Literature review

There is a large and well documented body of research on the need for appropriate and timely sharing of information between agencies and family violence providers to support effective family violence risk assessment and response (Attorney-General’s Department (Cth), Australian Institute of Judicial Administration Incorporated & University of Queensland 2019; Breckenridge et al. 2015; Domestic Violence Prevention Council (DVPC) 2016; Doyle 2015; Glanfield 2016; Justice and Community Safety Directorate (ACT) 2016; Steel, Blakeborough & Nicholas 2011).

Numerous Australian inquiries into domestic and family violence, often resulting from high profile homicides, have recommended that specific legislation be introduced or amended to improve information sharing arrangements between relevant entities (Adams & Lee-Jones 2016; Commonwealth of Australia 2016; Coroners Court of Victoria 2015; Domestic Violence Prevention Council (DVPC) 2016; Glanfield 2016; NSW Legislative Council 2012; Parliament of Western Australia 2012; Special Taskforce on Domestic and Family Violence in Queensland 2015; State of Victoria 2016a).

These recommendations have resulted in many Australian jurisdictions adopting family violence information sharing legislation including, most recently, Victoria’s introduction of Part 5A of the Family Violence Protection Act 2008 (Vic) (Jones 2016). Legislation currently operating in Australia (as of February 2020 when the final literature review research was undertaken) and relevant international jurisdictions is summarised in Appendix Three and includes specific family violence information sharing schemes (family violence ISSs) or provisions about family violence in similar child safety information sharing schemes (child safety ISSs).

This scoping review draws together findings from work on information sharing in the context of family violence from the past 10 years to identify key barriers and enablers to the effective implementation of family violence ISSs. It begins by explaining the search strategy employed to identify relevant literature followed by a brief overview of current Australian and international ISSs and subsequent evaluations of those schemes’ implementation. It goes on to summarise the barriers and enablers to information sharing identified in recent research literature, which are reported under three themes: legal, technological and organisational. Other factors that may be particularly relevant for the implementation of Part 5A of the Family Violence Protection Act 2008 (Vic), such as reporting on information sharing and defining prescribed entities in legislation, are outlined.

Literature review methodology

A search was undertaken using a range of databases including Scopus, Informit, Criminal Justice Abstracts and Google Scholar. The search strategy involved multiple keyword searches using the terms “information shar*’, “shar* information”, “family violence”, “domestic violence”, “domestic abuse” and “intimate partner violence”. The search was limited by language (English) and date (2011–2020). Studies were included if they addressed family violence information sharing schemes or reported on family violence -specific provisions in child safety information sharing schemes. Papers were excluded if they centred on distinct information sharing schemes, such as Domestic Violence Disclosure Schemes or the National Protection Order Scheme, as these schemes are targeted at disclosing specific information in limited contexts rather than multi-agency collaboration more broadly. The search was widened by a snowball approach based on reviewing citations within key articles (Keeley et al., 2015) to identify further articles of relevance that may not be listed in databases. Studies known to the research team, but which did not emerge from the initial searches, were also included. Individual searches of government websites were also conducted to identify specific legislative approaches and evaluations of similar information sharing schemes in Australia and internationally.

Current legislation and evaluations of information sharing schemes

Although Victoria is not the first Australian jurisdiction to introduce a specific family violence ISS (see Appendix Three), recent empirical literature and comprehensive evaluations specifically reviewing the barriers and enablers of ISSs in general are limited (Keeley et al. 2015; State of Victoria 2016a). As Keeley et al. (2015, p. 16) stated: ‘research has explored a range of barriers and enablers to collaboration, but less attention has been paid to inter-organisational information exchange as a specific issue’.

Australian legislative landscape

As of January 2020, the following states and territories currently have legislative provisions enabling family violence risk information sharing: Victoria (2017), QLD (2016), NSW (2014), Tasmania (since approximately 2004), ACT (since 1992, and 2005), NT (since 2019) and WA (Restraining Orders Act 1997). South Australia relies on non-legislative protocols developed by the SA Ombudsman. Internationally, the United Kingdom, and British Columbia rely on specific exceptions in their privacy acts to allow disclosure where family violence is present. In New Zealand and the United States, information sharing is enabled by specific provisions in their respective family violence/violence against women legislation.

