The broader context

Australia has two major levels of government – Federal and States/Territories. New South Wales is the most highly populated, multi-cultural and urbanised state. The Federal Government dominates public policy in the following areas – economic policy, social security, employment policies, immigration, health insurance and family law – and provides national strategic policy roles. States and Territories, however, have wide legislative powers and social policy remits. Child welfare services are primarily state responsibilities alongside provision of childcare, education, housing, health and social care services. In New South Wales, child welfare services are the responsibility of the Department of Family and Community Services (formerly the Department of Community Services (DOCS)).

Overarching trends in child welfare services in NSW and Australia concur with Gilbert et al.’s (Reference Gilbert, Parton and Skivenes2011) account of child protection orientated systems and the new child-centred orientation. In a comparative study of child welfare systems across welfare states in the 1990s, Gilbert (Reference Gilbert1997: 3) argued that Anglophone welfare states adopted a ‘child protection orientation’ while many Western European countries adopted ‘family service traditions’. The former emphasised ‘the protection of children from harm by degenerate relatives – the child-saving approach’ and legal-forensic child protection approaches alongside restricted provision of family support services (Gilbert, Reference Gilbert1997: 232). In contrast, the ‘family service orientation’ sought to promote child welfare by addressing ‘problems of family conflict/dysfunction stemming from social and psychological difficulties that are responsive to services and public aid’ working more collaboratively with families (Gilbert, Reference Gilbert1997: 232). Social services were more extensive and professionalised. These differences were associated with overarching welfare state and family policy characteristics (ibid.). Among Anglophone welfare states, the child protection orientation reflected well-established residual state welfare traditions and more recent neo-liberal welfare state reforms as well as liberal and conservative family policies (O'Connor et al., Reference O'Connor, Shola Orloff and Shaver1998). European family service traditions were alternatively aligned with more social democratic traditions as well as European social pedagogy service provision and Continental-Conservative family support policies.

In a follow up study, Gilbert and colleagues (Gilbert et al., Reference Gilbert, Parton and Skivenes2011) found Anglophone countries ‘adopting practices that moved them closer to a family service orientation’ and European countries establishing ‘policies and practice that leaned more towards child protection’ (Gilbert, Reference Gilbert2012: 533). These trends were a response to increasing child welfare concerns, changing social needs, broadening definitions of child maltreatment and international policy and provision trends. In addition, Gilbert et al. (Reference Gilbert, Parton and Skivenes2011) reported that a ‘new child-focused orientation’ had emerged which pursued more holistic, wide-ranging policies and programmes to promote child development and well-being, especially in early childhood. They highlighted the influence of children's rights campaigns that promoted children's rights to participation, provision and protection (Gilbert et al., Reference Gilbert, Parton and Skivenes2011). Additionally, Gilbert et al. (Reference Gilbert, Parton and Skivenes2011) highlighted the significance of Third Way social investment state perspectives, which, according to Hemerijck (Reference Hemerijck, Morel, Palier and Palmer2012: 46), seek to ‘actively mobilise the productive potential of citizens in order to mitigate new social risks’ while pursuing a ‘deliberate orientation towards “early identification” and “early action” targeted on the more vulnerable new risks groups’ to reduce social problems and their costs (in the most cost-effective way) (ibid.: 51). The concerns raised in a recent New South Wales (NSW) parliamentary briefing illustrated the emphasis on early childhood disadvantages in the Australian context:

The extract above additionally indicates the social policy significance placed on issues of parenting and family functioning in relation to children's outcomes in Australia which have led to greater social investments in policies and programmes to ‘strengthen and support families’ (Daly et al., Reference Daly, Bray, Bruckauf, Byrne, Margaria, Pecnik and Samms-Vaughan2015: 15).

