Introduction

Responding to allegations of child abuse and neglect typically requires the involvement of workers from different agencies and professional disciplines working in concert. Systems inquiries, critical incident and child death reviews commonly highlight poor information sharing across agencies and service coordination as key issues for severe child abuse cases (e.g., Child Protection Systems Royal Commission, 2016; Community Development and Justice Standing Committee, 2008). Providing a holistic and coordinated response is commonly identified as crucial to addressing these issues and fostering improvements in child safety and child wellbeing (e.g., Child Protection Systems Royal Commission, 2016; Community Development and Justice Standing Committee, 2008; NSW Ombudsman, 2017).

This paper focuses on severe cases of alleged child abuse, in which a police investigation is required. In these cases, police and child protection statutory authorities need to work together to conduct their investigations into whether there has been criminal conduct, and whether the child is safe in their present setting. Forensic medical evaluation may be required, and ideally interventions to address any remaining problems within the family environment and therapeutic services to assist the child and their non-offending parent(s) in recovery will be put in place.

While all workers are required to operate with the best interests of children in mind, how this is interpreted will likely differ based on professionals' disciplinary background and the roles of different organisations involved. These differences in interpretation and a lack of communication and understanding between agencies and workers can result in a poor response; causing confusion and distress for children and their families. Poor communication and coordination between agencies can also have other critical consequences where child-related risks are not properly identified or managed.

To better manage the response to cases of severe child abuse, many jurisdictions have implemented practice frameworks aimed at improving collaboration and coordination across agencies and disciplines. This paper:

• provides a brief description of multi-disciplinary team responses and an overview of important characteristics of responses;
• presents a national comparison of multi-disciplinary teams/centres for investigating cases of serious child abuse in Australian jurisdictions;
• summarises the key elements of planned cross-agency responses: integration of supportive and therapeutic services, co-location of workers, and the governance structures supporting cross-agency responses; and
• presents key areas for research and policy development.

What are multi-disciplinary team responses to abuse?

A multi-disciplinary team response to alleged child abuse involves a team of professionals from different disciplines and agencies working together. The degree to which teams are integrated, collaborative or consultative will vary between models but multi-disciplinary teams will usually have a process of case review or information sharing to coordinate and plan the response across agencies. Typically, these teams are focused on the effective investigation of allegations of harm to children and on children's and families' recovery and support needs post-disclosure. The purpose of these teams and the types of agencies involved will vary between models, although for the present review the multi-disciplinary teams will typically involve the police and child protection statutory authorities. The multi-disciplinary team may also involve medical, supportive and therapeutic professionals depending on the purpose of the model. Processes and procedures are often focused on avoiding police and child protection undertaking separate investigations of the same allegations in parallel, and children's and families' therapeutic and support needs being missed as statutory agencies focus on their investigative and safety imperatives.

Cross-agency arrangements differ in the degree of integration, from broad agreements between agencies acknowledging the need to share information, to fully integrated co-located teams in specialised facilities. Where no framework for collaboration exists, agencies (e.g., police and child protection) engage in information sharing and collaboration informally. Some frameworks do provide a process for information sharing and response planning, but agencies still operate and make their decisions individually. Full multi-disciplinary teams or centres provide a framework for cross-agency decision making drawing on the knowledge and expertise of different disciplines to plan and implement an effective response. An integrated cross-agency team, created by agreements across agencies, can operate virtually (e.g., by phone), or in a shared professional and victim-focused space. The most common examples of these integrated responses internationally are Child/Children's Advocacy Centres and Children's Houses or Barnahus (for an overview of these models see Box 1).

Important characteristics of multi-disciplinary team approaches

Multi-disciplinary teams acknowledge the multi-dimensional impacts of abuse and the needs of children and families affected by abuse by bringing together workers from different disciplines and agencies to discuss, plan and carry out a response to cases of child abuse. Bringing together the disciplines and agencies involved in the response to abuse to participate in case review, joint interviewing or other types of information sharing are inherent to multi-disciplinary team models, although the nature and degree of joint processes will vary (Newman & Dannenfelser, 2005; Newman, Dannenfelser, & Pendleton, 2005).

For most multi-disciplinary team models, collaboration between workers across agencies extends beyond holding case review meetings; the co-location of workers is assumed to build connection and professional relationships between team members, while also making it easier to undertake case consultation/review and other formal parts of the program (Green, Rockhill, & Burrus, 2008; Newman & Dannenfelser, 2005). Co-location may help to integrate different parts of the response resulting in a true team-based approach associated with higher levels of collaboration (Tye & Precey, 1999).

Broadly, providing services on site may help support effective collaborative teams (Edinburgh, Saewyc, & Levitt, 2008; Newman et al., 2005) and ensure that vulnerable families receive needed services by reducing some of the barriers to successful referrals for services (Burns et al., 2004). Many multi-disciplinary teams include support service providers as part of the response, using the initial contact with children and caregivers to build rapport in order to more effectively refer to services (Kemp, Marcenko, Hoagwood, & Vesneski, 2009) and to work with families to address any barriers to accessing services (Owens et al., 2002). The independence of these support workers from police and child protection statutory authorities may also be an important factor in engaging and supporting complex families (e.g., Powell & Wright, 2012).

