Identifying Families with Multiple Problems ... - Oxford Journals

British Journal of Social Work (2009) 39, 418–434 doi:10.1093/bjsw/bcm151 Advance Access publication January 21, 2008

Identifying Families with Multiple Problems: Perspectives of Practitioners and Managers in Three Nations Trevor Spratt and John Devaney Dr Trevor Spratt is Director of the Batchelor of Social Work (Relevant Graduate Route) and a Senior Lecturer in Social Work at the School of Sociology, Social Policy and Social Work at Queen’s University, Belfast. Dr John Devaney is a Lecturer in Social Work at the School of Sociology, Social Policy and Social Work at Queen’s University, Belfast. Correspondence to Dr Trevor Spratt, School of Sociology, Social Policy and Social Work, Queen’s University Belfast, 6 College Park, Belfast BT7 1LP, Northern Ireland. Email: [email protected]

Summary Whilst child welfare systems in the United Kingdom, Australia and the United States may share a number of common goals, they are not designed to identify families with multiple problems. Where system output measures have been utilised as proxy measures to detect such families they indicate the presence of families in the population served by child and family social work. In interviews with practitioners and managers working within contrasting welfare systems, we explore how families with multiple problems are identiifed, what repsonses they currently recieve and how their needs might be better met. Keywords: child welfare, child protection, multi problem families, system output measures, international comparisons

Introduction In the development of child and family social work over the past half-century, we can observe a tendency to capture the conditions of families within classification systems which reflect both ideological constructs of the issues and bureaucratization of agency responses (Spratt and Houston, 1999). In the course of this development, a certain commonality of language has emerged to permit a conversation across national

# The Author 2008. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved.

Identifying Families with Multiple Problems 419

boundaries. Broadly conceived, we can discuss families in certain ways which distinguish between those who are failing as a result of structural or environmental conditions (requiring support in order to achieve acceptable outcomes for their children); those who, for reasons of individual pathology or deviant lifestyle, may endanger their children (requiring a combination of social policing and support); and families who are unable to care for their children and require the state to share parental responsibility. Whilst child welfare systems reflect these particular priorities of the state (Gilbert, 1997), the perspectives of social workers within differing welfare regimes are not entirely shaped by local agency orthodoxy or particular service user classification systems. Rather, social workers make sense of agency strictures in their practice worlds, wherein dominant governing ideas may be tested against understandings of families, garnered from observation and experience (Lipsky, 1980; Spratt and Callan, 2004). In this article, we argue that a common conceptual language of social work offers the possibility of dialogue in relation to comparisons between the operation of national child welfare systems in recognizing and meeting the needs of children and families. Such comparisons are important at this time, for, whilst national variations may reflect differing historical trajectories, globalizing tendencies in the field of economic and social policy (Spratt, 2007 forthcoming) may challenge services to meet the needs of those families who require long-term help to meet complex needs. The research described here outlines the views of practitioners and managers across five research sites in the UK, Australia and the USA with regard to the identification of, and response to, families with multiple problems. We have chosen to use the term ‘families with multiple problems’ to describe those families who share a range of characteristics (including drug or alcohol abuse, the presence of domestic violence and the experience of physical or mental health difficulties on the part of one or both of the parents) which may result in repeated incidents of child maltreatment (English et al., 2000) or create environments detrimental to children. As noted by Cleaver and Freeman (1995), these families are characterized by the range and depth of their problems. Such circumstances are unlikely to be adequately addressed by short-term interventions on the part of social workers. Our particular interest is to ascertain if social work practitioners and managers working in different welfare regimes are able to identify families who are known to their agencies but whose needs are complex, requiring interventions on long timescales. Our intentions are to establish whether these families share similar characteristics and to ascertain what responses may need to be developed by agencies to meet their particular needs. Whilst we limit ourselves in this article to establishing how social work systems currently identify and seek to meet the needs (along dimensions of both policy development and measurement of service outcomes) of families with multiple problems, a further paper will reflect on the primary social and economic policy

420 Trevor Spratt and John Devaney

drivers influencing a renewed interest in these children (Spratt, 2007 forthcoming).