Australian child welfare services in the early 1990s could be described as child protection orientated in several ways. Mandatory reporting laws operated in all but one state, with legislative changes in the 1980s and 1990s progressively broadening their scope and significance (Lonne et al., Reference Lonne, Parton, Thomson and Harries2009). Legal-bureaucratic processes and decisions around substantiating and investigating child abuse and child neglect dominated state responses to reports of suspected child maltreatment (AIHW, 1993; Bromfield and Holzer, Reference Bromfield and Holzer2008). Family support and child welfare services for children and families in need and children at risk of maltreatment and neglect tended to be limited. Much provision in these areas was provided by third sector agencies which operated in fragmented ways, providing targeted social support to specific groups, children and families, and were largely delivered by low paid and low status practitioners (Carson and Kerr, Reference Carson and Kerr2014); while child protection services responded to child maltreatment and neglect concerns, and tended to provide short-term family casework orientated towards achieving ‘good enough parenting and family functioning’ to warrant withdrawal of child protection services (AIHW, 1993: 71). These features had broad historical roots, such as in terms of the British colonial legacy, residual welfare tendencies and conservative family policies (Jamrozik, Reference Jamrozik2009). Historic zealous state and charitable ‘child rescue’ practices, common in the first half of the twentieth century, also exemplified another dimension of the British colonial policies – discriminatory and detrimental treatment towards Australia's indigenous populations as well as the ‘undeserving poor’. Emphasis on parental responsibilities in relation to children and traditional gender roles led to limited public support for working mothers or state provided social care, childcare and family support services (O'Connor et al., Reference O'Connor, Shola Orloff and Shaver1998: 21). Further, healthcare and social services have long been largely provided by private and third sector agencies in Australia, supported by state funding and universal healthcare insurance (Jamrozik, Reference Jamrozik2009; Carson and Kerr, Reference Carson and Kerr2014). However, the ascending influence of neo-liberalism in the 1990s was another factor. Following a period of centre-left social welfare and family support, developments under the Labor Hawke Government (1983–91), the Federal Liberal–National Howard Coalition Government (1996–2007), pursued neo-liberal reforms with more rigour, and returned to socially conservative family policies. Social policies in this era reduced living wage agreements in the labour market, restricted state welfare provision to a greater extent, introduced conditional welfare measures, outsourced social welfare services to third sector providers, extended private sector provision such as in childcare and introduced New Public Management (NPM) reforms and cost-efficiency strategies (Carson and Kerr, Reference Carson and Kerr2014).

From the mid-1990s, however, there were increasing demands on – leading to increasing costs – and criticisms of state child welfare services across Australia. A proliferation of child protection scandals, child welfare services reviews and media campaigns pushed child welfare and child protection issues onto state and federal government agendas (HREOC, 1997; Lonne et al., Reference Lonne, Parton, Thomson and Harries2009). States were dealing with increasing numbers of child protection reports and cases, and rates of children in state care were increasing. These trends were particularly stark in NSW where the numbers of child protection reports doubled from 1999 to 2007 (Wood, Reference Wood2008). In 2005/6, official data found 8.4 children out of 1,000 were deemed ‘at risk of harm’, just over double the figure for 1999/00, which stood at 3.9 per 1,000 (AIHW, 2007: 25). Moreover, cases of child abuse and neglect had risen more among children from indigenous than non-indigenous backgrounds, with 12.9 per 1,000 children from indigenous backgrounds deemed at risk of harm compared to 4.7 per 1,000 children from non-indigenous backgrounds in 1996/97; and 44.2 per 1,000 children with indigenous backgrounds deemed at risk of harm compared to 6.9 per 1,000 from non-indigenous backgrounds in 2005/06 (AIHW, 2007: 27). From 1996 to 2004, there was an overall 45 per cent increase in the numbers of children and young people in state care across Australia (AIHW, 2004: 41). Again, indigenous children disproportionately made up a significant proportion of these children and young people.

These trends had multiple drivers, including changing social attitudes, legal changes and changes to institutional practices. However, they also reflected child welfare issues associated with socio-structural changes including widening socio-economic disadvantages. For example, the 1980s and 1990s saw increasing rates of lone motherhood, teenage motherhood, child poverty, substance misuse and diagnosed mental health problems across Australia – all of which potentially increase stresses on family relationships and parenting resources (Bromfield and Holzer, Reference Bromfield and Holzer2008). The combination of limited state welfare provision, insufficient support for working mothers, gendered family responsibilities and gender inequalities in the private and public sphere increased economic vulnerabilities for many families (O'Connor et al., Reference O'Connor, Shola Orloff and Shaver1998). For example, a study of several official large data sets found that in 1999 around 50 per cent of lone mothers were not in employment; 65 per cent of lone parent families relied on state financial support as their main source of income; and 70 per cent of lone mothers reported experiencing financial stress (de Vaus, 2004: 53). Social research in the 1990s and early 2000s demonstrated how children most at risk, such as indigenous children and youth, tended to suffer cumulative, compounding psycho-social and socio-economic adversities and issues (Daly, Reference Daly2006).