Case reviews are a key structure for collaboration within multi-disciplinary teams (Jones, Cross, Walsh, & Simone, 2005). Planned case review meetings allow teams to discuss and review actions on cases, while also presenting opportunities to build trust and rapport between workers (Jackson, 2012). Related to this, the use of cross-agency case-tracking systems and performance measurement allows agencies to more effectively plan and carry out their response with the full knowledge of actions by other agencies (Gragg, Cronin, & Schultz, 2006; Howell, Kelly, Palmer, & Mangum, 2004), and can frame the performance of multi-disciplinary teams across agencies (Bertram, 2008; Ells, 2000; Fargason, Barnes, Schneider, & Galloway, 1994).

Governance factors may also affect the quality of a multi-disciplinary team response. Having clear roles and cross-agency leadership (Lalayants, 2013), training and professional development (Darlington & Feeney, 2008; Haas, Bauer-Leffler, & Turley, 2011; Lalayants, 2013; Stanley, Miller, Foster, & Thomson, 2011; Szilassy, Carpenter, Patsios, & Hackett, 2013), a protocol or formal inter-agency agreement (Newman et al., 2005), and ongoing cross-agency review with procedures for conflict resolution (Barton & Welbourne, 2005; Lalayants, 2013) all seem to be important to an effective multi-disciplinary team response.

What is it that makes multi-disciplinary teams effective?

The authors undertook three reviews of the international literature and a survey of 316 US Child Advocacy Centre Directors to answer the following questions:

1. Are multi-disciplinary teams effective (Herbert & Bromfield, 2016, 2017a)?
2. What are the components of effective multi-disciplinary teams (Herbert & Bromfield, 2017b)?
3. What do Child Advocacy Centre Directors perceive as the most useful collaboration mechanisms within their multi-disciplinary teams (Herbert et al., 2017)?

There is evidence to support the idea that multi-disciplinary teams can result in improvements, particularly in criminal justice outcomes and increased referral to/uptake of therapeutic and support services (Elmquist, Shorey, Febres, & Zapor, 2015; Herbert & Bromfield, 2016, 2017a). Further, it appears that multi-disciplinary teams with demonstrated effectiveness in responding to severe child abuse allegations share many of the same characteristics (e.g., evidence-based child-interviewing protocols, information-sharing mechanisms, joint planning and independent child advocacy).

There is currently a lack of evidence comparing different types of cross-agency responses and components designed to enhance responses. What this means is that researchers don't know if there is a core set of components that are essential to effective multi-agency practice, whether model complexity can vary according to population size and needs, or if model effectiveness is enhanced with additional components. In the absence of empirical data for which components of the model are essential, the authors take the position that model components need to be carefully planned with a clear theory for why key elements are included or excluded (Bromfield, Salveron, & Main, 2017).

Comparison of characteristics of cross-agency responses to severe child abuse in Australian jurisdictions

Distinct responses within Australian jurisdictions

Table 1 provides a summary of the specialist multi-disciplinary teams or models for responding to serious child abuse, inclusive of a police investigation, operating within Australian jurisdictions at February 2017. Where more than one model or approach was identified within a jurisdiction, these have been separated into distinct responses for this comparison. Different approaches for metropolitan versus regional or remote areas were common and there were several pilot approaches being trialled. For some jurisdictions, there were minor differences in the response; these were discussed in terms of a single response. For example, in South Australia a different process for interviewing exists depending on the age and language competency of children; however, the cross-agency communication and investigation process was common across different child groupings.

Note: ^aThe WA Police were undertaking an additional pilot within their Perth-Metro response involving removing police interviewers from the joint interviewing pool and having them work from the investigations floor with Child Abuse Squad detectives. The pilot also made it possible for Child Abuse Squad detectives to observe specialist child interviews. Department of Child Protection and Family Support interviewers were able to observe interviews in this pilot but not conduct them. This has been treated as part of the Perth-Metro response for the purposes of this comparison paper as it involves only minor differences to the regular interview process.

1. General characteristics

Table 2 provides a comparison of the general characteristics of each of the distinct responses identified in Australia. These differ in terms of the degree to which the response is built around decentralised resources. New South Wales, Victoria and Queensland have the specialist resources for responding to child abuse distributed across the state. By comparison, Western Australia, South Australia, Tasmania and the Northern Territory operate responses that are centred on assets and resources in capital cities.

Only one jurisdiction (Tasmania) does not have specialist child abuse or sexual assault police. In Western Australia and South Australia, for some regional/remote cases, non-specialist police units undertake investigations.