Common goals of child welfare systems Waldfogal (2001) contends that Australia, the USA and the UK share a common Anglo-Saxon political and legal heritage that combines to make them more similar to each other than they are to other nations. However, in spite of this shared heritage, the welfare systems within each country have evolved in different ways, representing variations on Esping-Andersen’s (1990) neo-liberal regime. Child welfare systems in these industrialized nations have developed a number of common goals. These include systems designed to identify and differentiate between those families who maltreat children and those requiring family support (Spratt, 2000). Over the past ten years, there has been a move to refocus protective services to intervene at an earlier stage in a more supportive fashion, and to rebalance the reported emphasis on the protection of children from their parents, to the support of children with their parents. The refocusing on the welfare needs of families which continues to take place across the three nations cannot, however, be read as a simple return to the broad welfare perspective of the 1960s. Rather, the common goals of child welfare systems have evolved through experience and research, with regard to both greater understandings as to the differentiated nature of referred populations and appreciation of the positive and negative effects of state intervention in family life. In recent years, there has been discernable growth in interest in comparing child welfare services across the three nations (Gilbert, 1997; McDonald et al., 2003; Mallucio et al., 2001). While those families referred to child welfare systems and who have their circumstances understood as requiring child protection services are relatively easy to identify, it is more difficult to identify subgroups with particular needs and characteristics. This is because data are not routinely collected by agencies on the number of problems families experience or on the numbers of families experiencing problems. Cleaver and Freeman (1995) developed a classificatory system of families subject to child protection interventions in the UK and identified some 43 per cent of such families as having multiple problems. In Northern Ireland, Devaney (2007) found that all subjects in his study of children who were the subject of multiple child protection registrations, who remained for long periods on the child protection register or who were subject to repeated episodes of harm whilst registered were members of families with multiple problems. Essentially, such families may be said to live in


a constant state of struggle against a range of internally and externally located stressors, episodically coming to the attention of child welfare professionals.

Background to the research Our interest in the issues of how families with multiple problems may be identified and their needs met arose out of research undertaken by one of the authors in Northern Ireland (Devaney, 2004). One of the objectives of this research was to examine the career and characteristics of children who seemed trapped in the child protection system (Devaney, 2007), together with how professionals and managers involved with the child welfare system responded to the needs of such families (Devaney, under review). As part of this research, twenty-eight subjects (experienced practitioners and managers from a range of professional backgrounds) were interviewed. There being no dedicated measures to assess numbers and profiles of families with multiple problems, children’s names added to the child protection register in excess of two years or on more than one occasion were used as proxy indicators. Data emanating from this study were used in conjunction with the findings from Cleaver and Freeman’s (1995) research on the impact of social work investigations upon children and families to inform the interview schedule designed to explore these issues with social worker’s and managers in Australia and the USA by the other author. Contacts with agencies in Australia and the USA were made through academic and practitioner colleagues. The author responsible for this stage of the research was granted access to two contrasting research sites within each nation. The participating research sites were one urban and one rural site within the state of California in the USA and one urban site in the state of New South Wales and one rural site in the state of Victoria in Australia. Contacts in each of the sites were asked to facilitate interviews with social workers and managers who could respond to questions on policy and operational issues, particularly with regard to identification of, and service provision to, families with multiple problems. In all, forty subjects were interviewed across the four sites in the USA and Australia and twenty-eight in Northern Ireland. The Northern Irish interviews were carried out in 2003/04 with the USA and Australian interviews took place in 2005. In almost all cases, the interviews were digitally recorded and subsequently transcribed for analysis. Ethical approval was sought for each of the parts of the studies reported through the ethics process within the authors’ university, and, in addition, the employing authority for the staff interviewed provided permission for the individuals to partake in the studies.


Aims of the research Given its scale, the study is essentially exploratory in nature, with the intention being to begin to identify families with multiple problems within child welfare systems. Our aim is to better understand how child welfare agencies identify and provide services to families with multiple problems in ways not captured by statistical outputs. The responses of practitioners and managers will be used to identify themes, together with suggestions as to how child welfare systems may respond to the challenges of recognizing and developing services for families with multiple problems.

Description of the research sites Notwithstanding the common goals of nations identified above, the complex and nuanced nature of history, politics, legislation, policy and practice combine to create differing contexts for work in each nation. The following descriptions, whilst providing explanatory outlines, cannot fully capture the richness and difference of local conditions.