Critical analysis of child welfare services further pointed to systemic issues. Waldfogel's (Reference Waldfogel1998) account of five systemic problems facing Anglophone child protection systems was highly relevant. This emphasised issues of: over-inclusion (for example, unnecessary child protection reports); under-inclusion (such as, inadequate support and protection for vulnerable children, particularly in response to sexual abuse and exploitation); service capacity (for example, severe funding shortages; gaps in services); service delivery (such as, adversarial relationships with parents; discrimination towards poor families); and service orientation (for example, residual family support; crisis intervention orientated) (Bromfield and Holzer, Reference Bromfield and Holzer2008; Lonne et al., Reference Lonne, Parton, Thomson and Harries2009). Additional problematic issues highlighted in public reviews and research studies were practical, cultural and linguistic barriers to accessing support services; poor inter-agency information sharing and cooperation; fragmentation between state and federal services; inadequate cooperation and consistency between different professional groups and service providers; inadequate risk and needs assessments; inadequate funding; and workforce development issues (Bromfield and Holzer, Reference Bromfield and Holzer2008).

In response, states introduced several reforms. Two important features were differential response and family support reforms. Differential response reforms sought to address issues of over-inclusion and under-inclusion in the child protection system. They attempted to broaden state responses to child protection reports to incorporate more systematic referrals to, and provision of, support and services for children and families in need. New South Wales (NSW) introduced differential response reforms in the late 1990s following the introduction of the 1998 Children and Young Persons (Care and Protection) Act – the primary child welfare legislation (although in amended form) in NSW today. Changes were made so that reports of suspected child maltreatment were assessed with reference to evidence-based risk assessment tools to distinguish between ‘child welfare reports’ about children in need but not at risk of harm and ‘child maltreatment reports’ which became ‘notifications’ that were subsequently investigated (an alternative no response needed option was also possible). The 1998 Act also provided state child welfare services with legal powers to assess a child's needs and make referrals to services for the provision of social support. However, the child protection orientation remained highly dominant at this point in NSW and the 1998 Act did not provide more extensive child or family entitlements to needs assessments or service provision. In contrast other Australian states at this time were moving more towards the family service orientation, increasing the threshold for child protection concerns to ‘at risk of significant harm’ (rather than at risk of harm) and providing more extensive funding for family and child welfare services. For example, the state of Western Australia developed the ‘signs of safety’ approach and many states introduced the New Zealand Family Group Conferencing (FGC) approach (Arney and Scott, Reference Arney and Scott2013). The Signs of Safety approach sought to improve family casework with families in which there are serious child welfare concerns (Arney and Scott, Reference Arney and Scott2013). Informed by a family strengths approach, it encouraged comprehensive family engagement in thorough assessments of family risks, needs and safety followed by family casework and family support provision orientated to addressing child welfare concerns, promoting positive family and community relationships of support, and building on family strengths (Arney and Scott, Reference Arney and Scott2013). The FGC approach included the involvement of family and community members in decisions and safety plans for children at risk (ibid.).

Alongside these developments, the public health prevention approach gained ground led by campaigners and researchers. In 1991, for example, the National Child Protection Council (NCPC) was launched with a key role to develop public health child maltreatment prevention research and strategies. Informed by the medical model, ‘risk and protection factors’ research, and evaluation studies, the public health approach to child maltreatment prevention and intervention contains the following key features: ‘define and measure the problem’, ‘identify causal, risk, and protective factors’, ‘develop and determine effectiveness of interventions’, and ‘implement interventions with ongoing monitoring’ (MacMillan et al., Reference MacMillan, Nadine Wathen, Barlow, Fergusson, Leventhal and Taussig2009: 250). It also advocates the pursuit of integrated primary, secondary and tertiary prevention strategies targeted at ‘different groups and communities with varying degrees of risk’ (Bromfield and Holzer, Reference Bromfield and Holzer2008: 55). Primary prevention includes universal initiatives that ‘target whole communities in order to build public resources and attend to the factors that contribute to child maltreatment’ (ibid.: 53). This would include universal children's and family services, material support measures, public awareness campaigns around domestic or sexual violence, parent education programmes, community development and health promotion campaigns. The important feature of primary prevention is its universality – it intends to address socio-structural factors, promote social change and reduce social stigma. Secondary prevention measures, conceptualised as ‘early intervention’, target ‘children and families at risk due to the presence of one or more risk factors for child maltreatment’ (ibid.: 56). Although conceptual and theoretical perspectives are disputed, studies have highlighted, for example, the following factors as highly correlated with reported and substantiated cases of child maltreatment and child neglect: severe poverty, social isolation, substance misuse, severe health problems, domestic violence, harmful parenting and harmful cultural attitudes (Bromfield and Holzer, Reference Bromfield and Holzer2008; MacMillan et al., Reference MacMillan, Nadine Wathen, Barlow, Fergusson, Leventhal and Taussig2009). Early intervention aims to reduce risk factors and strengthen protective factors, the latter correlated with lower incidences of child welfare concerns as well as greater social protection for children and young people; and enhanced individual and community resilience in response to individual and social adversities. In relation to child welfare concerns, protective factors include access to resources and support (for example, time, material resources, human capital and social support); nurturing, responsive parenting and care; and nurturing and supportive family and friendship relationships (Macmillan et al., Reference MacMillan, Nadine Wathen, Barlow, Fergusson, Leventhal and Taussig2009). Tertiary prevention refers to remedial and protective interventions, targeted at ‘families where child maltreatment has occurred’, including state child welfare interventions (ibid.: 56). Another central commitment to the public health approach is a focus on evidence-based policies, programmes and practice in relation to initiatives and approaches where rigorous independent studies have evidenced effectiveness, and service consultations support these claims with evidence of high service satisfaction and appropriateness among service users.