Jurisdictions differ in terms of the scope of their specialist police child abuse units. New South Wales, Queensland and Western Australia have specialist policing units specifically for child abuse offences; other states/territories have units or groups within sex crime divisions that include adult sexual offences (i.e., Victoria, the Northern Territory and the Australian Capital Territory), or family violence (i.e., South Australia) units along with child abuse investigations.

All jurisdictions have some protocol or process in place for police and child protection agencies to collaborate on cases and share information, although jurisdictions varied in terms of the extent to which the response involves an integrated cross-agency team, and what other agencies are involved in the response. Differences also exist in the scope and stage of the cross-agency collaboration.

In terms of co-location, the Multi-Disciplinary Centres pilots in Victoria, the Multiagency Investigation & Support Team pilot in Western Australia, the Child Abuse Taskforce in the Northern Territory, and more than half of the Joint Investigation Response Teams operating in New South Wales include co-located workers. Many of the co-located centres involve separate sections or floors for different agencies, at least in part due to operational reasons. This may affect the degree to which these teams are integrated in practice.

For most of the responses, the key agencies involved are police, child protection and health. Tasmania only includes police and child protection agencies as part of their response, although responders have close contact with the hospitals and services they refer to for forensic medical examinations and counselling services. The Multi-Disciplinary Centres pilots (Victoria), Multiagency Investigation & Support Team pilot (WA), and the Wraparound response (the ACT) all have non-government agencies integrated into their responses as service providers (e.g., sexual assault counselling and casework services).

Most jurisdictions reported the use of joint investigations, except Queensland, Tasmania and the ACT; however, the nature of these joint investigations varies. New South Wales is distinct in involving police, child protection and health agencies from the point of intake, with collaboration between agencies over the course of the case. Victoria and South Australia had similar descriptions of collaboration between child protection agencies and police in response planning, information sharing, interviewing and investigation (although in SA for children under 7 years old and in some other instances, this also involves assessment by clinicians at one of SA Health's Child Protection Units. See Table 6 for a full list of conditions where the Child Protection Unit is involved).

Notes: ^aCross-agency team operating out of a single centre, which also has onsite facilities for interviews with children. ^bEleven of 22 Joint Investigation Response Teams are fully co-located. In December 2016 NSW Police announced their intention to move from shared accommodations to a model of "proximal co-location", with the Joint Investigation Response Teams operating with separate but nearby accommodation (in some cases still in the same building) in each area. ^cRegional/Remote responses are conducted by the local detective team rather than a specialist squad. ^dChild protection and police are co-located in the metro response but this does not include the statutory child protection workers who have responsibility for the case. ^eThe Australian Federal Police are only co-located in Darwin.

2. Centre characteristics

Of the centre-type approaches, all involve the co-location of police and child protection workers. The New South Wales Joint Investigation Response Team model is unique in having workers from the state health agency co-located or available nearby to attend interviews, providing a support role and facilitating referrals to NSW Health services (i.e., medical and counselling services); although this was also the situation for some South Australian cases. The Multi-Disciplinary Centres pilots (Victoria) and Multiagency Investigation & Support Team pilot (WA) are distinct in their co-location with a non-government agency providing therapeutic and support services. The Child Abuse Taskforce (NT) and the Perth-Metro response involve the co-location of police and child protection; however, in Western Australia this does not include the child protection workers responsible for responding to the case.

Note: ^aEleven of 22 Joint Investigative Response Teams are fully co-located.

All centre-based or "one stop shop" type responses have onsite interviewing suites with facilities for recording interviews; all are either specifically designed for children or designed to be victim sensitive (i.e., adult victims are also interviewed in the facilities). The Joint Investigation Response Teams (NSW), Multi-Disciplinary Centres pilot sites (Victoria), Perth-Metro (WA), Child Abuse Taskforce (NT), and Multiagency Investigation & Support Team pilot (WA) all have onsite interview suites. In NSW, all the co-located Joint Investigation Response Team sites have onsite interviewing facilities; the non-co-located Joint Investigation Response Team sites use interview suites at the local Child Abuse Squad or other community facilities. In the Multi-Disciplinary Centres pilots (Victoria), Multiagency Investigation & Support Team pilot (WA) and Perth-Metro almost all of their forensic interviewing takes place in their specialist suites. As the Child Abuse Taskforce (NT) responds to cases across the Northern Territory, interviews can be conducted in any place that is deemed to be safe and provide privacy, in addition to the interview suites at the taskforce building.

For responses that were not centre-based, jurisdictions varied in their requirement for interviews to be conducted in specialised suites. The Victorian Standard Response, Queensland statewide response, and the ACT response have specialist interviewing suites, although the specialist police in the Victorian Standard Response has provision for interviews to take place in other settings, reflecting that not all teams have access to interviewing facilities in their area. The Queensland statewide response, the Victorian Standard Response and the Regional/Remote WA response have provision to conduct interviews in safe settings. Arrangements for interviews differed based on the child's age and capacity to communicate in South Australia; children under 14, or older children with difficulty communicating, are interviewed in specialist facilities.