Northern Ireland The legislation governing social work practice in Northern Ireland is essentially the same as that of England and Wales, as the 1995 Children (NI) Order closely mirrors the 1989 Children Act. The agency context, however, differs in that since 1972, in Northern Ireland, health and social services have been integrated and delivered through Health and Social Services Boards (and, more latterly, Trusts). This differs from England and Wales, where health and social services are delivered separately, then later through Local Authorities under the control of elected councils. Social services have a duality of function in relation to work with children and families; they have to ensure that children are protected from suffering significant harm at the hands of their care-givers and also to identify and provide services to children whose health and developmental needs will not be met without the provision of services. Given, however, the historical dominance of child protection perspectives, child protection work has tended to undermine more recent attempts to ensure balanced delivery of functions and this tension continues to be played out in the everyday work of social workers (Spratt, 2000, 2001; Spratt and Callan, 2004).

The United States of America In recent times, the concern of the Federal government has been to ensure the provision of permanency plans for children brought into state care


through the imposition of time limits on assessment processes. The 1997 Safe Families and Children Act stipulates that children received into the care of the state should have their circumstances assessed and plans for reunification or adoption made within tightly prescribed timescales. Such developments parallel the time impositions created by welfare policies at state level and reflect a concern to cap spiralling costs associated with increased numbers of children received into care, wherein nine out of every ten agency dollars are spent on provision of services to such children (Lindsey, 2004). Whilst individual states have wide executive powers to legislate for their own system priorities in relation to child welfare matters, at the core is a child protection service designed to address child abuse. In common with other states in recent years, California has introduced legislation (2002 California Child Welfare Outcomes and Accountability Act) to promote a move away from a system that had come to be seen as punitive and adversarial (Pelton, 1998) towards one which embraces notions of early intervention, collaboration with families and promotion of supportive service provision through community-based providers. Such efforts, however, are constrained by the interplay between federal and state funding, wherein the former provides non-capped funding for children who are received into state care but capped funds for wider initiatives aimed at prevention, thus leaving state and county-level budget holders responsible for funding any expansion of these services. This creates a perverse incentive for local providers and compounds the difficulty of redesigning services to create differential response in recognition that most referred families do not meet criteria for child protection processes but may require tailored packages of services to meet differing sets of needs.

Australia Across Australian states, there has been a general move towards re-balancing child welfare systems by focusing on the need to identify and provide support to families before child protection concerns are indicated (Australian Institute of Health and Welfare, 2007). Both New South Wales (NSW) and Victoria have, as policy objectives, placed emphasis on the creation of early intervention and prevention programmes, although these are driven through contrasting organizational strategies and exist within differing legislative frameworks (Department of Human Services, 2004). Changes in the legislation (1998 Children and Young Persons (Care and Protection) Act) governing the Department of Community Services (DoCS) in NSW expanded the range of mandatory reporters and lowered the reporting threshold, resulting in greatly inflated numbers of referrals to the agency. The agency responded by setting up a centralized telephone helpline, on which staff initially assess referrals and prioritize them. DoCS is committed to prevention and early intervention


strategies through the funding of community-based services, establishment of partnerships with other providers, as well as providing its own early intervention casework service (New South Wales Department of Community Services, 2002). In Victoria, there is a move to reconfigure services to reflect the larger proportions of families requiring supportive services. This has resulted in the establishment of Family Support Innovations Projects. The state has invested to set up community-based services to divert families notified to child protection services, with the stated goal of providing earlier and more flexible services to minimize child protection re-notifications (Department of Human Services, 2004). As in California, child protection investigations in Victoria are limited (ninety days). It is recognized that such time limiting ‘does not address the highly complex and relapsing problems facing families coming into contact with Child Protection’ (Department of Human Services, 2004, p. 65).

Families with multiple problems: themes In the following sections, the views of respondents are captured in relation to their responses to the research questions. These are represented as a number of themes which illuminate the contexts in which they work and indicate possibilities in relation to future work with families with multiple problems. The origin of respondents is denoted by the use of (ni) Northern Ireland, (nsw) New South Wales, (vic) Victoria, (cal-r) California rural and (cal-u) California urban.