In addition, the early 2000s saw the development of the national agenda for early childhood campaign which in 2009 led the Council of Australian Governments (COAG) to adopt the Investing in the Early Years Strategy (COAG, 2009a). The strategy aims to ‘build an effective early child development system’ with a strong focus on expansion of early years services including pre-school, childcare and family support provision; new entitlements to universal pre-school provision for four year olds and universal parenting support provision for parents of young children (COAG, 2009: 16) as well as greater investment in research and official data about national patterns of early childhood development, welfare and socio-economic disadvantages (Montoya, Reference Montoya2014). Early childhood policies in Australian have likewise adopted public health approaches. Increasingly, the State and Federal Governments were moving towards child-centred active social policies, social investment perspectives and social prevention strategies.

The New South Wales reforms: Families NSW and Keep Them Safe

In 1998, the NSW Government launched the Families First initiative to provide ‘a coordinated network of prevention and early intervention services to promote health and well-being among children aged 8 and under’ (valentine et al., Reference valentine, Fisher and Thompson2006: 416) supported by $52m funding for 1998–2004 (NSW DOCS, 1998; The Cabinet Office, 2002). Renamed Families NSW in 2006, the reforms were a cross-departmental initiative, including health, disability, education and housing services as well as the Department for Community Services (DOCS). Area-based service partnerships were established involving service providers to produce area-based service plans and improve inter-agency working. Investment in new services prioritised provision for pregnant women, families with young children (particularly those aged three and under), vulnerable families with young children and economically deprived areas. Families NSW introduced new universal health home visiting services and sustained (targeted) health home visiting services – the former providing home-based health visits to all new mothers within two weeks of giving birth and subsequently when children were six to eight weeks old and six to nine months old. Having a broader role, health visitors were to work in child-centred, family-focused and preventative ways, ‘picking up on the wider family needs and parental vulnerabilities’ (Thompson et al., Reference Thompson, valentine and Fisher2006: 20). Health visitors could refer families for sustained health home visiting via a referral to the DOCS. The sustained home visiting services entailed child and family health nurses, knowledgeable about maternal and child health, providing healthcare, family support and parenting advice, and promoting access to support and services (NSW DOCS, 2005). They were based on the US Nurse Partnership programme, which targeted young poor mothers, providing intensive home visiting services for up to two years (UNODC, 2009). The United Nations lists the US programme as an evidence-based family skills programme that can reduce the risk of child maltreatment and improve maternal health, child health and development, parent–child interaction, parenting and family functioning, home conditions and home environments (UNODC, 2009). The NSW sustained home visiting services were described as ‘intensive’, providing fortnightly home visits for up to two years. In addition, Families NSW funded intensive family support workers based in third sector organisations to undertake family casework with families with multiple, more complex needs based on referrals from the DOCS and Families NSW service area networks. These workers undertook case management and family casework roles to improve parenting and family functioning, as well as provide family support and access to additional services. The reforms also funded parents-as-teachers and schools-as-communities programmes, both orientated towards improving children's school engagement and pre-school learning, and delivering community-based parenting and family support (Thompson et al., Reference Thompson, valentine and Fisher2006). Further, new childcare services were established, and funding support was provided for community groups and community integrated service centres, including family centres, playgroups and peer support schemes (NSW DOCS, 1998; The Cabinet Office, 2002).