3. Intake characteristics

Only New South Wales has a consolidated cross-agency intake process in the form of the Family and Community Services Helpline (NSW). Matters meeting the Joint Investigation Response Team criteria are referred by the helpline to the Joint Investigation Response Team Referral Unit for tri-agency assessment - accepted matters meeting the Joint Investigation Response Team criteria are referred to a local Joint Investigation Response Team for a response. At the time of comparison, a similar scheme has recently been accepted by the South Australian Government as part of their response to the Nyland report (Child Protection Systems Royal Commission, 2016).

Most jurisdictions have a process for a discussion or review across agencies as to whether the matter should be accepted for a cross-agency response. In Queensland, the specialist cross-agency Suspected Child Abuse and Neglect team response applies to cases where Child Safety Services have made a notification and determined that coordination of a multi-agency response is required. The Child Abuse Taskforce (NT) has daily cross-agency discussions of cases received by Territory Families as to whether a matter should be accepted for a joint or police-only investigation. For Perth-Metro, Regional-Remote WA and Multiagency Investigation & Support Team pilot responses, each agency undertakes their intake process and then brings the relevant information to a strategy meeting to decide if the case receives a joint, police-only or child-protection-only response. For all other jurisdictions, agencies pass relevant referrals on to each other and agencies make individual decisions about whether to accept a matter for investigation.

The criteria for intake to a cross-agency response differ across jurisdictions. In most cases, the response is directed towards child abuse cases where child protection authorities are likely to be involved in parallel investigations. For some jurisdictions, the threshold for the cross-agency response is primarily the threshold for the involvement of the specialist child-protection/sexual-assault police team, with child protection authorities involved in a much wider range of cases. The Suspected Child Abuse and Neglect teams (part of the Queensland Statewide response) are unique in terms of accepting cases after the point of notification or intervention by Child Safety Services, with this agency determining whether a multi-agency response is required. The Child Abuse Taskforce (NT) also specifically targets complex cases (i.e., intra-familial and criminal) for a taskforce response.

4. Information sharing and case planning

Every jurisdiction has a process in place for the sharing and exchange of information across agencies, although there are differences in the agencies included in this. There are also differences in the formality around information-sharing processes and the legislation in place to allow information exchange from different types of agencies.

The Joint Investigation Response Team model (NSW) has a comprehensive process for information exchange and case discussion in their local area response protocol; this includes planning initial contact with children and families, collecting and exchanging information available about the family, planning the interview and coordinating the response following the interview. South Australia has a similar series of processes for information sharing and follow-up.

By comparison, most other jurisdictions have much more informal processes of information exchange and case discussion. Queensland, the Northern Territory, the ACT and Western Australia (all models) have similar processes of conducting strategy meetings to plan the response for each agency, with individual follow-up between workers undertaken informally. These meetings take place regularly in WA (weekly), the Northern Territory (daily), and the ACT (monthly), but can also be scheduled as needed. The Victorian models and the Tasmanian model rely on informal consultation between workers to coordinate their response, rather than arranging regular cross-agency meetings.

Jurisdictions also differ in terms of the information-sharing provisions permitted within the state/territory. New South Wales, South Australia and the Northern Territory each have comprehensive schemes allowing for the exchange of information between a wide variety of agencies when the information relates to the safety and wellbeing of children. Information sharing is more restrictive in Queensland and Western Australia where only specified government and non-government agencies are permitted to exchange information about children. In Tasmania and the ACT, information exchange is restricted to organisations (Tas.) and professionals (ACT) that deal with children, allowing them only to exchange information with the statutory child protection authority in their jurisdiction. In Victoria, the provisions are even more restrictive, allowing only the provision of information from professionals to child protection statutory authorities.

Notes: ^aPrescribed bodies include all organisations that wholly or partly provide services to children (including all kinds of educational, health and child service providers). ^bDespite the legislative change, written consent to exchange information between government and non-government agencies is still the norm (Herbert & Bromfield, 2017c, p. 52).

New South Wales has the most comprehensive training and professional development for cross-agency work, with an induction and foundation skills course run across professional and agency groups. Although few jurisdictions run cross-agency training, mostly agencies provide training and professional development on working with other agencies within their own professional groups.

5. Interviewing

Across the country, almost all interviewing of children is done by the investigating officer from the specialist child abuse/sexual assault unit. The main exception to this is in Western Australia (Perth-Metro and Multiagency Investigation & Support Team pilot), where a pool of trained child interviewers from both police and child protection agencies conduct interviews. In South Australia, age and communication capacity determine who interviews children (see Table 6), and especially vulnerable children are interviewed by a worker from the Child Protection Service (SA Health). As noted previously, in some regions within jurisdictions the investigating officer is not from a specialist child abuse/sexual assault unit.

