Child Abuse and Neglect Research: Past and Present

Some time ago, the broad field of child abuse and neglect research was described as methodologically messy (Socolar, Runyan, & Amaya-Jackson, Reference Socolar, Runyan and Amaya-Jackson1995) and largely irrelevant to key policy questions facing the field of child welfare (Besharov, Reference Besharov1981). The weakness of the research in this area pertained to collective research and methodological factors such as variations in the definition of child abuse and neglect, poor study design, measures with no appropriate psychometric properties, lack of taxonomic delineation of abuse and neglect, issues of contributing risk factors; causality and reporting biases, as well as constraints on the generalisation of research findings (Besharov, Reference Besharov1981; Briere, Reference Briere1992; McGee, Wolfe, Yuen, Wilson, & Carnochan, Reference McGee, Wolfe, Yuen, Wilson and Carnochan1995; Socolar et al., Reference Socolar, Runyan and Amaya-Jackson1995). However, considerable advances – and some progress – have been made to address some of these gaps; in particular, appropriate screening methods (Hoft & Haddad, Reference Hoft and Haddad2017); addressing measurement issues and methodological limitations (Camilo, Garrido, & Calheiros, Reference Camilo, Garrido and Calheiros2016; Fallon et al., Reference Fallon, Trocmé, Fluke, MacLaurin, Tonmyr and Yuan2010); trauma-informed practices and therapeutic care (Fratto, Reference Fratto2016; Jackson, Waters, Meehan, Hunter, & Corlett, Reference Jackson, Waters, Meehan, Hunter and Corlett2013); advances in understanding the neurobiological and psychosocial sequelae of child abuse and neglect on brain development (Beal & Greiner, Reference Beal and Greiner2016; Kaufman & Charney, Reference Kaufman and Charney2001; Perry, Reference Perry2009; Teicher & Samson, Reference Teicher and Samson2016; Thomason & Marusak, Reference Thomason and Marusak2017; Weber, Jud, & Landolt, Reference Weber, Jud and Landolt2016); as well as practical intervention and prevention activities and programmes (Barth, Reference Barth2009; Constantino, Reference Constantino2016; James, Reference James2000; Lane, Reference Lane2014; MacMillan et al., Reference MacMillan, Wathen, Barlow, Fergusson, Leventhal and Taussig2009).

However, the aforementioned citations are largely non-Australian-based clinical research studies. Within the context of Australian child welfare systems and related practices, there is a poor evidence base for the various issues pertaining to children in out-of-home care, and significant research gaps in key areas (Bromfield et al., Reference Bromfield, Higgins, Osborn, Panozzo and Richardson2005; Bromfield & Osborn, Reference Bromfield and Osborn2007; Cashmore & Ainsworth, Reference Cashmore and Ainsworth2004). The key question remains as to the validity of generalising internationally based research findings to Australian children and families. Similarly, of critical importance is the argument that in terms of Aboriginal and Torres Strait Islander children and families ‘there is relatively little scope for international research to inform direction in this area due to the unique cultural needs’ (Bromfield et al., Reference Bromfield, Higgins, Osborn, Panozzo and Richardson2005, p. 25). Nevertheless, it is important to systematically examine and analyse the extent to which international pioneering research advances and emerging Australian-based research has been influential in informing and shaping the development of effective child protection practices and clinical service delivery within Australian child welfare systems. Currently, there has been no systematic attempt to synthesise the available research literature and related policies to address this question.

Effective Interventions and Programmes

The development, implementation and evaluation of evidence-based practices (EBP) for children in out-of-home care, as well as effective preventative interventions, remains a challenge for child welfare organisations in Australia (Frederico, Long, McNamara, McPherson, & Rose, Reference Frederico, Long, McNamara, McPherson and Rose2016; Hanson, Self-Brown, Rostad, & Jackson, Reference Hanson, Self-Brown, Rostad and Jackson2016). Further, the overall concept of evidence in protecting children in general, and EBP in social work in particular, is both complex and controversial (Appleton & Stanley, Reference Appleton and Stanley2008; Webb, Reference Webb2001). Despite this, governments, child welfare-related policymakers and funding organisations place considerable emphasis on the development and implementation of EBP and/or manualised programmes that have been endorsed as effective and impactful through rigorous research and evaluation methodologies, such as randomised controlled clinical trials (Barth, Reference Barth2008; Buckley, Tonmyr, Lewig, & Jack, Reference Buckley, Tonmyr, Lewig and Jack2014). Similarly, the National Research Agenda for Protecting Australian Children (Babington, Reference Babington2011; COAG, 2009) prioritises the focus on instrumental research into what interventions work best, and for whom.