Remembering past times Most respondents had worked with providers of child welfare services for some considerable period of time, including some since the 1970s. Remembering times past, they characterized this period as being less prescriptive in terms of agency policy, less preoccupied with issues of child abuse and more conducive to the forming of partnerships with families: I guess when I first came in [1982] it was more the long-term work, they had families that used to stay around for ever basically and the case worker became almost like part of the family and it was a very patriarchal old sort of system (nsw). We used to have a level of services when I started as a social worker, that were just general family social workers, [they] just got swallowed up by child abuse and neglect response (cal-r). I think the people who have been around a bit longer have so much to bring to family and child care. . ..[they] have been around the block a bit more in terms of knowing the risks, can sit down and work with people (ni).


Constraints of the present: time limits By way of contrast with the past, social work today was perceived as often restricted by bureaucratic environments wherein a virtue was made out of time-limited response. It was evident that workers in child welfare organizations across all three nations worked within such constraints, these being variously represented by legal restriction, agency policy and culture: If [families] don’t reach the level that we could file on you can’t keep the case open any longer. A Family Maintenance Order is only for twelve to eighteen months (cal-r). A level 1 is sort of, go out there straight away or within 24 hours and then level 2 is within 48 hours and level 3 is 10 days, level 4 is over 10 days. . .. The manager’s job is to look at the reports and then prioritize them because you know that there’s not a whole lot that you’re going to be able to allocate (nsw). I always thought that when the Children Order came in it was all about more preventative work, but that never really happened. You know, it’s still reactive stuff (ni).

The impact of such time limits upon the ability of child welfare agencies to meet the needs of families was noted by a number of respondents, although the role played by families in creating episodic response to longerterm needs was also noted: You don’t have the time to put these band aid solutions . . . how can you change a lifetime addiction to drugs with only [a] 4 month programme? (nsw). A lot of the time they don’t take up the services offered to them, for a very long time, you know, they’re offered and offered and they go for a short while and they pull back, then there is another crisis (ni).

Constraints of the present: restricted mandates In order to control inflow, all agencies in the study had developed systems to differentiate referred populations and prioritize those who should receive a response. Even though the Australian and American systems involved cases being processed through a child protection intake system, such systems used risk assessment tools to differentiate between those cases requiring a child protection response and those requiring family support or falling outside of the agency remit. In the USA, respondents noted major developments in diversion of family support cases to providers of services based in voluntary organizations and, increasingly, within communities, albeit access to such services was through a child protection hotline or post child protection investigation:

426 Trevor Spratt and John Devaney With the differential response (we’ve actually had this for 1 year) and basically the referral comes from the screening unit, where they’ve figured out, no we’re not going to open a case but the family has agreed for services (cal-r). The only way to get to family preservation services now is to first have an investigation by emergency response so then they decide that the family needs help, like the chronic user, not so chronic though because to get into family preservation that’s supposed to be time limited to 6 months (cal-r).

This pattern was also evident in Australia; however, with a less restrictive mandate, as well as referring family support work to other providers, agencies were developing their own capacity to respond: Assertive engagement would mean not giving up . . . so it can be contact, bit of a chat – give you a call in a couple of weeks and see how it is, sometimes if you hang in there long enough there’ll be something that they actually feel that they need (vic). Our office is one of the enhanced service delivery offices where just last week the intervention team has managed to set up . . . doing intervention work, not necessarily children at risk but trying to help those families before they get continuously reported (nsw).

Features of families referred on multiple occasions Because of the time constraints on service provision together with restrictions imposed by mandates, referred populations whose longer-term needs were not captured in agency measures were recognized in a haphazard fashion (in parallel with their inferential presence in agency service measures). Where such families were identified, similar problems were noted, as one respondent put it: ‘I think it’s very much the same issues everywhere’ (nsw). When asked what the features of this population were, respondents were remarkably consistent in identifying the following characteristics: Economic problems ‘They [come from] economically challenged communities’ (cal-r). Single parenthood ‘In my experience; lone parent families’ (ni). Domestic violence ‘We get a lot of domestic violence clients’ (nsw). Drug and alcohol abuse ‘Most of the parents that we have . . . the majority of them are recovering from substance abuse or alcohol’ (cal-r). Mental health of the parent ‘[There are] a lot of mental health issues’ (nsw). Neglect ‘Chronic neglectful families’ (cal-r).