Notes: ^aWA Police were running a pilot to better integrate their interviewing and investigation. Police specialist interviewers still conducted interviews, but the investigating officer observes, along with a Child Protection interviewer. ^bThe SA Government have committed to establishing an additional Child Protection Service unit at Lyell McEwin Hospital.

Almost all jurisdictions have provision for a recommendation for child protection authorities to be involved in interview planning and to be present for a child interview to minimise the need for additional interviews and disclosures. New South Wales is unique in that it includes NSW Health to provide specialist knowledge and input into response planning; although, in South Australia, the Child Protection Service (SA Health) is involved in conducting interviews for especially vulnerable children. Most jurisdictions outline the role of child protection authorities in their protocols as observers of the interview. This is different in WA (Perth-Metro and Multiagency Investigation & Support Team pilot) as child protection interviewers either conduct or provide feedback on interviews (see the footnotes in Table 6).

6. Support and advocacy services

The models differ as to the degree to which supportive services and advocacy are included as part of the response. Primarily the models that do have these services are centre-based, co-located responses. In this context, advocates are defined as holistic and independent workers with a role to listen to and act for children and families affected by abuse.

Five of the 12 distinct responses identified have some kind of professional onsite support service available (Joint Investigation Response Teams, Multi-Disciplinary Centres pilots, Perth-Metro, Multiagency Investigation & Support Team, and the ACT). In South Australia, the interviewer from the health-based Child Protection Service (who interviews especially vulnerable children) also serves as a support person as part of providing interviewing and assessment independent from police and child protection agencies. Support during and directly following the interview is provided by health clinicians in both NSW responses (Co-located Joint Investigation Response Teams and Non-co-located Joint Investigation Response Teams) and by a worker from the Department of Child Protection and Family Support based in the interview unit in the Perth-Metro response. The rest of the responses involve support services from the non-government sector built into the response.

Only three responses have onsite advocacy services as part of the response (Perth-Metro, Multiagency Investigation & Support Team pilot, and Multi-Disciplinary Centres pilot). The ACT response includes a mobile service where support workers from the Canberra Rape Crisis Centre attend when children report abuse to the police. Centres Against Sexual Assault are closely identified with the Sexual Offences and Child Abuse Investigation Teams in Victoria, and often provide a similar advocacy support role even where they are not co-located with police (Sexual Offences and Child Abuse Investigation Teams are required to notify Centres Against Sexual Assault within two hours for a response). For other jurisdictions, many of them have close ties with equivalent services in their jurisdiction, but support and advocacy providers are not embedded into the statutory response to the same degree as in the ACT and the Victorian Standard Response.

7. Integration of therapeutic services

The jurisdictions differed in the degree to which support services were embedded in the response to ensure that children and families receive needed services. Few responses have capacity for referral to onsite support and therapeutic services for children and families; both the Multiagency Investigation & Support Team pilot (WA) and the Multi-Disciplinary Centres pilots (Victoria) provide a specialist support and therapeutic service co-located with the tertiary child protection response. Children seen by the Child Protection Service (SA) at the Adelaide Women and Children's Hospital could also receive a referral to onsite services.

Some jurisdictions provide a facilitated referral to offsite services. In New South Wales this involves a NSW Health worker coordinating therapeutic and other health services. The ACT Wraparound response involves service providers in their case discussions, meaning that support services were closely connected with planning the response. Informally this is also the case in Tasmania. It is also worth noting that cases with ongoing involvement from child protection statutory authorities may include referral to services, potentially as part of an order or plan.

Note: ^aThis table distinguishes between an in-house service, a facilitated referral and a suggested referral. In an in-house service, the support or therapeutic service is provided on the same site. In a facilitated referral, the referrer will contact the service and assist with making the arrangements for the service. In a suggested referral, the referral is simply suggesting an appropriate service for the child or young person and their family.

All jurisdictions have close ties with authorities that conduct forensic medical examinations, although no sites in Australia have this service co-located with the statutory response (i.e., policing and child protection agencies). Although not a co-located response, children seen at the Child Protection Service (SA Health) in South Australia can receive forensic medical examinations at the same site where they are interviewed; and where cases are seen at Adelaide Women and Children's Hospital they can receive supportive and therapeutic services. In New South Wales, health clinicians arrange for forensic medical examinations at the closest facility. Similarly, in Queensland, Western Australia (Multiagency Investigation & Support Team pilot and Perth-Metro), South Australia, Victoria and the ACT, health agencies are closely involved in the response, participating in case planning and information exchange, allowing for easy access to forensic medical examinations. Regional/Remote WA, Tasmania and the Northern Territory agencies make referrals as needed through informal connections to their state health services. In some states with resources centralised in the state capital (e.g., Western Australia), children can be required to travel long distances to attend a specialist forensic medical examination facility.