What is the best practice is a justifiable question, yet both the implementation and evaluation of EBP in child welfare are complex processes that require in-depth consideration of various factors and levels, including the dynamics of the care system, organisational factors, service provision, method of delivery and the characteristics of the consumers (Aarons & Palinkas, Reference Aarons and Palinkas2007; Atkins & Frederico, Reference Atkins and Frederico2017). The discussion of these domains is beyond the scope of this brief paper, but generally there is a critical gap in Australian-based research on the barriers, challenges and facilitators to the development, implementation and evaluation of EBP in child welfare. Further, Aarons and Palinkas (Reference Aarons and Palinkas2007) highlight that there is even less understanding of service provider perspectives on these issues. Further research and evaluation activities in the aforementioned domains are likely to advance and progress the Australian evidence base.

Nevertheless, one of the key challenges related to EBP that faces child abuse and neglect researchers within the child welfare system is to delineate the specifics of any form of intervention and/or programme and its direct impact on the primary clients of child protection, namely the children. In other words, the systematic evaluation and comparison of practice and service models. The comprehensive report entitled ‘Supporting the Roadmap for Reform: Evidence-informed Practice’ (Moore et al., Reference Moore, Beatson, Rushton, Powers, Deery, Arefadib and West2016) utilised a rapid evidence assessment methodology and identified 33 Australian-based programmes that met the criteria of efficacy in addressing various facets of child maltreatment. Specifically, trauma-informed specialist programmes such as Take Two, a Victorian state-wide therapeutic service (Jackson, Frederico, Tanti, & Black, Reference Jackson, Frederico, Tanti and Black2009), as well as the Evolve therapeutic services in Queensland (Klag et al., Reference Klag, Fox, Martin, Eadie, Bergh, Keegan and Raeburn2016), are considered promising programmes (Bath, Reference Bath2015; Moore et al., Reference Moore, Beatson, Rushton, Powers, Deery, Arefadib and West2016); yet this field of research requires considerable translational research if the goal of effective prevention, and the provision of strength and needs-based treatment is to be progressed (Kemp, Marcenko, Lyons, & Kruzich, Reference Kemp, Marcenko, Lyons and Kruzich2014). Overall, at the current juncture, the efficacy of interventions and/or programmes in this area remains debatable.

There remains a critical need to ascertain the effectiveness of interventions and programmes. From an analytical perspective, a powerful tool to estimate practice effectiveness and/or effect sizes across interventions and programmes is the meta-analytic technique. This method allows ranking practices with minimal subjectivity and identifies the potential usefulness of interventions or programmes by accounting for factors that may impact programme efficacy. However, there has been no Australian research utilising meta-analytical techniques to systematically review the efficacy of interventions and/or programmes that have employed a specified treatment modality to ameliorate the impact of child abuse and neglect. On the other hand, from a clinical research perspective, further innovation and development in the area of intervention and practice is needed, along with methodologically rigorous evaluation activities. This is likely to require substantial investment, resources and active collaboration between child abuse and neglect researchers and practitioners/professionals.

Further, there is little Australian research about what aspects of mainstream interventions and/or programmes are replicable and flexible enough to be adapted to diverse populations and settings, such as to strengthen the social and emotional well-being of Aboriginal and Torres Strait Islander children and families (Day & Francisco, Reference Day and Francisco2013; Day, Nakata, & Miller, Reference Day, Nakata and Miller2016; Newton, Day, Gillies, & Fernandez, Reference Newton, Day, Gillies and Fernandez2015). Critically, there is an overall shortage of Australian-based research about what works with, and for, culturally and linguistically diverse groups in general, and Aboriginal and Torres Strait Islander families and children in particular (Bromfield & Osborn, Reference Bromfield and Osborn2007; Cashmore & Ainsworth, Reference Cashmore and Ainsworth2004). Many questions remain unanswered about the best ways to prevent and/or treat the impact of child abuse and neglect across various cultural and ethnic groups (Widom, Reference Widom2013). It is also plausible to argue that a ‘scientifically-endorsed’ effective treatment modality or programme may not necessarily be endorsed and/or perceived as meeting the particular cultural needs of these diverse groups. This argument warrants further research and investigation.

