Scottish Government

It's everyone's job to make sure "I'm alright" Literature Review

Part III: Addressing the problem and 'what works' in child protection
Messages from elsewhere

This chapter briefly considers the ways in which some other countries address child protection concerns.

(i) Standardised risk assessment and differentiated response systems

Precise risk assessment tools have been used in North America and Australia with the aim of ensuring judgements about eligibility for services or the need for investigation and intervention are standardised and comprehensive. In many ways the Australian child protection system is similar to that found in the UK. Some states have, however, introduced much more standardised risk assessment and case response differentiation mechanisms than found here (Hill 2002). Most services have adopted some form of structured risk assessment tool or checklist. Attempts have also been made to more efficiently tailor the response to reports of child maltreatment through the introduction of some form of differentiated response system or the streaming of reports based on an initial assessment of the extent to which reported concerns require/do not require a child protection assessment. Central intake teams have been introduced. Some agencies have pioneered new ways of helping families, for example, the Strengthening Families Model in Victoria provides support to families 'at risk' to prevent them becoming child protection clients. Staff work primarily to build on families' strengths rather than modify their deficiencies and seek to engage families in developing their own solutions. The benefits of these new practices in Australia are that families are not unduly traumatised by inappropriate protective investigations and are more likely to accept assistance. Family problems can be comprehensively assessed and appropriate services put in place to address them. The new risk assessment schedules, however, have been found to take up a lot of staff time and lead to a more rigid response to families and their effective use is dependent on appropriate training. Most importantly, there needs to be adequate investment in the resources available to support families, otherwise an approach based on central intake and a differentiated response system merely improves investigatory processes with limited impact on clients (Tomison and Poole 2000; Tomison 2002).

Four types of risk assessment system are used in the US:

• The Matrix approach which contains 16 to 35 factors which describe a level of child, parent or family functioning each of which is rated on a 3- to 5-point scale in terms of its contribution to low, moderate or high levels of risk to a child.
• The empirical predictors method which focuses on a small number of risk factors more predictive of child abuse or neglect. The child, parent or family characteristics associated with child abuse or neglect are considered but not included in the final set of risk factors unless they actually predict the re-occurrence of one or more types of child abuse or neglect.
• Family assessment scales where assessment of child and family functioning is the primary focus of the instrument rather than identification of risk factors. The family is rated at multiple points during casework with a view to identifying risk factors as well as family strengths and resources.
• Child at risk fields which use an ecological approach and are organised around five forcefields - child, parent, family, maltreatment and intervention. A series of open-ended questions and rating scales are used to help workers identify risk influences that may be operating in the family situation. This is one of the most comprehensive forms of risk-assessment systems available in that it considers a variety of risk influences, helps workers make decisions about initial safety and promotes the use of risk assessment throughout the entire casework process.

At least 42 states have now adopted risk-assessment schedules. There is evidence that they can be helpful but also that they vary considerably in their definitions, purposes and the quality of evaluation (Dent 1998 Haggell 1998). DePanfilis and Scannapieco (1994) recommend that controlled studies which compare safety outcomes between groups of maltreated children whose caseworkers use models for assessing risk and safety and those who do not use any particular model or criteria should be carried out.

Risk assessment tools have been used far less in the UK although there is currently interest in developing them. There has also been very little research into the effectiveness of different ways of assessing risk in the UK in comparison with the US where risk-assessment models have been assessed for their effectiveness in terms of validity and reliability (Sargent 1999). The Department of Health have recently introduced a framework for the Assessment of Children in Need and their families (Department of Health et al 2000). Research is currently being carried out into the effectiveness of the framework. There are to date no similar national guidelines in place in Scotland although work has been taking place in relation to assessment frameworks. For example, the Assessment Development Project, a joint initiative between Dundee University and local authority social work departments. The project began following an evaluative study of assessment in cases of neglect which indicated the need for a flexible framework for assessment with guidance on how to move from the gathering of information to the development of a purposeful plan of intervention (Daniel 1999a). The Glasgow Assessment framework (Glasgow Assessment Framework Guidance Notes (undated)) is being devised so that all assessment activity which involves services to children can be considered within a single framework. It is based on the Department of Health framework but has been adapted to local circumstances.