Table 9 outlines the child witness protections available in the eight Australian legal jurisdictions (while noting that different arrangements are currently in place in NSW due to the Child Sexual Offence Evidence pilot). The jurisdictions are similar in terms of the state child witness provisions with a range of special conditions. All jurisdictions allow for a recorded interview as the child's evidence-in-chief and almost all allow for the pre-recording of a cross-examination of children, although the conditions for this differ depending on the age of the child. Pre-recording of the cross-examination was only recently introduced in New South Wales as part of the Child Sexual Evidence Pilot in Newcastle and the Downing Centre in Sydney; in other areas the cross-examination is still by closed-circuit television (CCTV). All jurisdictions provide witness support services, with slight differences in terms of their roles (i.e., whether they provide support or merely advice). New South Wales and South Australia each have a scheme for intermediaries: independent professionals who assess and advise on the capacity of the child to communicate with the court and other professionals.

8. Governance

Jurisdictions differ in the degree to which their response is prescribed and documented in a cross-agency protocol. The Joint Investigation Response Teams have comprehensive documentation detailing the response and the roles and responsibilities for agencies (see Table 10). Queensland, the NT, Victoria, the ACT, South Australia and the Multiagency Investigation & Support Team pilot response have detailed cross-agency protocols. For the other responses in Western Australia, agencies each have their own documented processes to follow. Tasmania does not have specific policy or protocols for their cross-agency response, but the response operates through informal information-sharing between agencies under a broad memorandum of understanding.

All jurisdictions have processes in place for review and discussion of conflicts and difficulties across agencies. Some jurisdictions hold these as regular review sessions (all NSW, Queensland, Multiagency Investigation & Support Team pilot [WA], ACT, NT), while other jurisdictions identified processes for addressing problems either at the case or policy level (Victoria, SA, Tasmania). Five responses have cross-agency data systems for monitoring the delivery of the case response and for monitoring outcomes from the response. New South Wales, Queensland, Western Australia and the ACT each have shared cross-agency data systems. More informally, the Northern Territory records case outcomes on a shared spreadsheet.

Notes: ^aThe Joint Investigative Response Team's Memorandum of Understanding is under review and may be revised as a part of the current review of the Joint Investigative Response Team by the NSW Ombudsman's Office. ^bPolicy and practices in Queensland are currently under review and may be subject to change.

Conclusion

This paper outlines some of the differences between cross-agency responses to severe child abuse in Australian jurisdictions, separated out into 12 distinct responses within the eight Australian legal jurisdictions. These responses differ in the degree of co-location and integration of agencies, the connection between the investigation and support/treatment response, and the degree to which the response is de-centralised.

Areas for research and policy development

It is hoped that this paper will prompt discussion within and between jurisdictions about the arrangements in place to support cross-agency and inter-agency responses to severe abuse, and the need for ongoing research and evaluation into these systems. We note that there has been limited comparative research and evaluation in Australia, and that system responses change quickly, limiting the opportunity for evidence-based policy development.

Overall, there is a need to evaluate the outcomes of existing system responses in Australian jurisdictions. This would ideally include undertaking work to clarify the operation and assumptions of responses across agencies, and to identify key outcomes to be monitored. Monitoring agreed cross-agency outcomes presents an opportunity to obtain a baseline for when new initiatives are implemented (e.g., witness intermediaries in NSW), as well as measuring performance over time. The measurement of key outcomes across agencies may also serve to break down the pre-occupation with individual agency performance and highlight overall system functioning and the impact of this system on agencies, children and families. There may also be some value in exploring a set of quality and outcome measures that could be applied in any jurisdiction, including those without arrangements for cross-agency responses.

The effect of advocacy and built-in therapeutic services is still not well-researched internationally, particularly in terms of their impact on the uptake of services and how they may influence the criminal justice or child protection process. A related issue is that models that rely upon referral to externally provided therapeutic services rely upon those services providing a high-quality evidence-based response for the referred aspects of their model to be effective.

Ultimately, the quality of a cross-agency response depends on how the workers operate within the frameworks they have. A comprehensive policy framework needs to build in measures that enhance the capacity of these workers, and recognise the challenging environment they operate in.

Areas for consideration in enhancing responses

The introduction highlighted a number of characteristics of multi-disciplinary teams that, based on the existing literature, seem to be associated with a more effective response across domains (i.e., criminal justice, child protection and therapeutic). This comparison suggests that Australian jurisdictions have many of these characteristics.

While research suggests that a planned cross-agency response is effective compared to an unplanned response (Herbert & Bromfield, 2016, 2017a), there is limited research comparing different types of cross-agency responses in improving children's safety, prosecuting offenders and addressing the effects of abuse experienced by children and their non-abusive family members (Herbert & Bromfield, 2017a). This means there is a limited basis on which to recommend characteristics or components as a strategy to improve responses. While recommending further research and policy development in this area, the authors would like to highlight four key areas for consideration to enhance responses. These issues have broader application to informing the planning and provision of other interventions in which multiple disciplines or agencies are involved in the response.

