Introduction
As a preface to providing a cogent case for family inclusion in child protection practice, Thorpe (Reference Thorpe2008) listed some well-intentioned but regrettable historical episodes of poor child protection practice in Australia. After establishing that ‘respectful practice with parents has never been very common’ (p. 6), she referred to data showing that children and young people in care typically want more contact even with people (parents) who may have caused them harm. This is consistent with findings reported by Briggs and Hawkins (Reference Briggs and Hawkins1997) and points to the importance of keeping parents involved, even in cases requiring removal of the child from the family.
Thorpe's paper highlighted the relevance of attachment theory as the theoretical foundation of inclusive practice and the important role of an ethical approach to practice. Current practice was criticised for being disrespectful to parents, heavy-handed in the use of power, unsustainable, harmful to children and a contributor to the mental health services load. Thorpe cited alternative, positive examples of respectful practice and provided guidance about the principles and resources that would be valuable in building an alternative parent inclusive approach.
Since then, while therapeutic approaches to child abuse have sometimes emphasised the family as well as the perpetrator (e.g., Keane, Guest, & Padbury, Reference Keane, Guest and Padbury2013), family inclusive child protection at the policy and service intervention level, although increasingly advocated (e.g., Hawkins, Reference Hawkins2013; Lonne, Parton, Thomson, & Harrie, Reference Lonne, Parton, Thomson and Harrie2009; Scott, Arney, & Vimpani, Reference Scott, Arney, Vimpani, Arney and Scott2010), remains less than widespread. The reasons for this are ideological, political and practical, and tied up with policy and theory. This paper will review some of these considerations and argues that while a persuasive case for increased family inclusive practice exists, a shortage of high-quality evidence inhibits improvements and reform in the child protection system, and that the sought-after changes towards family inclusive practice will be well served by a much greater engagement with high-quality research evaluations.
Child Protection Practice
Murphy (Reference Murphy2010) has shown that mothers are sometimes simultaneously victims and perpetrators; that is, they are subjected to violence at the hands of their partners and suspected of abusive acts toward their own children. In such volatile and complex cases, critics of the existing child protection system see the mothers as open to potential re-victimisation by a child protection system that views them as perpetrators, while proponents see instead the potential for an abuse report as an opportunity for service delivery and support (Murphy, Reference Murphy2010).
Hughes, Chau and Poff (Reference Hughes, Chau and Poff2011) reported that women who experienced intimate-partner violence described the child protection services as intrusive and unhelpful. In this study, workers were described by the women as usually not listening to their concerns or offering support. Conversely, Broadhurst, Holt and Doherty (Reference Broadhurst, Holt and Doherty2012) reported ‘problems of engaging parents who display both active and passive forms of resistance’ (p. 518).
Collectively, these illustrative studies show: the potential value of including parents in child protection interventions, that attempts to help have not always been appreciated and that changes to practice are needed if service outcomes are to be worthwhile.
There is an argument that a paradigm-level change in the child protection system is required. Hansen and Ainsworth (Reference Hansen and Ainsworth2013) have noted that the Australian child protection system ‘with its heavy emphasis on mandatory reporting, the move across time to a forensic/investigative/prosecutorial model of practice, the increased reliance on practice tools to assist decision making, that is Structured Decision making (Children's Research Centre, 2009), and the risk-averse climate of child protection agencies in all states and territories’ (p. 104) is an example of ‘taking action at the wrong level’ (Watzlawick, Weakland, & Fisch, Reference Watzlawick, Weakland and Fisch1974, p. 39). According to Hansen and Ainsworth's (Reference Hansen and Ainsworth2013) argument, the mandatory reporting system has resulted in an overflow of reports that has made the system unmanageable and the reliance on a forensic/investigative/prosecutorial model has resulted in a system that seeks to identify and punish offending parents. This approach is particularly criticised for failing to recognise the impact of social disadvantage on parenting practices. Hansen and Ainsworth's pessimistic assessment of the current Australian child protection system describes it as an expensive ‘game without end’ (Hansen & Ainsworth, Reference Hansen and Ainsworth2013, p. 106).
