees observed that there now appears to be little on-going specific training in child protection and child protection orders. Asked whether professionals had undertaken joint-training for their work in child protection, interviewees gave mixed responses, with some indicating that this is an area that requires considerable development.
The study suggests that while some in-house and inter-agency training is being provided, the degree to which this is structured and co-ordinated is extremely variable. Moreover, the impact of training on professional practice does not appear to be consistent across the authorities. Responses also indicate that, apart from police and health visitors, training does not always encompass the full range of professionals or agencies that are involved in child protection work.
The role of child protection case conferences in decision making
The timing and place of child protection case conferences in relation to an application for a child protection order is a complex issue. Arranging a child protection case conference prior to seeking a CPO was generally seen as impractical in light of the emergency nature of such applications. However, where children are already known to social workers, interviewees reported that information from an earlier case conference might usefully be included in the child protection order application.
There was general agreement that holding a case conference prior to seeking a child protection order in cases involving an unborn child was important. However, using the child protection case conference in these circumstances can also raise dilemmas for practitioners, such as whether or not to invite the mother for fear that sharing thinking at this stage could result in the mother disappearing.
The general view appeared to be that a case conference would be more likely to follow the granting of a child protection order. However, views varied on this point, suggesting that the role of the case conference at this stage is uncertain in the minds of practitioners. The decision to hold a case conference after a CPO has been granted appears to hinge on whether the workers concerned regard this as a necessary means of safeguarding the child when other measures, such as a children's hearing or a looked after child review, have been triggered by the order.
This issue highlights the complexity of the relationship between different aspects of the decision making system in applications for child protection orders, including the courts, the children's hearing system and the looked after children system.