1. Integration of supportive and therapeutic services

For many of the responses, a key aspect missing was specific resourcing in place to address the distress, uncertainty and trauma experienced by children and their families. We note that in every jurisdiction there are specialist services and advocacy organisations; however, the degree to which these independent services are built into the response varied. Further research is needed to unpack the role and better understand the impact of built-in advocacy and support services. The integration and resourcing of services to ameliorate the effects of abuse appear to increase the likelihood of children and families receiving timely referrals to therapeutic and support services to reduce the impacts of abuse.

2. Co-location of workers

While co-location is by no means a panacea to improving cross-agency collaboration, sharing working space (or otherwise working nearby) and meeting regularly appear to be important to working effectively across agencies. Building individual rapport between workers, understanding and respect for each other's work, and quicker and easier discussion and information exchange all may play a part in improving responses and seem likely to be enabled by co-location.

3. Systematised mechanisms for case planning and review

Having requirements for the frequency, timing and nature of case planning between agencies seems to be an important part of ensuring that these points of contacts between agencies are used effectively. Having these interactions run more informally may cause problems between agencies in getting all the information needed and involving workers from other agencies that may have other imperatives.

4. Governance structures

A clear cross-agency protocol, a structure to help manage the response across agencies and ownership from constituent agencies are important parts of a cross-agency response. Without a clear understanding of the process and the roles of individuals, there may be cross-agency conflict. Without a means of managing or responding to conflict across agencies, a breakdown in teams and process can evolve. Without consistent buy-in from constituent agencies, adherence to agreed processes may be subject to the personalities of individual workers.