The complaint that the Australian system results in many official enquiries at great cost (Hansen & Ainsworth, Reference Hansen and Ainsworth2013) is also evident in the United Kingdom, where there is a current debate about the value or otherwise of serious case reviews. These ‘have to be undertaken when, as a consequence of abuse or neglect, a child or young person is seriously hurt, impaired or dies, or where there are concerns about how agencies have acted to protect the child’ (Jones, Reference Jones2013, online). Jones reported that such reviews cost from £40,000 to £100,000 for more complex cases (with an average of three cases published per week). Consequences include harassment and abuse, and ‘blame piled on those who daily give their working lives to helping and seeking to protect children’ (Jones, Reference Jones2013, online). Nonetheless, the system is being embraced, with the National Society for the Prevention of Cruelty to Children (NSPCC) in the United Kingdom currently offering courses for people to improve the quality of children's serious case reviews through support and training. An NSPCC director, Carol Long, has wanted to improve the case-review approach. She acknowledged that ‘the majority of social workers do not read serious case reviews and so miss out on what they teach us’. Her suggested remedy is to ‘teach social workers some of the same critical reasoning skills that are fundamental to the training of lawyers . . . social workers need to become stronger at gathering facts, sifting information, testing hypotheses and triangulating evidence, and using experienced managers to challenge and question so that decisions are well informed and hopefully safer’ (Long, Reference Long2014, online). Is this another example of taking action at the wrong level?
In both Australia and the United Kingdom child protection systems are struggling. In Australia, criticisms of child protection workers for failing to remove abused children from violent parents reached (yet another) crisis point in July 2013 following a number of child deaths in New South Wales. A television programme (Lateline) highlighted the polemic between a child protection worker who characterised his co-workers as heroes and tried to convey the genuine concern and compassion he observed in colleagues, versus the views of relatives of a fatally abused child who saw the system as culpable for being slow to remove children from obviously abusive situations.
The Victorian Law Reform Commission's (2010) final report into protection applications in the children's court stated that there had been nine major reviews of Victoria's child protection system in the past 33 years. Such reviews are often the result of high-profile child abuse cases involving perceived failure of the protection system. The Victorian Department of Human Services report Child protection workforce: The case for change noted that entry-level child protection practitioners, who represent 50 per cent of the case-carrying workforce, have been leaving at a rate of approximately 25 per cent a year (Victorian Government Department of Human Services, 2011).
Australian Institute of Family Studies data (2013) showed 252,000 notifications to protection services in Australia in 2011–2012, that 39,621 children were in out-of-home care on 30 June 2012, that the number of children in out-of-home care has risen each year from 2000 to 2012 and that the national rate of Aboriginal children in out-of-home care is almost ten times the rate for other children.
Dissatisfaction with the child protection system among the public, the clients and amongst child protection workers themselves, the failure of multiple judicial and policy reviews over time to change fundamentally the reported problems, and evidence of the size and scope of the child abuse problem all point to the need to consider an approach that is not simply more of the same.
Research in Child Protection Has Not Been Done Well Enough
Tomison (Reference Tomison2000) noted that for child abuse prevention programmes ‘very few rigorous evaluations have been done in Australia or internationally’ (p. 3), and that ‘with a few exceptions, no systematic research has preceded the implementation of primary and secondary prevention programs in Australia’ (p. 4). His paper explained why gold-standard experimental approaches to evaluation (i.e., randomised controlled trials – RCTs) may not always be practical, and argued for a more pluralist approach to evaluation that recognises the limits of empiricism and makes room for such alternative approaches as realist evaluation (Pawson & Tilley, Reference Pawson and Tilley1998) and what he called a Comprehensive Evaluation Framework.
Within the constraints to research so well described by Tomison (Reference Tomison2000), and putting to one side the debate about the virtues of competing research methodologies, it is maintained that a failure to engage in high-quality evaluative studies will disadvantage the goal of a move towards increased family inclusive practice. While the rhetoric underpinning the virtues of family inclusive practice seems persuasive, empirical evidence of such factors as better client outcomes and reduced staff turnover would be a powerful force for change.
We know how to improve the quality of evaluation studies. The Cochrane collection is testimony to the idea that clinical activity can be significantly improved by questioning accepted practice. For example, the clinical practice of using human albumin in emergency treatments where restoration of blood volume is urgent was revised following the use of randomised control trials, which showed that the risk of death in patients receiving albumin was 14% compared to only 9% in control groups, i.e., an established clinical practice was killing people (Alderson et al., Reference Alderson, Bunn, Lefebvre, Li, Li, Roberts and Schierhout2002). The Cochrane collection web site offers training opportunities for those wishing to develop their research skills (http://www.cochrane.org/training). The alternative Campbell collection may be even more relevant as it has a focus on assessing the effects of social interventions, primarily via the use of systematic reviews (http://www.campbellcollaboration.org/about_us/index.php).
Methodological pluralism in evaluation practices should be encouraged to work around the limitations associated with any single methodology, but evaluations certainly need to progress into more sophisticated territory than simple pre–post client satisfaction measures.
What is Changing for the Better?