From 1999 to 2002, however, there was a substantial increases in the rates of child protection notifications in NSW. The number of notifications rose from 30,398 in 1999/00 to 55,208 in 2001/02 (AIHW, 2007: 17). The number of substantiated cases also rose from 6,477 to 8,606 over the same period (ibid.: 18). In addition, a comprehensive review by the NSW Ombudsman was highly critical of state child welfare services (NSW DOCS, 2005). In response the NSW government introduced further reforms and expanded the child protection services workforce supported by $1.2 bn of funding including greater investment in Triple P Parenting Programmes. Listed as an evidence-based family skills programme by the UN, Triple P programmes are provided on a group-based, home-based or multi-component basis (UNODC, 2009). Independent random controlled trails (RCTs) have found these programmes were effective in promoting more nurturing and supportive parenting practices for children and young people, and strengthening social support for families which in turn contributed to reduced behavioural and emotional problems for children and young people, and in improvements in child and adolescent development and well-being (Holzer et al., Reference Holzer, Higgins, Bromfield, Richardson and Higgins2006). The Brighter Futures programme was also introduced, targeted at higher need families with children eight years and under, predominantly with the following ‘family risks factors and vulnerabilities’ (as assessed by the DOCS): ‘domestic violence; parental drug and alcohol misuse; parental mental health issues; a lack of extended family or social supports; parents with significant learning difficulties or intellectual disability; child behaviour management problems; and lack of parenting skills/inadequate supervision’ (NSW DOCS, 2005). The program was delivered by state child welfare and third sector services. Families referred were allocated a case manager who, working collaboratively with families, assessed family needs, designed family support plans and brokered and organised services. Core services delivered were Triple P programmes, intensive family support services, childcare and sustained home visiting. In 2006, Parenting Responsibility Contracts were introduced to improve engagement with the Brighter Futures programme. These were contracts between parents and the DOCS in which parents agreed to engage with services and change behaviours, such as their consumption of alcohol.

In the next few years, however, the demands on the child protection system continued to intensify. Rates of notifications increased by nearly 50 per cent from 2002/03 to 2005/06 (AIHW, 2007: 23). In addition, following child protection scandals, the NSW government commissioned the Special Commission of Inquiry into Child Protection in New South Wales. Similar to the 1997 Royal Commission, this Inquiry was led by Justice Wood. Based on an analysis of NSW child protection for 2005/06 and research reviews, the Wood Report (Reference Wood2008) highlighted four critical issues. The first was the problem of over-inclusion – excessive unnecessary child protection reports which overloaded child welfare services, utilised resources and reduced the quality of initial risk assessments. These increased the likelihood vulnerable children would ‘slip through the net’ (Wood, Reference Wood2008: 23). According to the Inquiry, the problem was due to the low threshold for child protection concerns, that of child ‘at risk of harm’, set by the 1998 Children and Young Persons (Care and Protection) Act. Wood (Reference Wood2008) recommended the 1998 Children and Young Persons (Care and Protection) Act threshold for child protection concerns be raised from child ‘at risk of harm’ to at ‘risk of significant harm’. The second problem was inadequate responses to child protection reports for which Wood (Reference Wood2008) recommended a series of reforms to the ways in child protection reports were assessed and dealt with, widening to some degree differential responses within NSW state child welfare services. The third problem was increasing rates of repeat child protection reports. Wood (Reference Wood2008: 58) found 50 per cent of reports in 2005/06 related to approximately 20 per cent of the overall number of children reported. This problem was related to poor initial risk assessments as well as inadequate service responses, particularly in relation to domestic violence, poor parental mental health, housing insecurity, sexual abuse, child neglect, substance misuse and indigenous children. In particular, service provision particularly failed to engage and improve outcomes among families with complex needs, including severe poor health, substance misuse and domestic violence. Further provision in Triple P programmes, the Brighter Futures programme, sustained home visiting and intensive family support was deemed needed. The Inquiry also recommended these services be outsourced more to third sector agencies, arguing that they provide cheaper and less stigmatising services. The Inquiry (Wood, Reference Wood2008) recommended greater use of conditional welfare measures in relation to serious poor parenting concerns. The fourth problem involving gaps in early intervention and prevention services – including for fathers, indigenous communities and vulnerable young people aged nine to fourteen – was exacerbated by inadequate inter-agency working, service fragmentation, under-funding, inadequate staff training and workforce issues (Wood, Reference Wood2008).