References

• Barton, A., & Welbourne, P. (2005). Context and its significance identifying "what works" in child protection. Child Abuse Review, 14(3), 177-194. doi: 10.1002/car.892
• Bertram, R. M. (2008). Establishing a basis for multi-system collaboration: Systemic team development. Journal of Sociology and Social Welfare, 35(4), 9-27.
• Bromfield, L., Salveron, M., & Main, J. (2017). Strengthening what works for children: Aligning target group, theory of change and program components to outcomes. Melbourne: Child Family Community Australia, Australian Institute of Family Studies.
• Burns, B. J., Phillips, S. D., Wagner, G. A., Barth, R. P., Kolko, D., Campbell, Y., & Landsverk, J. (2004). Mental health need and access to mental health services: Youths involved with child welfare. A national survey. Journal of the American Academy of Child & Adolescent Psychiatry, 43(8), 960-970.
• Child Protection Systems Royal Commission. (2016). The life they deserve: Child Protection Systems Royal Commission report. Adelaide: Government of South Australia.
• Community Development and Justice Standing Committee. (2008). Inquiry into the prosecution of assaults and sexual offences. Perth: Legislative Assembly, Parliament of Western Australia.
• Darlington, Y., & Feeney, J. A. (2008). Collaboration between mental health and child protection services: Professionals' perceptions of best practice. Children and Youth Services Review, 30(2), 187-198. doi: 10.1016/j.childyouth.2007.09.005
• Department of Justice Canada. (2013). Children's Advocacy Centres Next Steps Meeting. Ottawa: Department of Justice Canada. Retrieved from <cac-cae.ca/wp-content/uploads/fr_-DOJ_NSM-Research-Update_April-11+12_-S.McDonald-Eng-Version.pdf>.
• Edinburgh, L., Saewyc, E., & Levitt, C. (2008). Caring for young adolescent sexual abuse victims in a hospital-based children's advocacy center. Child Abuse & Neglect, 32(12), 1119-1126. doi: 10.1016/j.chiabu.2008.05.006
• Ells, M. (2000). Forming a multidisciplinary team to investigate child abuse. Rockville: Office of Juvenile Justice and Delinquency Prevention.
• Elmquist, J., Shorey, R. C., Febres, J., & Zapor, H. (2015). A review of Children's Advocacy Centers' (CACs) response to cases of child maltreatment in the United States. Aggression and Violent Behavior. doi: 10.1016/j.avb.2015.07.002
• Fargason, C. A., Jr., Barnes, D., Schneider, D., & Galloway, B. W. (1994). Enhancing multi-agency collaboration in the management of child sexual abuse. Child Abuse & Neglect, 18(10), 859. doi: https://doi.org/10.1016/0145-2134(94)90065-5
• Gragg, F., Cronin, R., & Schultz, D. (2006). Processing and outcomes of child abuse and neglect cases in three sites. National Evaluation of the Safe Kids/Safe Streets Program. Washington D. C.: Office of Juvenile Justice and Deliquency Prevention.
• Green, B. L., Rockhill, A., & Burrus, S. (2008). The Role of Interagency Collaboration for Substance-Abusing Families Involved with Child Welfare. Child Welfare, 87(1), 29-61.
• Guobrandsson, B. (2014). Towards a child-friendly justice and support for child victims of sexual abuse. Protecting children from sexual violence: A comprehensive approach. Strasbourg Cedex: Council of Europe. Retrieved from <www.coe.int/t/dg3/children/1in5/WhatWeKnow/Publication_en.asp>.
• Haas, S. M., Bauer-Leffler, S., & Turley, E. (2011). Evaluation of cross-disciplinary training on the co-occurrence of domestic violence and child victimization: Overcoming barriers to collaboration. Journal of Health and Human Services Administration, 34(3), 352-386.
• Herbert, J. L., & Bromfield, L. (2016). Evidence for the efficacy of the Child Advocacy Center model: A systematic review. Trauma Violence & Abuse, 17(3), 341-357. doi: 10.1177/1524838015585319
• Herbert, J. L., & Bromfield, L. (2017a). Better Together? A Review of Evidence for Multi-Disciplinary Teams Responding to Physical and Sexual Child Abuse. Trauma, Violence, & Abuse. doi: 10.1177/1524838017697268
• Herbert, J. L., & Bromfield, L. (2017b). Components of effective cross-agency responses to abuse. A report for the NSW Ombudsman's Office. doi: 10.4226/78/599e195d13749
• Herbert, J. L., & Bromfield, L. (2017c). Multiagency Investigation & Support Team (MIST) pilot: Evaluation report. Adelaide:Australian Centre for Child Protection Division of Education, Arts and Social Sciences, University of South Australia.doi: 10.4226/78/5976de366dceb
• Herbert, J. L., & Bromfield, L. (2017d). National cross-agency comparison of practice in responding to criminal child abuse. A report for the NSW Ombudsman's office. doi: 10.4226/78/599e1844bdba2
• Herbert, J. L., Walsh, W., & Bromfield, L. (2017). A National Survey of Characteristics of Child Advocacy Centers in the United States. Child Abuse & Neglect. doi: 10.1016/j.chiabu.2017.09.030
• Howell, J. C., Kelly, M. R., Palmer, J., & Mangum, R. L. (2004). Integrating child welfare, juvenile justice, and other agencies in a continuum of services. Child Welfare, 83(2), 143-156.
• Jackson, S. L. (2012). Results from the Virginia Multidisciplinary Team Knowledge and Functioning Survey: The importance of differentiating by groups affiliated with a child advocacy center. Children and Youth Services Review, 34(7), 1243-1250. doi: 10.1016/j.childyouth.2012.02.015
• Jones, L. M., Cross, T. P., Walsh, W. A., & Simone, M. (2005). Criminal investigations of child abuse: the research behind "best practices". Trauma Violence & Abuse, 6(3), 254-268.
• Kemp, S. P., Marcenko, M. O., Hoagwood, K., & Vesneski, W. (2009). Engaging parents in child welfare services: Bridging family needs and child welfare mandates. Child Welfare, 88(1), 101-126.
• Lalayants, M. (2013). Multidisciplinary collaboration in child protective clinical consultations: perceptions of best practices. Journal of Public Child Welfare, 7(3), 253.
• National Children's Alliance. (2016). Annual Report 2015. Washington, D. C.:National Children's Alliance. Retrieved from <www.nationalchildrensalliance.org/sites/default/files/downloads/2015-NCA-Annual-Report.pdf>.
• National Children's Alliance. (2017). Standards for Accredited Members. Washington D.C.: National Children's Alliance.
• Newman, B. S., & Dannenfelser, P. L. (2005). Children's protective services and law enforcement: Fostering partnerships in investigations of child abuse. Journal of Child Sexual Abuse, 14(2), 97-111.
• Newman, B. S., Dannenfelser, P. L., & Pendleton, D. (2005). Child Aabuse investigations: Reasons for using Child Advocacy Centers and suggestions for improvement. Child & Adolescent Social Work Journal, 22(2), 165-181. doi: 10.1007/s10560-005-3416-9
• NSW Ombudsman. (2017). NSW Ombudsman inquiry into the operation of the JIRT Program. Sydney: NSW Ombudsman.
• Owens, P. L., Hoagwood, K., Horwitz, S. M., Leaf, P. J., Poduska, J. M., Kellam, S. G., & Ialongo, N. S. (2002). Barriers to children's mental health services. Journal of the American Academy of Child & Adolescent Psychiatry, 41(6), 731-738.
• Powell, M. B., & Wright, R. (2012). Professional's perceptions of a new model of sexual assault investigation adopted by Victorian Police. Current Issues in Criminal Justice, 23, (333-352).
• Stanley, N., Miller, P., Foster, H. R., & Thomson, G. (2011). Children's experiences of domestic violence: developing an integrated response from police and child protection services. Journal of Interpersonal Violence, 26(12), 2372-2391. doi: 10.1177/0886260510383030
• Szilassy, E., Carpenter, J., Patsios, D., & Hackett, S. (2013). Outcomes of short course interprofessional training in domestic violence and child protection. Violence Against Women, 19(11), 1370.
• Tye, C., & Precey, G. (1999). Building bridges: The interface between adult mental health and child protection. Child Abuse Review, 8(3), 164-171.