Benedetti (Reference Benedetti2012) has described a ‘continuous stream’ of innovations to improve child safety, which have included a pronounced reliance on evidence-based practice and on programme efficacy trials (p. 1). Benedetti's important review acknowledges the multiplicity of causal factors of abuse, the importance of theory to help understand this complexity and to guide interventions, and the need for a strategic plan to integrate programmes within services. Her review is presented in terms of eight major trends in child protection, some of which relate specifically to family inclusive practice. A discussion of each of these eight areas of child protection practices follows.
Trend 1: Advances in Neuroscience Highlight the Negative Impacts of Poor Parenting and Stress on a Child's Developing Brain
One of the many examples presented by Benedetti is identified here as it illustrates that good-quality evaluation has led to recognition. The Multidimensional Treatment Foster Care Program is a community-based programme designed to decrease problem behaviour and improve social and behavioural skills for children aged between 3 and 17. One-on-one therapy with children is used, and training is also provided to foster parents and biological parents. RCTs have demonstrated efficacy when the programme was compared with regular foster care (Fisher, Kim, & Pears, Reference Fisher, Kim and Pears2009), and with services as usual and a no child welfare system involvement group (Leve, Fisher, & Chamberlain, Reference Leve, Fisher and Chamberlain2009).
Trend 2: Social Context and Culture Can Protect the Developing Child and Strengthen Parental Capacity in Important Ways that Can Buffer Against Individual and Contextual Risk Factors
An Australian example showed, in an RCT of high-risk female caregivers compared to a wait-list control group, that mothers with a history of, or at risk of, maltreating their children who received Parent–Child Interaction Therapy for 12 weeks were ‘observed to have better parent-child interaction and reported better child behaviour and decreased stress’ (Thomas & Zimmer-Gembeck, Reference Thomas and Zimmer-Gembeck2011, p. 177).
Trend 3: Promising Community Prevention Strategies Create New Opportunities and Challenges in Intervention Design, Implementation and Evaluation
Benedetti included the Triple P system in her list of promising community prevention strategies. Triple P is a well-known parenting intervention programme developed in Australia by Sanders and colleagues (http://www.triplep.net/glo-en/home). The interventions are designed to improve outcomes for children by working with their parents to develop knowledge, skill and confidence (Mazzucchelli & Sanders, Reference Mazzucchelli and Sanders2010). Triple P stands out as having been particularly well evaluated with a very large number of research studies, including a large-scale population trial involving a large random sample of 3000 households from Brisbane, Sydney and Melbourne, with a 2-year follow-up interval (Sanders et al., Reference Sanders, Ralph, Sofronoff, Gardiner, Thompson, Dwyer and Bidwell2008). An RCT has also shown Triple P to work effectively in an online mode (Sanders, Baker, & Turner, Reference Sanders, Baker and Turner2012).
Trend 4: An Increasing Number of Federal Policy Initiatives are Directing Public Investments towards Evidence-based Programmes
The US Maternal, Infant, and Early Childhood Home Visiting program (MIECHV) will allocate $US1.5 billion worth of grants to states to implement evidence-based home visiting programmes over 5 years from 2010. By way of comparison, in April 2012 the Australian government Department of Social Services announced an additional 3.1 million dollars to deliver projects which will help protect Australia's children, in addition to the $A63.1 million committed over 4 years under the National Framework for Protecting Australia's Children 2009–2020. With some exceptions, the list of funded programmes (http://www.dss.gov.au/our-responsibilities/families-and-children/programs-services/protecting-australias-children/protecting-australia-s-children-funding-a-further-31million-to-protect-australia-s-children) does not appear to emphasise evidence-based programmes or rigorous evaluation.
Trend 5: New Research Findings Continue to Underscore the Importance of Addressing the Needs of New Parents and Young Children
Benedetti reported the outcomes of many large-scale visiting schemes (often derived from the well-known Hawaii Healthy Start programme which started nearly 40 years ago). RCT-level data showed mixed results, however, with no difference in intimate-partner violence, for example, between treatment and control groups in the Hawaii Healthy Start programme, but significantly fewer acts of serious child abuse from mothers in the Healthy Families New York programme. The lack of consistent findings underlines the importance of undertaking research to differentiate successful from unsuccessful programmes.