Legislative provisions enabling information sharing to protect children’s safety and wellbeing where there are concerns about family violence currently exist in: ACT (2016), WA (2011, amended to include family violence provisions in 2016), NSW (2009), Tasmania (approx. 2004 and 2009), QLD (since 2004) and Victoria (2019).

Evaluations of Australian information sharing schemes

In the past five years, relevant family violence and child safety ISSs in NSW, Tasmania and Western Australia have been subject of reviews or evaluations (see Appendix Three). The South Australian Ombudsman also reviewed their non-legislative information sharing Guidelines in 2012 (Ombudsman SA 2013).

However, as noted above, government-funded evaluations and recent academic literature in this space largely focus on the broader mechanisms of multiagency coordination and collaboration, rather than information sharing specifically (Keeley et al. 2015). For example, Tasmania’s 2014 review of the Safe at Home program only briefly summarised the enabling effect of section 37 of the Family Violence Act 2004 on information sharing while BOCSAR’s 2017 evaluation of NSW’s Safer Pathway Program noted information sharing should be the subject of future reviews (Department of Justice (Tas) 2015; Trimboli 2017). Academic literature also tends to focus on information sharing arrangements for specific, limited types of information, such as the National Protection Order Scheme or Domestic Violence Disclosure Schemes (Taylor et al. 2017).

Evaluations of Australian child safety ISSs are more common and include similar themes to family violence schemes (Cassells et al. 2014; Keeley et al. 2015; Parliament of Western Australia 2012). The most recent comprehensive reviews of inter-organisation information sharing in Australia include Cassells et al. (2014) and Keeley et al.’s (2015) evaluations of the NSW child information sharing provisions contained in Chapter 16A of the Children and Young Persons (Care and Protection) Act 1998 (Chapter 16A). Those evaluations have informed subsequent Australian State and Territory government reviews, including the Victorian Royal Commission into Family Violence (State of Victoria 2016a) and consequently provided a foundation for identifying many of the key barriers and enablers to information sharing covered in this literature review.

Barriers and enablers to information sharing schemes

Although there are limited empirical studies into the effectiveness of information sharing schemes specifically, the existing evidence base identifies three categories of barriers and enablers for effective information sharing: political and legal, technological/operational and organisational (Gil-Garcia & Sayogo 2016; Keeley et al. 2015; Yang & Maxwell 2011). These were recognised in the RCFV’s justification for introducing a legislative information sharing scheme in Victoria, which focused on:

The fact that legislation and policy governing information sharing are complex, confusing and restrictive [;] the lack of an information-sharing culture and leadership[;] [and] reliance on outdated IT systems, which impedes information sharing (State of Victoria 2016a, p. 170; 2016c).

These findings have been replicated in other Australian government family violence inquiries, including the COAG Advisory Panel on Reducing Violence against Women and their Children, which heard similar evidence that agencies and service providers were inhibited from collaborating in their responses to domestic and family violence by cultural, financial, human resource, policy and legal barriers (Commonwealth of Australia 2016). Conversely, the 2016 Australian Capital Territory Glanfield Inquiry concluded:

It appears that clear legislative authority to share information, coupled with training and practical mechanisms requiring or supporting this, can result in better information sharing between agencies which facilitates better outcomes for vulnerable families (Glanfield 2016, p. 92).

However, key literature concludes that these three themes vary in significance, with organisational factors being more significant in inhibiting and enabling information sharing than technological factors such as IT systems or legislative frameworks (Adams & Lee-Jones 2016; Keeley et al. 2015). As such, it is frequently emphasised that information sharing is one aspect of multi-agency collaboration that assists responses to family violence (Healey, Humphreys & Wilcox 2013), and although addressing the barriers and enablers outlined below may improve the effectiveness of information sharing schemes, collaboration itself:

Is a developing process that is challenging and time-consuming, and that successful collaboration is based on a need to work together to improve services, so that the benefits outweigh the difficulties (Keeley et al. 2015, p. 18).