Data Collection Strategies

Data collection strategies in child abuse and neglect studies adopt qualitative, quantitative and/or mixed methods. They include the use of surveys, questionnaires, participant observation, ethnographic fieldwork, drawings, photography, narrative techniques, standardised measures, telephone and/or face-to-face interviews and drop-and-collect data collection practices (Fargas-Malet, McSherry, Larkin, & Robinson, Reference Fargas-Malet, McSherry, Larkin and Robinson2010; Johnson & Turner, Reference Johnson, Turner, Tashakkori and Teddlie2003). These are useful methods, but child abuse and neglect researchers operating within social welfare systems predominately rely on clinicians, practising social workers and other professionals for data collection. Generally, the data collected are derived from clinical practice; in other words, case-loads, which are not random samples and/or research participants. This significantly limits the generalisability of research findings.

In addition, professionals working with and for children impacted by abuse and neglect may use a range of inconsistent methods to collect the information that informs their assessment and intervention planning. For example, some clinical practitioners may utilise standardised measures to inform their practice and treatment planning (Jacob, Edbrooke-Childs, Law, & Wolpert, Reference Jacob, Edbrooke-Childs, Law and Wolpert2017), while others may consider these tools as a bureaucratic tick-box exercise (Badham & Minds, Reference Badham and Minds2011; Wolpert, Fugard, Deighton, & Görzig, Reference Wolpert, Fugard, Deighton and Görzig2012). It appears that methods used to collect clinically-relevant information are influenced by practitioners’ theoretical orientation, professional training and clinical experience, as well as attitudes towards the use of measures in practice (Norman, Dean, Hansford, & Ford, Reference Norman, Dean, Hansford and Ford2014). Overall, there is controversy with regard to how to measure matters related to children and their families, in a way that is clinically meaningful to inform best practice and treatment. Nevertheless, the inconsistent and varied methods of clinical data collection limit the quality and usefulness of such data for research purposes. Hence, data useful for sophisticated analyses are often not available and/or are of poor quality. This is among the many factors that limit the translation and uptake of research findings by service providers and professional workers.

Further, research and evaluation activities in the field of child abuse and neglect within Australian child welfare organisations and therapeutic programmes face various methodological limitations pertaining to study design, instruments and survey construction; as well as data-related limitations, including lack of attention to basic psychometric properties. For example, listening to children in care is a well-established priority within the Australian child welfare system, but there is a paucity of Australian-based studies with systematic attempts to obtain high quality data regarding children's experiences and satisfaction with out-of-home care-related services (Barber & Delfabbro, Reference Barber and Delfabbro2005; Holland, Reference Holland2009). However, a recent Australian-based study utilised an innovative data analysis approach to address the limitations within the available secondary data, yet also highlighted that the ‘voices of the primary clients of child protection systems are rarely heard’ (Withington, Duplock, Burton, Eivers, & Lonne, Reference Withington, Duplock, Burton, Eivers and Lonne2017). A collection of obstacles and barriers has been identified that impede systematic data collection about the views and experiences of children in the care system (Gilbertson & Barber, Reference Gilbertson and Barber2002). These include a set of limitations including recruitment difficulties, high non-response rates, missing or transient participants, lack of cooperation from social and other workers, declining to participate, placement instability, lack of follow-up by social and other workers, participants being assessed by social workers as too distressed to participate in research and/or too dangerous to interview, as well as participants not keeping research-related appointments (Gilbertson & Barber, Reference Gilbertson and Barber2002).