Trend 6: Implementation Science Offers Programme Managers Effective Research Frameworks to Monitor and Strengthen the Service Delivery Process and to Improve the Odds of Replicating Model Programmes with Fidelity and Quality
The gap between scientist/researcher and practitioner is well recognised in the social sciences, with scientists complaining that research results are ignored in the field and practitioners complaining that research isn't relevant to real-world problems, that problems encountered in the field are not adequately investigated by scientists or that the results of scientific studies cannot be translated into practice. For some time attempts have been made to bridge this gap in psychology (e.g., Kazdin, Reference Kazdin2008) and in social work (e.g., Mitchell, Lunt, & Shaw, Reference Mitchell, Lunt and Shaw2010). The study of knowledge transfer has developed (Hoeijmakers, Harting, & Jansen, Reference Hoeijmakers, Harting and Jansen2013) and fields as diverse as human immunodeficiency virus (HIV) research (Pinto, Wall, & Spector, Reference Pinto, Wall and Spector2014), autism (Parsons et al., Reference Parsons, Charman, Faulkner, Ragan, Wallace and Wittemeyer2013), childhood practice (Wood, Drogan, & Janney, Reference Wood, Drogan and Janney2013) and education (Klingner, Boardman, & McMaster, Reference Klingner, Boardman and McMaster2013) have described what it takes to scale up and sustain evidence-based practices.
Partnerships between practitioners and clinicians in psychology (e.g., Barkham, Hardy, & Mellor-Clark, Reference Barkham, Hardy and Mellor-Clark2010; Castonguay, Locke, & Hayes, Reference Castonguay, Locke and Hayes2011) offer a model for cooperative evaluative endeavours, and community-based participatory research also offers opportunities to improve knowledge translation (Wallerstein & Duran, Reference Wallerstein and Duran2010).
Some attempts have been made to address the research–practitioner gap in child protection (e.g., Chagnon, Pouliot, Malo, Gervais, & Pigeon, Reference Chagnon, Pouliot, Malo, Gervais and Pigeon2010). Application of research findings in the field has sometimes been very problematic, as illustrated by a study of the use of decision-making tools by child safety staff in Queensland (Gillingham & Humphreys, Reference Gillingham and Humphreys2010).
Benedetti cited Aarons, Hurlburt and Horwitz (Reference Aarons, Hurlburt and Horwitz2011) as having produced the first implementation model for child welfare practice, noting that the work provides a strong conceptual framework for child welfare practitioners.
Fidelity research is also central to the integration of research and practice, and Benedetti cites the work of Gearing et al. (Reference Gearing, El-Bassel, Ghesquiere, Baldwin, Gillies and Ngeow2011), which provides a review of the past 30 years of fidelity research, and the work of Kaye and Osteen (Reference Kaye and Osteen2011), which addresses some of the challenges in measuring fidelity in child welfare systems.
Trend 7: Maximising Population-level Change Requires New Understanding of How to Construct and Sustain Effective State Systems, Local Community Collaboration and Robust Community-based Organisations
Benedetti reviewed some of the literature on organisational change and inter-agency cooperation. Australian Triple P research from Sanders and Murphy-Brennan (Reference Sanders and Murphy-Brennan2010) was notable as a prototype for how large-scale dissemination of evidence-based systems can be achieved.
Trend 8: New Technologies Offer Important, Cost-effective Opportunities for Advancing our Reach into New Populations and Supporting Direct Service Providers
Benedetti noted examples of the use of technology, such as iPhones to record videos of home visitors (Jabaley, Lutzker, Whitaker, & Self-Brown, Reference Jabaley, Lutzker, Whitaker and Self-Brown2011) and laptops for home-visiting data collection (O’Connor, Laszewski, Hammel, & Durkin, Reference O’Connor, Laszewski, Hammel and Durkin2011).
Australian research, again from the Triple P programme, was cited as showing (in an RCT) the effectiveness of an educational television programme on children's behaviour and a range of parenting outcomes, including dysfunctional parenting, parental anger, depression and self-efficacy (Calam, Sanders, Miller, Sadhnani, & Carmont, Reference Calam, Sanders, Miller, Sadhnani and Carmont2008).
Other technological innovations included a text-messaging intervention called Text4Baby.org. Expecting mothers can sign up and receive free texts, from the prenatal period until the baby's first birthday. Benedetti recommends the development of an online community and parent resource, including a video library, based on the approach taken for maths and science education by Khanacademy.org.
Conclusion
Family inclusive child protection practice has a sound theoretical and value-based rationale, and offers a remedy to some of the failures observed in current practice, with the potential to improve client outcomes and staff job satisfaction. The pace of adoption of this model may be enhanced if a team approach between researchers and practitioners is able to develop good-quality evaluation designs, which might produce objective data for the consideration of bureaucrats and those with the power to influence policy and practice. The success and international recognition of the Triple P programme seems due, in no small part, to its focus on high-quality evaluation. Methodologically pluralist, but rigorous evaluation work, especially studies that combine the talent of frontline workers and researchers, may be strategic.