(ii) Family support

The child protection systems of Sweden, France, Germany and Belgium have a much stronger emphasis on family support and mediation than those of the UK, North America and Australia. In Sweden and Belgium child protection is rooted in traditional social policies that seek to provide social assistance and public services on a comprehensive basis (Hill 2002). In Sweden the system works in partnership with parents, as part of the general system of social welfare offered as a right, voluntarily and with resources to support families. Social workers usually have good relations with service users. An example of the extent to which professionals work with families to keep them together while ensuring appropriate care is the possibility of housing whole families together for four months for assessment (Khoo et al 2002).

In Germany the concept of family rights is very strong. Judges are often reluctant to order the removal of a child from the family home against the wishes of parents even where the reasons for doing so are compelling. Indirectly this forces local authorities to allocate more resources to preventative work. Germany's Children and Youth Services Act has been successful in persuading local authorities to allocate a greater share of resources to family support work. The emphasis in family support is on helping the family to help themselves by identifying problems and building on strengths. The emphasis on family rights has been successful in reducing the number of children in public care but there is concern that this is at the expense of the rights of the child to protection (Buchanan 1996).

Formal intervention in the French system is framed in terms of a package of support and education rather than child protection. The Children's Judge acts as a kind of case manager combining a judicial, therapeutic, social and moral function. S/he rarely uses his or her authority to impose measures, but instead to develop trust and negotiate outcomes with professionals and families. Social workers together with the Inspecteur or the Children's Judge are freer to take risks to keep the family together. There is an investigation of risk as in the UK but this appears to result in more preventive action rather than investigation and surveillance and parents seem less fearful of the system (Buchanan 1996, Cooper 2002).

(iii) The confidential doctor service

The confidential doctor service which is predominant in the Netherlands, Belgium and parts of Germany offers an alternative and therapeutic approach to dealing with child protection. It was first developed in the Netherlands and is essentially based on the notion that parents with difficulties, or those who have abused or neglected their children, should be able to come of their own free will, to an agency which they can be confident will give them help without the risk of being judged or prosecuted (Borthwick and Hutchinson 1996; Madge and Attridge, 1996; Marneffe 1992).

The confidential doctor centres in Belgium are located in hospital settings and directed by a consultant psychiatrist leading a multi-disciplinary team of professionals including social workers, psychologists, nurses, speech therapists and health visitors. Services offered include crisis intervention; telephone counselling; child, individual, couple and family therapy; and residential accommodation in the hospital. The centres also offer support and counselling to professionals involved in child protection work, together with training, information and research. The aim is to help parents acknowledge their action and take responsibility for not harming their children in the future. Emphasis is put on supporting the non-abusing parents' capacity to protect the child. Families are followed up over a considerable period of time. There are a high number of self referrals to the confidential doctor service. Self-referrals or referrals from the parents themselves make up more than 30 per cent of cases. Incidence of reabuse has been found to be low (Borthwick and Hutchinson 1996; Madge and Attridge 1996; Marneffe 1992). A number of concerns have, however, been expressed about the confidential doctor system:

• the child's interests might be subordinated to the parents' rights and wishes;
• children might undergo continuing abuse while agencies seek to work with their families; and
• family therapy may not address issues of power within families, particularly power imbalances related to gender (Hill 2002).

(iv) Family Group Conferencing

Family Group Conferences were originally used as an alternative way of dealing with juvenile justice. They have since been used in the area of child protection where they operate on the basis that extended family networks are the prime repositories for creative solutions to the problems of child abuse and neglect. The Family Group Conference (FGC) was developed in New Zealand. It is a legal process based on traditional Maori decision-making practices which brings the family and the state together in a shared decision-making process. Established under the Children, Young Persons and their families Act 1989 Family Group Conferences were an attempt to address the cultural gaps in service delivery and to place the family, particularly the extended family, at the heart of the child protection process. The Act moved decision making about children in need of care and protection and young offenders from the court room and professional office to the Family Group Conference, reduced the power of social workers and increased the role and authority of the child's extended family (Connolly and McKenzie 1999; Swain 1995).