The NSW Keep Them Safe Strategy (2009–14) (NSW Government, 2009) then took forward most of the recommendations set out in the Wood Report. The Strategy sought to refocus child protection on serving children most at risk while ‘extending and enhancing early intervention and family support services’ (ibid.: 10). It introduced the legislative changes and differential response reforms recommended by Wood (Reference Wood2008). It invested in service expansion, pledging to deliver by 2010 more universal services to children, parents and families, including community child and family centres, pre-school provision for four year olds, Triple P parenting programmes for all parents with children age three to eight years and universal maternal mental health screening. Third sector agencies became the primary providers of the Brighter Futures programme, which was expanded, as were the sustained intensive home visiting services and intensive family support services with these services also being tailored to Aboriginal and ‘culturally and linguistically diverse’ (CALD) families.

The key features of the NSW approach to early intervention and family services reforms were mirrored in wider Australian developments. The National Framework for Child Protection 2009–2020 (COAG, 2009b) responded to influential campaigns for a national child maltreatment prevention strategy based on the public health approach (Bromfield and Holzer, Reference Bromfield and Holzer2008). The National Framework also focused on national data generation and analysis, and research programmes, to monitor child welfare trends and promote evidence-based practice. It also funded state–federal family services initiatives (ibid.). Other Federal level developments echoed Wood's (Reference Wood2008) recommendations for increasing the use of conditional welfare measures as mechanisms for promoting parental responsibilities and family changes in response to child welfare concerns. The 2007 Northern Territory intervention provided a stark example of the use of conditional welfare measures in response to concerns about parental responsibilities and child welfare (Fawcett and Hanlon, Reference Fawcett and Hanlon2009).

Critical policy and provision issues

The NSW reforms, therefore, invested in new universal and targeted family services orientated towards preventing child maltreatment and promoting child welfare. In addition, to promote child welfare they sought to improve the coordination and integration of services, and strengthen community support for families. Lastly, they aimed to re-orientate frontline practice across services towards more holistic, preventative, child-centred and family-focused approaches.

In the NSW context, there was substantial expansion of services for children and families. The introduction of universal home health visiting for all new mothers brought NSW more towards European levels of services. In relation to intensive family support services, official data stated that 6,584 children were in receipt of services in 2010/11, which was reported as a significant increase on previous years (AIHW, 2012: 45).

Commitment to evidence-based policies led to considerable investment in service evaluations. Several state-funded evaluation studies reported high levels of cost-effectiveness among the new universal and targeted services. Kemp et al. (Reference Kemp, Harris, McMahon, Impani, Anderson, Schmeid, Aslam and Zapart2011) reported the findings of a randomised controlled trail (RCT) study of an intensive health home visiting service delivered to families living in a multi-cultural deprived neighbourhood in NSW. The study examined maternal, child and family outcomes among 208 mothers and their children over a thirty month period, comparing outcomes before and after receipt of the intensive health home visiting service (Kemp et al., Reference Kemp, Harris, McMahon, Impani, Anderson, Schmeid, Aslam and Zapart2011). These mothers, children and families also received the universal health home service. Outcomes among these families were also compared to outcomes among a comparison group which only received the universal health home visiting service. This study found, compared to the comparison group, the mothers who received the sustained (intensive) home visiting services had better mental health scores, were ‘more emotionally and vocally responsive’ towards their children and breastfed their children for longer 30 months after commencing the service (Kemp et al., Reference Kemp, Harris, McMahon, Impani, Anderson, Schmeid, Aslam and Zapart2011: 537). The children in receipt of the intensive service had comparatively better cognitive development and mental health at eighteen months (ibid.). In the sustained home visiting group, home conditions, access to social support and engagement with other services were areas of improvement post-intervention to a greater degree than the comparison group (ibid.). Mothers born overseas, young mothers (under twenty-five years) and poorer families benefitted from the sustained service the most (ibid.). It is possible that one of the main reasons for the improvements in the Early Childhood Development Index findings for children in NSW from 2009 to 2012 was the provision of these services. Although rates remained nearly double the state average among indigenous children, official data found overall 19.9 per cent of NSW children aged one were classified as ‘developmentally vulnerable’ in 2012 compared 23.3 per cent in 2009 (Montoya, Reference Montoya2014: 8).