It is important to note that apart from neglect, the other forms of child abuse—physical abuse, sexual abuse and emotional abuse—did not


feature in answers because questions were focused upon the identification of those families who were known episodically to the agencies. Our particular interest was the features underlying presenting problems. Most respondents choose not to identify such characteristics in isolation, but rather saw families as exhibiting multiple features over time. ‘Chronic over a long period of time; domestic violence, mental health with mum or dad, [and] usually poverty’ (nsw). In contrast with respondents in Northern Ireland, those in the USA and in Australia noted the issue of colour as a feature of families who had multiple problems. In California, this was expressed in high proportions of children of colour in the care system. One urban Californian respondent stated that ‘six per cent of the city’s population is black whilst sixty percent of children in care are black’. Conversely, in Australia, a concern for the ‘lost generation’ of aboriginal children raised in the care of white parents as part of an assimilation policy had led to a reluctance to respond to the needs of aboriginal families through use of the care system.

Filtering and engagement processes Despite contrasting welfare regimes, there was recognition that families with the characteristics identified above required help, often as a preventative measure to ensure that they were not re-referred for more serious reasons in the future. It was apparent that respondents took differing approaches to the issue of engagement with such families. Those working in systems designed to deal with matters clearly designated as child protection were under pressure to differentiate between referred families (especially where mandatory reporting laws generated large volumes of referrals). This could result in those families who exhibited needs not meeting agency thresholds being effectively denied access to the system: We start [at the hotline] asking them questions about the family then what we’ll also be doing is scoring what the risk assessment is and if the score comes out and meets the criteria we will assign it (cal-u). The priority has been child protection . . . it’s been to assess the risk of harm reports and do a risk of harm assessment and then refer to a [family support] agency set-up—funded to deal with those families . . . I would say the focus has been very much on short term intervention (nsw).

Despite the volume of work and agency filtering systems, some respondents indicated that they would try to respond to the needs of such families either through the use of developing differential response resources or, where the agency permitted it, some direct engagement with the families. In Northern Ireland, respondents held a wider and more flexible remit to designate the children of such families as requiring support, but again cited a range of issues which made wider engagement with families problematic (see Spratt and Callan, 2004, for a fuller discussion of the issue of


worker engagement with families using a sample of social workers in Northern Ireland): Staffing levels is the big issue at the minute with social work . . . so much change of staff, change of faces for families . . . they don’t get a chance to really get to know people (ni).

Engagement is, of course, a mutual activity and may be restricted not only by agency factors, but also by negative perceptions on the part of some families in relation to the roles and functions of social workers. Respondents in all three nations noted a common antipathy amongst the general population in relation to being in receipt of social work services. Whilst such attitudes were most pronounced in America—‘It’s like the Gestapo coming to your house’ (cal-r)—they were echoed in Australia— ‘They are apprehensive; particularly if they’ve got a history with [the agency]’ (nsw); ‘I’ve been called a Dingo, with reference to the baby that was taken and never found’ (vic)—and in Northern Ireland—‘[The parents] resent the intrusion of social services, yet they can’t manage without the support of social services’ (ni). Respondents spoke of the directness and honesty required to engage such families—‘[Families] have allowed me in after the cursing and have sat and given me answers to every question in completely open ways’ (vic). ‘I don’t think people respond to people who are controlling what you do, it’s better if it’s a partnership’ (ni).

What would help? Early intervention Respondents hypothesized that earlier interventions would make a difference in the lives of families. Moreover, they were able to detect changes of philosophy within their agencies in harmony with this idea: We’re trying to build a system that can support families where there are neglect or abuse issues looming (cal-r). In terms of families that have never been reported; children under a certain age where we can look at doing some early intervention work, not necessarily children at risk but trying to help those families before the families get continually reported . . . the families will be voluntary which is a new thing for us (nsw).

Many respondents cited necessary changes to take forward these aspirations. Some of these were at a cultural level, especially in the USA. ‘The only time we take care of people is when they’re hurt, otherwise our society think that people should take care of themselves’ (cal-r). Other respondents were concerned to identify resource constraints: ‘I do feel that we need resources early, because I feel if we got in there early enough, with good resources, maybe we could change some of their attitudes and


break this cycle that I feel contributes to poor parenting or poor relationships’ (ni). And some pointed to the need for services to co-operate: ‘More money in early intervention and a lot more support in projects working in schools because it still seems to be very much a silo system’ (vic).