Ultimately, as the Victorian Behavioural Insights Unit concluded, ‘information sharing is not an end in itself’ (Behavioural Insights Unit (Vic) 2017, p. 5).

Balancing privacy and safety

A common undercurrent of existing research and literature on information sharing schemes in the context of family violence is the need to balance concerns about client privacy and confidentiality with the protection of clients from potential risks (Adams & Lee-Jones 2017; Justice and Community Safety Directorate (ACT) 2016; Keeley et al. 2015; Parliament of Western Australia 2012; Special Taskforce on Domestic and Family Violence in Queensland 2015). Addressing concerns about privacy and confidentiality are critical in the context of family violence as a lack of confidence in how information is shared can result in victim/survivors’ reluctance to report family violence or seek support including by ‘destroying relationships of trust between a service provider and a client, leading to disengagement of a client, and becoming a barrier to victim/survivors’ willingness to seek help’ (Special Taskforce on Domestic and Family Violence in Queensland 2015, p. 231; State of Victoria 2016a). Interprofessional differences on this key issue shape stakeholders’ views on whether information sharing schemes should require consent from clients, which can then underpin the key factors inhibiting effective information sharing, such as administrative delays because of the perceived need to seek consent (HM Inspectorate of Probation et al. 2017), different professional cultures, and technological concerns about securing data. The need for proportionality in decision-making about the sharing and protection of information is emphasised in existing research literature and given significant weight in legislative drafting (Home Office (UK) 2014; Keeley et al. 2015; State of Victoria 2016a; Victorian Government 2017). Even where consent is not required by legislation, seeking consent where possible is still recognised as best practice, as this facilitates trust between the client and information sharer (Keeley et al. 2015; NSW Legislative Council 2012; Special Taskforce on Domestic and Family Violence in Queensland 2015). As such, the tensions in balancing privacy and safety are a persistent theme throughout the literature in relation to effectively implementing information sharing arrangements.

Political and legal factors

Barriers

Legislation and policy can inhibit information sharing, either by directly prohibiting disclosure of personal information or through complex and confusing regulatory frameworks (Keeley et al. 2015; DVPC 2016). According to Adams and Lee-Jones (2017, p. 1351), ‘the legal framework for decision making can become a problem if it does not find, or does not clearly articulate, the appropriate balance between competing rights’. In their study into sharing information relating to child sexual abuse, they found that legislation and policy has tended to emphasise the protection of personal information about children and their families, leading to over-caution by workers and ‘sometimes tragic outcomes’ (Adams & Lee-Jones 2017, p. 1355).

A review of the Australian Capital Territory Family Violence Intervention Program in 2012 similarly concluded: ‘information sharing is hampered by lack of interagency protocols and a legislative base to ensure that information is adequately provided and protected’ (Cussen et al. 2012). Internationally, New Zealand’s Ministry of Justice found that although its Privacy Act allows the disclosure of personal information where there is a serious threat to life, such a high threshold can create perceptions that disclosure of personal information for family violence risk assessment purposes is very limited (Ministry of Justice (NZ) 2015). Similarly, in NSW the Legislative Council expressed concern about:

The evidence from many other participants that present information sharing provisions are significantly impairing services’ ability to work together, and ultimately, are impairing positive outcomes for clients, not least their safety (NSW Legislative Council 2012, p. 81).

Recognition of these legislative barriers underpinned the introduction of Part 5A of the Family Violence Protection Act 2008 (Vic), as illustrated in the Bill’s Second Reading Speech:

Current laws are complicated, confusing and restrictive for those who work with victims and perpetrators. … This bill squarely addresses this gap. … The regime provides a clear authority for organisations responding to family violence to share relevant information as needed for family violence risk assessment and management, cutting through current complexity (Victorian Government 2017, p. 2118).