Several studies evaluated the new intensive family support services. The findings of the Brighter Futures evaluation echoed common conclusions:

The impact evaluation element of the Brighter Futures study found, compared to the comparison group, that there were greater reductions in child welfare ‘risk of harm’ reports among families engaged with the programme for 12 months and more, and fewer placements of children in state care (ibid.: 105, 114). This study included an economic evaluation which reported substantial cost-effectiveness in terms of the financial costs of service delivery vs financial savings from family impacts (ibid.: 188). The programme also predominantly served the priority at risk target groups.

Studies indicated three key features of effective services. Firstly, it was important that service provision that was responsive, relevant, and appropriate to family needs and circumstances. For example, mothers in receipt of sustained home visiting valued the service because it provided ‘parenting help’, ‘convenient home visits’, ‘baby development monitoring and support’, ‘help with other children’, ‘links to community networks’ and ‘referrals to other services’ (Zapart et al., Reference Zapart, Knight and Kemp2015: 3). For these mothers, these features enabled them to ‘improve parenting knowledge, understanding and skills’ and ‘create a stimulating environment for young children’ which in turn improved child development and welfare (ibid.). In these regards, the higher and longer levels of service provision among intensive family service seemed to be associated with better outcomes for higher need children, parents and families. For example, families who participated in the Brighter Families programme for twelve months or more benefitted more than families who participated for six months or less; and those who engaged for two years benefitted the most (SPRU, 2010). However, other studies criticised the lack of flexibility with respect to levels and durations of services (Arney and Scott, 2015). Key factors associated with service provision matched to needs and circumstances were: additional services availability and service resources, flexibility and ethos. Secondly, several studies stressed the importance of the professional–service user relationship. Here fostering effective, respectful, enabling, empowering and supportive relationships with individuals and families was crucial. Kemp et al. (Reference Kemp, Harris, McMahon, Impani, Anderson, Schmeid, Aslam and Zapart2011: 543) discussed the importance of ‘helping characteristics’ and helping processes’ within positive professional–service user relationships. Practitioner skills, resources (including within longer-term intensive services the critical resource of time), appropriate caseloads (including small caseloads within intensive services) and ethos (for example, whole family engagement, including children, fathers, relatives and friends; strengths-based practice; partnership working) were all critical factors associated with positive professional–service user relationships (Arney and Scott, 2015; Zapart et al., Reference Zapart, Knight and Kemp2015). Thirdly, effective inter-agency working and multi-agency relationships enhanced service effectiveness. The Families NSW evaluation evidenced the ways in which area-based service partnerships improved access to support and services for children and families. These networks enhanced relationships, resources and information links between services, and promoted joint-planning and joint-working, such as new referral pathways and service provision possibilities between child welfare and health services or health and housing services or family support and health services (valentine et al., Reference valentine, Fisher and Thompson2006: 415). These relationships also enabled intensive services to arrange follow-up support for families where needed (SPRU, 2010). However, inter-agency working required resources (time, admin and funding) and was furnished by commitments to work in holistic, collaborative, ecological and preventative ways.

Conversely, deficiencies in these respects could inhibit service effectiveness. In this sense, the NSW reforms suffered from several implementation problems. These included inadequate service funding, for example in relation to intensive home visiting and open-access community-based children's and family centres (Thompson et al., Reference Thompson, valentine and Fisher2006). It included inadequate resources (time, personnel and funding) for training, including Triple P programmes training, and involvement in area service networks (valentine et al., Reference valentine, Fisher and Thompson2006). Further it included insufficient attention to addressing practical, cultural, organisational and linguistic barriers to accessing services, particularly those faced by Aboriginal families, disabled people and CALD families (SPRU, 2010). In addition lower levels of inter-agency engagement in some area service networks was an issue which inhibited awareness of the new services and approach among many frontline practitioners (valentine et al., Reference valentine, Fisher and Thompson2006). Lastly, service capacity and quality issues were raised in relation to third sector agencies and Aboriginal programmes (SPRU, 2010).