What would help? Long-term intervention For families already known to agencies and whose circumstances indicate that the opportunity for early intervention had been missed, the outlook was rather more pessimistic: Long term families, you know, how do we stop them from continually coming back, it’s too late for early intervention for them, we’ve got to look at them as well (vic). We had really dysfunctional families whose parents cannot manage their own lives, never mind manage their children’s lives. I think those are people that early supports may not turn around (ni).

Opinions were divided on how best to help such families, and these varied from sourcing help in the community—‘The family worker will work on trying to get the family connected with the different agencies so that they can take over as the people to add support and kind of wean them off our system and that really is the whole idea. Connecting them with the community people who can help them lead an independent life’ (cal-u)— through developing services—‘[We require] resources to back up the agencies’ viability of doing long term work because families are really chronic’ (vic)—to some form of continuing agency contact—‘I think they [referring to families] always need a level of support and a level of social work involvement, you know, just to try and keep them on the right track’ (ni).

What prospects are there for systems to develop in directions that recognize the need for early intervention or for long-term service provision?

As we have observed, the present designs of child welfare systems across the three nations have been influenced by a predominant concern to manage child protection risks. The measures designed to track progress through such systems have largely been concerned with the identification and management of child protection risks. However, as one respondent put it, ‘There are all sorts of different milestones that are meant to be achieved but they’re administrative achievements, they’re not achievements for the family or a guide as to how well the family are moving’


(vic). The recognition by respondents that present measures do not identify and track families whose characteristics cannot be captured by agency categories is to some extent shared by the welfare agencies. The movements we have noted point to recognition that the population of families requiring help from the state is more differentiated than the traditional categorizations of child protection, family support and children in state care would suggest. In the USA, diversionary strategies have been adopted, whereas, in Australia, agency remit has been softened to include a wider range of families whose circumstances fall outside a child protection remit compared with Northern Ireland, which has attempted to engage a population with longer-term needs. If the potential for change created by such recognition is to be built upon, it will need externally located drivers. Such drivers are to be found in social and economic investment models built upon requirements for market competitiveness within a global economy (Spratt, 2007 forthcoming). We can, however, detect some signs in the responses from social workers and mangers that such investment strategies may be influencing current changes in child welfare systems. For example, some respondents noted the underlying economic premise that money spent on early intervention would reap return on investment in the longer term: The research shows that if kids are at an optimum level by the time they go to school—that money put in at that end is good investment (nsw). With this shift towards outcomes in general . . . everything is about outcomes and accountability now . . . not only is it the right thing to do for children and families to serve them, but it is cost effective . . . it would be obvious that every dollar you put in you would get back (cal-u). There’s been another wave of information that’s hit New South Wales and the rest of Australia about early intervention and prevention and investment in the early years and so what is occurring now I think is a shift of this organisation . . . to move to early intervention and prevention . . . [The] research base has improved so substantially over the last decade or so about the societal and actual monetary returns to society for investing in early intervention. They’re more amenable to spending money in that direction. . . . Invest a dollar now and you save, depending on what you read, 3 or 7 or 14 or 21 dollars thirty years down the tracks (nsw).

Whilst such investment strategies were generally welcomed, some respondents were apprehensive as to how child welfare systems might adapt to meet the needs of a wider and more differentiated population: My one worry is because we’ve got such a volume of these crisis cases, that might actually swamp the early intervention team . . . we may need an increase in the availability of services (nsw).

American respondents in particular were acutely aware of the cultural and systemic obstacles to progress:

Identifying Families with Multiple Problems 431 Here parents aren’t feeling as deserving of long term investment (cal-u). You can make the case obviously that we’re paying way more money for prison and all the other costs, mental health etc—for all these kids that keep growing up with all these problems. Getting that whole mind-set to change to really go to a model—that is thinking about cost benefits, I don’t think anybody in the government sector has confidence it will get there (cal-u).