Despite these limitations, the Australian-based CREATE Foundation Report (McDowall, Reference McDowall2013), examined the views and experiences of over 1,000 children in the care system across jurisdictions utilising adequate data collection methodologies. A key recommendation included supporting the active participation of children in making decisions about their lives, which is consistent with the United Nation Convention on the Rights of the Child (McDowall, Reference McDowall2013). Further, from a research point of view, it is recommended that ‘governments need to improve access to children and young people in care so that research can be conducted more easily and social workers need to recognise the importance of research and make their clients more amenable to the process’ (Bromfield et al., Reference Bromfield, Higgins, Osborn, Panozzo and Richardson2005, p. 18). Overall, many of the challenges that child abuse and neglect researchers face in recruiting and engaging children in research relate to broader ‘administrative, political, legal, and pragmatic barriers’ (Berrick, Frasch, & Fox, Reference Berrick, Frasch and Fox2000, p. 119), which considerably limit researchers’ capacity to progress and strengthen the evidence base in this domain. The dominant socio-political philosophy of the care system, to protect children impacted by trauma from the perceived adverse effects of participating in research, is one of the many key factors contributing to greater difficulties facing child abuse and neglect researchers (Berrick et al., Reference Berrick, Frasch and Fox2000).

Generally, data collection strategies within the child welfare system are often derived from general surveys, anecdotal information or routinely collected administrative data, such as referral, case management and treatment service datasets (Buckley et al., Reference Buckley, Tonmyr, Lewig and Jack2014; Chipungu & Bent-Goodley, Reference Chipungu and Bent-Goodley2004). These data collection strategies and outputs can complement each other to create opportunities for innovative research and contribute to translation of knowledge into effective practices. For example, when Australian jurisdictions adopt a common data collection strategy and a consistent administrative data system, then child abuse and neglect researchers have opportunities to conduct extensive comparative analyses of various factors such as ‘caseload dynamics, entries and exits from care, reunification and adoption’ (Berrick et al., Reference Berrick, Frasch and Fox2000, p. 119).

However, the aforementioned methodologies and data collection strategies also have significant limitations and shortcomings. For example, the types of child abuse and neglect reported and parental co-occurring health and risk factors are often based on the subjective opinion of the referrer, in most cases a case manager. In addition, these may not have been identified as substantiated concerns and this is not always clearly reported. Also, ambiguities in the classification of abuse and neglect limit the validity of such administrative data, since these are based on referrers’ identification and documentation of child abuse and neglect. Further, care system priorities are to keep the child safe rather than the precise measurement and description of their past and/or current abuse and neglect experiences. In addition, the main function of administrative and secondary datasets is not research and evaluation, but to comply with record management standards and to meet the funding and operational requirements of any given organisation and/or programme within the child welfare system (Drake & Jonson-Reid, Reference Drake and Jonson-Reid1999). Nevertheless, to combat this limitation, enhancing quality data collection and availability with a focus on systematic approaches to research and data collection is one of the key factors contributing to innovative research in the child abuse and neglect domain.

Data Linkage and Opportunities for Collaboration

In Australia, the Australian Institute of Health and Welfare (AIHW) works with the States and Territories to collate and annually report on rates of child abuse and neglect notifications, investigations and substantiations (AlEissa et al., Reference AlEissa, Fluke, Gerbaka, Goldbeck, Gray, Hunter and Tonmyr2009). However, this has been reported as not being a ‘high quality surveillance data system’ (Broadley, Goddard, & Tucci, Reference Broadley, Goddard and Tucci2014, p. 16), and identified as unreliable with many data-related deficiencies and limitations (Broadley & Goddard, Reference Broadley and Goddard2014). One of the implications of these child protection data-related limitations is that they may substantially limit the ability of government and the public health community to respond to the problem of child abuse and neglect from a public health perspective (Broadley & Goddard, Reference Broadley and Goddard2014). Consequently, there is an urgent need to consistently collect relevant information that is comparable across Australian jurisdictions. In fact, Broadley and colleagues (Reference Broadley, Goddard and Tucci2014) strongly recommend a unified child protection system across Australia with consistent data collection strategies.