An FGC is convened by a Care and Protection co-ordinator and has three phases:

• information sharing;
• private deliberation; and
• reaching agreement.

During private deliberation the professionals withdraw and the family consider whether the child is in need of care or protection and make decisions on the basis of this discussion. Only a very small percentage of FGCs fail to reach agreement and when they do not, dispute is usually between family members, not between family and professionals. Families from other cultural groups as well as Maori have welcomed the opportunity to be involved in decision making about their children. There has been a world-wide interest in the FGC approach as it seems to offer a vehicle for creating a genuine partnership between families and services and a strong likelihood of children maintaining links with their families. The FGC model is increasingly being used in Canada, Australia and the US and there have been a range of local pilot FGCs in the UK, for example, the Hampshire Social Services Department pilot (Connolly and McKenzie 1999; Swain 1995; Marsh 1996).

The benefits of using FGCs include the healing of family rifts, indications of a lower reabuse rate and the potential for services to have an improved public image. Gilling et al (1995) found that the FGC process was a positive experience for some family members and assisted them in addressing care and protection difficulties. Family Group Conferences are not necessarily beneficial for every child, however, and a number of problems have been associated with their use:

• family members are not always given sufficient information about the circumstances of the care and protection situation or about the FGC process;
• the assumption that family placement is inherently better than non-family placement is not universally valid and the automatic inclusion of family is not always a straightforward matter: extended family networks might not exist or someone may be alienated from family members; the FGC may arouse previously dormant family problems or disputes and protection for vulnerable family members may need to be provided; the power of more dominant family members may need to be challenged otherwise less assertive but important family members may not be heard;
• the family decision-making process enables children's views to be heard but within some traditional indigenous practices, for example, some Pacific Island cultures children's opinions are not privileged and disagreement with adults is not necessarily sanctioned;
• there have been concerns about inadequate resourcing of FGCs. The process can be expensive in terms of getting family members to meetings and in the need for a range of supportive services and programmes to enable families to succeed with their new responsibilities. It is difficult to say whether FGCs save money but the fact that fewer families are referred to court would suggest savings are likely. There may also be savings in terms of placements with families instead of outside placements; and
• there are a lack of effective systems for monitoring the implementation of conference decisions and for taking action where the decisions are not implemented as agreed.

It has also been suggested that children and young people's interests will be compromised or that families involved will be dysfunctional and unable to make decisions but the experience of FGCs has found that most of these criticisms are of limited significance (Connolly and McKenzie 1999; Swain 1995).

Initial findings from research in the UK suggest that Family Group Conferences may be useful in that they are based on a 'strengths' model, seeing families as having positives to offer rather than a 'deficit' model which assumes that families are not fit to plan for their child's welfare (Morris, 1996). Families seem to make safe plans and experience ownership in the plan. People who have experienced child protection conferences and FGCs point to stark differences between the two. They find conferences intimidating and do not like being with strangers who talk about their lives. FGCs make them feel more at ease because it is just family rather than strangers. The Hampshire research found that:

• 73% of people felt positive about being asked to participate;
• neutral ground away from social services buildings and family homes was felt to be important by nearly everyone;
• the large majority appreciated being left to deliberate in private without professionals involved;
• 80% of people felt they were listened to;
• 37% would have liked more information about the FGC process as they were not clear how the family meeting would work;
• 80% of people were satisfied with their plan and 12% partially satisfied; only 8% were dissatisfied;
• 86% said the FGC process was good, 10% were undecided, only 4% felt it was bad;
• the great majority (86%) of social workers endorsed the model wanting FGCs to remain available; and
• social workers reported that setting up the meetings could be very time consuming (Nixon et al 1996; Crow 1996).