The reactions of the public, media and politicians to child deaths and the power of this to distort the priorities of child welfare systems were also noted: Even though we try to move in a direction . . . what happens is when there’s a child death or a serious injury the way that the environment, the media, the policy leaders blame the child welfare system often drives the system in turmoil to sometimes scapegoat the workers or really blame them—the blame comes down—and then what happens is more children start coming into care because workers don’t feel at ease (cal-u).

Conclusion It may be too early to identify changes in child welfare systems that will inevitably lead to a move away from narrow investigative functions and enable the development of dedicated services to children and families with multiple problems. What appears to be happening at present is a slow and incremental move towards the development of routing mechanisms that recognize something of the diversity in the population of those referred and re-referred to child welfare agencies. The particular configurations of such mechanisms reflect differences in the historical development of child welfare systems, including the various political and ideological climates in which they are sited. These have the effect of either inhibiting or promoting the ability of agencies to detect the presence of families with multiple problems and develop the necessary services and partnerships to address these. The development of measures designed to track system outputs are beginning to be complemented by those that track outcomes for children and families. Such measures, however, still reflect the historical concerns of child welfare systems to address short-term child protection issues together with those children who are received into the care of the state. In the UK, there has been attention paid to the recognition of families with multiple problems within the youth justice sector (Respect Task Force, 2006) and there has been more recent interest in locating such families within those referred to social services (HM Treasury and Department for Education and Skills, 2007); however, as in the other nations, there are no dedicated measures to identify such families. Instead, we are reliant on a diverse range of proxy measures to indicate their presence. The practitioners and managers in our study were sensitive to the presence


of such families in their respective systems, but also generally lacked the means to identify these families by means of standard agency categorization, such as ‘child protection’ or ‘child in need’. It is important to note, however, that child welfare policy environments across all three nations in our study have become more sensitive to the issue of diverse subpopulations and there is some official recognition that short-term interventions are not appropriate to serve the needs of all families. It is in the context of a wider consideration of long-term investment by the state in relation to children that we may detect the driver for future change in welfare systems (Spratt, 2007 forthcoming). The necessity to increase the percentage of people in work to meet future health and pension costs is likely to increase political interest in systems that identify populations who will require investment to prevent their children becoming economically dependent on the state (Esping-Anderson, 2003). Families with multiple needs may therefore benefit from a shift in policy which sees investment in them—although initially costly, still justified when measured against projected lifetime costs to the state. If child welfare agencies anticipate such developments, they may also be in a position to influence them by utilizing the experiences of practitioners and the work of researchers to support the case for investment and to establish common cause with economists and policy makers. The actual shape of service provisions to families with multiple problems is a more complex matter and it is likely that the particular differences in agency contexts across the three nations would mean that diverse patterns would prevail. In Australia and in the USA, it is likely that these would include the encouragement of child welfare agencies to continue to recover something of their historical role in broadening their remit to (re)include wider family support functions in addition to their child protection remit. In the UK, with the government setting targets for the eradication of child poverty by 2020 and deeming four of the 11.7 million children in the UK to be in need of extra services in order that normative health and developmental expectations may be met (Department of Health et al., 2000), economic and child and family policies are converging. Whether or not such conditions for convergence are realizable across the three nations is debatable. The respective governments of Victoria and New South Wales are beginning to recognize the necessity to develop different modes of response to families with multiple problems. Similar echoes are to be found in California, where the aspiration by state government is to move beyond traditional narrow child protection service configurations, but this remains hampered by the distorting effect of the provision of federal funding to foster-care but not to family support services. In all three nations, there exists the possibility that child welfare agencies may be by-passed as service providers for families if the span between child protection functions and new welfare responsibilities remains too hard to bridge (Spratt and Callan, 2004). In this scenario, the role of identifying families with multiple problems may


remain, but such families would quickly be re-routed to other services provided by government, community and voluntary agencies. Whilst future predictions are always fraught with uncertainties, it is nonetheless clear that child welfare systems are developing in ways that challenge the privileging of the child protection lens as a means to understand the diversity of the referred population. The concern of the authors has been to identify existing system measures as proxy indicators for the presence of a population of families that we conceptualize as having multiple problems and to seek the views of practitioners and managers as to the characteristics of such families, the nature of their needs and how these might be met. The emerging picture is characterized by a diversity of views reflecting differences in local context. Nevertheless, clear themes have emerged, demonstrating a common understanding as to the shared characteristics of these families, together with indication as to how welfare systems may better adapt to meet their needs.