Consistent and comparable national child protection statistics would offer enormous opportunities for innovative research, such as a multi-method approach to data linkage incorporating national and/or state data with secondary data, such as data from law enforcement organisations, emergency and health services (in particular family and child services), employment, education, disability services and data registries available in the public domain. The Western Australian Data Linkage System is an example that has demonstrated the benefits and contributions of data linkage activities to generate knowledge and identify areas for progress in population health (Holman et al., Reference Holman, Bass, Rosman, Smith, Semmens, Glasson and Stanley2008). Similarly, the Victorian Data Linkage System could be creatively utilised to undertake innovative research using linkable data drawn from existing datasets within Victoria. Examples where such research is occurring include the proposal for collaborative efforts between Harvard University (United States) and the Department of Prime Minister and Cabinet (Australia), utilising creative research designs (e.g., Victorian-based data linkage encouragement design, Ames & Hiscox, Reference Ames and Hiscox2016) to evaluate the efficacy and clinical impact of the Berry Street Take Two programme.

Australian-based research indicates that data linkage research is a cost-effective way to enhance the utility and policy relevance of both primary data collections and routinely collected data (Brook, Rosman, & Holman, Reference Brook, Rosman and Holman2008; Roos, Menec, & Currie, Reference Roos, Menec and Currie2004). Accordingly, data linkage offers an innovative, powerful and methodologically robust approach to create a rich source of data for a fraction of the cost of studies designed to collect primary and/or prospective data. Further, enhanced collaboration between researchers, clinicians, service providers and policymakers to link and share data and information across the child social welfare organisations provide enormous opportunities to undertake translational research that ensures that research and services synergistically work together to reach and benefit vulnerable children and families (Toth & Cicchetti, Reference Toth and Cicchetti2011). However, the uptake of data linkage to enhance and progress research and translating knowledge into improved practices and policies within the child abuse and neglect field has been limited in Australia.

Translational Research on Child Abuse and Neglect

Despite various definitions, the core concept of translational research is translating research into practice (Woolf, Reference Woolf2008), including attempts to bridge the gap between understanding and intervention (Gunnar et al., Reference Gunnar, Fisher, Dozier, Fox, Levine, Neal and Vazquez2006). In terms of child abuse and neglect, researchers may draw upon translational research to better understand (a) multiple risk factors, (b) the impact of abuse and neglect on basic developmental processes, and to (c) advance knowledge pertaining to the development, implementation and evaluation of empirically and culturally informed intervention strategies for children and families most in need (Toth & Cicchetti, Reference Toth and Cicchetti2011). Generally, translational research includes two interrelated areas of research, namely T1 and T2 (Woolf, Reference Woolf2008). T1 refers to the application and translation of research findings derived from basic research (e.g., risk factors) to inform and develop new methods of assessment, diagnosis, early detection and prevention, new treatment options and performing controlled trials involving human participants. T2 research translates and applies clinical research findings to everyday clinical practice that is likely to influence health and policy decision making (Woolf, Reference Woolf2008). It is argued that the field of child abuse and neglect in general, and prevention in particular, is likely to be best advanced through a national translational research framework.

There remains a critical need to ‘bridge the gap between basic and applied research and its application to problems of clinical importance’ (Toth & Gravener, Reference Toth and Gravener2012, p. 135). In terms of Australian out-of-home care research, there appears to be few examples of T1 and T2 research, with a considerable gap in the strategies linking T1 and T2 to inform practices and related policies. Further, more than a decade ago Cashmore, Higgins, Bromfield, and Scott (Reference Cashmore, Higgins, Bromfield and Scott2006) highlighted significant gaps in Australian out-of-home care research and strongly highlighted the urgent need to develop a national research agenda with adequate resources and funding, to support multi-site cross-jurisdictional studies, as well as enhanced collaboration between researchers, policymakers and practitioners. In fact, both child abuse and neglect researchers and practitioners are urged to ‘stretch beyond traditional boundaries’ (Toth & Gravener, Reference Toth and Gravener2012, p. 136). For example, as a social worker, Jennifer Lehmann's (Reference Lehmann2015) narrative reflections on out-of-home care stretch traditional clinical boundaries and hence are sometimes perceived as rule breakers ‘in order to achieve excellence’ for these children and young people and their families (Broadley, Hunt, & Goddard, Reference Broadley, Hunt and Goddard2015, p. 288). Similarly, child abuse and neglect researchers need to stretch the research boundaries to innovatively explore the many unanswered questions in this field.

Role of Funding Source

Department of Health and Human Services Victoria, Australia, funds the Take Two programme.