Three final issues will be discussed which illustrate the complexities and challenges of the public health approach in practice as well as more entrenched features of policy and provision path dependencies, and ideological influences. The first set of issues relate to the conceptual, practical and value-based complexities and conflicts around providing ‘early intervention and prevention’ services, based on deceptively simplistic notions of thresholds and types of risks, needs and concerns. Two pressing problems were firstly, agency complaints that DOCS operated high thresholds for early intervention and child protection service referrals which inhibited access to support and services; and secondly, third sector agencies nsufficiently meeting the needs of ‘higher risk’ families. valentine et al. (Reference valentine, Fisher and Thompson2006) found health visitors in pursuit of child-centred, family-focused preventative goals attempted to refer families to intensive services but found DOCS assessed these families as not ‘moderate to high risk’ and not eligible for intensive support services. Some health visitors as a consequence reported these families as having child protection concerns which was more likely to enhance access to intensive support services for families but with detrimental consequences – increasing problems of over-inclusion in the child protection system and increasingly the likelihood of more adversarial professional-family relations (SPRU, 2010). Further, valentine and Katz (Reference valentine and Katz2015: 123) found many third sector agencies defined their service role as working exclusively with ‘low to medium risk’ families, and were reluctant to support families perceived as ‘difficult to engage or at crisis point’. These researchers were critical of these judgements by agencies which categorised families in stigmatising ways and inhibited access to early intervention services.

Other issues were challenges in engaging and helping families with multiple, complex problems and adversities – a primary target group. Several studies reported higher service drop-out rates and lower beneficial impacts for ‘those who were relatively more disadvantaged socially and economically’, ‘indigenous families’ and ‘families with parental drug and alcohol problems or intellectual disability, families with children reported for neglect, and families with a long reporting history’ (SPRU, 2010: 3). These issues related to the critical points above. They also reflected service design limitations. For example, the Brighter Futures programme provided a limited case manager role for family workers who reported discontentment and who sought more therapeutic engagement with families (SPRU, 2010). Additionally, the programme relied on core services that provided short-term periods of intensive family support worker support (three to six months periods) and group parenting interventions. UK evaluations of intensive and integrated family support services in relation to child neglect, and services for families affected by parental mental health and/or substance misuse have alternatively evidenced more positive results when services are based on ‘the team around the child’ approach and include specialist health, therapeutic and social worker practitioners with flexibility, resources, support and expertise to provide tailored, multi-faceted support and interventions (Devaney et al., Reference Devaney, Canavan, Landy and Gillen2013).

This discussion points to another overarching limitation – the neglect of professional development issues. Valentine and Katz (Reference valentine and Katz2015: 125) argued the NSW reforms marginalised critical and pressing issues about workforce and professional development and training, and called for a stronger focus on investment in ‘a skilled, resourced and authoritative workforce across different sectors and services’ (ibid.). Central areas of concern were professional training, status, working conditions and pay, supervision and support. These issues were particularly pressing for third sector agencies, given their increasing role in serving higher need families. Based on their empirical study in this area, valentine and Katz (Reference valentine and Katz2015: 124) concluded ‘families with complex needs were being made the responsibility of NOGs, but NGOs were not being properly resourced to reflect these responsibilities’. These criticisms related to the limitations of the NSW approach to evidence-based practice. Arney and Scott (Reference Arney and Scott2013: 17) argue that practitioners need to develop ‘prescriptive knowledge’ – practice wisdom, expertise, skills and judgements orientated to considerations of ‘how to intervene and support this particular family’, and criticised narrow views of evidence-based practice that overly emphasised ‘descriptive knowledge’ (i.e. knowledge of population trends; programme evaluation findings).

Finally, studies reported high levels of need and gaps in service provision, particularly in terms of ethnic minority families, refugees and asylum seekers, fathers and young people as well as services for families affected by disabilities (Arney and Scott, Reference Arney and Scott2013: Thompson et al., Reference Thompson, valentine and Fisher2006, 2010). The reforms focused on new services, primarily involving mothers and young children, at the expense of engagement and services for fathers and young people (Arney and Scott, Reference Arney and Scott2013). In practice, ‘family-centred’ largely entailed mother-focused practice with evaluation studies providing scant discussion of the specific needs, experiences and impacts for fathers and men; or levels of child and youth engagement. In these ways, the reforms potentially reinforced maternal responsibilities for children; provided limited support and interventions tailored to fathers and men; and marginalised children's rights perspectives. However, there were also accounts of services targeted at fathers, such as domestic violence programmes, which effectively engaged and supported fathers; and of good practice in youth engagement (Arney and Scott, Reference Arney and Scott2013).