References Australian Institute of Health and Welfare (2007) Child Protection Australia 2005 – 06, Child Welfare Series Number 40, Canberra, Australian Institute of Health and Welfare. Cleaver, H. and Freeman, P. (1995) Parental Perspectives in Cases of Suspected Child Abuse, Studies in Child Protection, London, HMSO. Department of Health, Department for Education and Employment and the Home Office (2000) Framework for the Assessment of Children in Need and their Families, London, The Stationery Office. Department of Human Services (2004) Protecting Children: Ten Priorities for Children’s Wellbeing and Safety in Victoria: Technical Options Paper, Melbourne, Victorian Government. Devaney, J. (2004) ‘Chronic child abuse: Failings in systems, policy or practice’, unpublished Ph.D. thesis, Queen’s University Belfast. Devaney, J. (2007) ‘Chronic child abuse: The characteristics and careers of children caught in the child protection system’, British Journal of Social Work Advance Access, 10.1093/bjsw/bcm089. Devaney, J. (under review) ‘Inter-professional working in child protection with families with long-term and chronic needs’. English, D., Wingard, T., Marshall, D. and Orme, A. (2000) ‘Alternative responses to child protection services: Emerging issues and concerns’, Child Abuse and Neglect, 24, pp. 375 – 88. Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism, Cambridge, Polity Press. Esping-Andersen, G. (2003) ‘Against social inheritance’, in Browne, M., Thompson, P. and Sainsbury, F. (eds), Progressive Futures: New Ideas for the Centre Left, London, Policy Network.

434 Trevor Spratt and John Devaney Gilbert, N. (1997) Combating Child Abuse: International Perspectives and Trends, Oxford, Oxford University Press. HM Treasury and Department for Education and Skills (2007) Policy Review of Children and Young People: A Discussion Paper, London, The Stationery Office. Lindsey, D. (2004) The Welfare of Children, 2nd edn, Oxford, Oxford University Press. Lipsky, M. (1980) Street-Level Bureaucracy: Dilemmas of the Individual in Public Services, New York, Russel Sage Foundation. Mallucio, A., Ainsworth, F. and Thoburn, J. (2001) Child Welfare Outcome Research in the United States, the United Kingdom and Australia, Washington, DC, CWLA Press. McDonald, C., Harris, J. and Wintersteen, R. (2003) ‘Contingent on context? Social work and the state in Australia, Britain and the USA’, British Journal of Social Work, 33, pp. 191 – 208. New South Wales Department of Community Services (2002) Annual Report 2002, Sydney, NSW Department of Community Services. New South Wales Department of Human Services (2005) DoCS Corporate Plan 2004/ 05 – 2008/09, Sydney, NSW Department of Human Services. Pelton, L. (1998) ‘Four commentaries: How we can better protect children from abuse and neglect’, Protecting Children from Abuse and Neglect, 8(1), pp. 120 – 32. Respect Task Force (2006) Respect: Family Intervention, available online at http:// respect.gov.uk/article.aspx?id=9072&terms=domestic+violence&. Spratt, T. (2000) ‘Decision making by senior social workers at point of first referral’, British Journal of Social Work, 30, pp. 597 – 618. Spratt, T. (2001) ‘The influence of child protection orientation on child welfare practice’, British Journal of Social Work, 31, pp. 933 – 54. Spratt, T. (2007) ‘Identifying families with multiple problems: Possible responses from child and family social work to current policy developments’, British Journal of Social Work, forthcoming. Spratt, T. and Callan, J. (2004) ‘Parents’ views on social work interventions in child welfare cases’, British Journal of Social Work, 34, pp. 199 – 244. Spratt, T. and Houston, S. (1999) ‘Developing critical social work in theory and in practice: Child protection and communicative reason’, Child and Family Social Work, 4(4), pp. 315 – 24. Waldfogal, J. (2001) The Future of Child Protection, London, Harvard University Press.

Acknowledgements This research has been undertaken with assistance from the NSW Department of Community Services. However the information and views contained in this study do not necessarily, or at all, reflect the views or information held by the NSW Government, the Minister for Community Services, or the Department.