This justification reflected the RCFV’s findings that a legislative scheme was preferable because it would provide a single, clear authority for organisations and workforces to confidently share information and would be cheaper than developing alternatives such as a code of practice (State of Victoria 2016a).

These findings from recent government inquiries and academic literature highlight that lack of legislative or regulatory authority can impede information sharing and, as Peterson and Schroeder (2017, footnote 621) emphasised in relation to Canada’s legislation: ‘the limitation brings home the need for governments to consult subject matter experts in connection with the detailed wording of statutes.’

Notably, confusion about the regulatory authority for sharing information is more likely to inhibit information sharing than legislation itself, and broader restrictive policy factors identified include the prioritisation of certain programs, institutional and professional politics, privatisation and competitiveness (Adams & Lee-Jones 2016; Keeley et al. 2015). In sum, while literature and key Australian government reports suggest introducing information sharing schemes reduces the hesitation of agencies to exchange information, key studies establish that ‘[m]ost barriers occurred in the interpretation of the legal and policy constraints rather than in the actual legal or policy provisions’ (Keeley et al. 2015, p. 3).

Enablers

Other work suggests that legislation and policy can enable information sharing where it provides a clear authority for appropriate disclosure, particularly where disclosures are mandated rather than permitted and the legislation explicitly defines when information can be shared (Adams & Lee-Jones 2016; Yang & Maxwell 2011). The enabling impact of legislation is reflected in the number of recommendations from Australian inquiries and reviews into family violence responses that legislative schemes need to be introduced, and in the subsequent implementation of those schemes by most State and Territory governments (see Appendix Three). For example, in recommending specific legislation be introduced in relation to family violence, a 2012 Western Australian review of the Children and Community Services Act 2004 (WA) concluded:

Specific legislation which enables relevant information to be shared in good faith, and which provides corresponding protections from liability for doing so, would provide the sector with greater certainty and confidence in responding to family and domestic violence (Parliament of Western Australia 2012, p. 12).

Similarly, Keeley et al.’s (2015, p. 8) evaluation of the NSW child information sharing scheme noted ‘the introduction of specific legislative authority has clearly been helpful in the ongoing development of a culture of appropriate information sharing’ and s 37 of the Family Violence Act 2004 (Tas) has been identified as a key enabler for interagency information sharing in Tasmania (Cassells et al. 2014; Department of Justice (Tas) 2015). These findings are reflective of earlier work by Yang and Maxwell which showed that ‘legal and policy regulations can facilitate relationship building, risk reduction, and trust development in inter-organizational information sharing projects when specific guidance such as how to utilize information is proposed’ (2011, p. 170). Legislative and policy frameworks can also enhance the public’s trust in the government’s handling of information by creating standards for protecting and storing data (Gil-Garcia & Sayogo 2016; Yang & Maxwell 2011), which is critical in the family violence context, noting ongoing concerns about confidentiality and privacy.

However, existing literature notes that the enabling effect of legislation is limited (Adams & Lee-Jones 2016; Taylor et al. 2015). For example, Cassells et al. (2014) found that the messaging around the implementation of Chapter 16A in NSW was more effective in encouraging information sharing to protect children than the actual legislation. They found that:

Although the provisions of Chapter 16A of the Children and Young Persons (Care and Protection) Act 1998 that enable information sharing are highly valued, there were issues reported in implementation and the continuing challenges around information sharing (Cassells et al. 2014, p. 55).

In relation to sharing information about domestic violence protection orders, Taylor et al. (2015) concluded that protocols and memoranda of understanding are not enough; the workforce needs to have sufficient knowledge about sharing information. Keeley and colleagues found this knowledge lacking in relation to proactive child safety information sharing in NSW:

Notwithstanding that such proactivity is permitted by Chapter 16A, the research team was unable to identify any statement of policy within the policy documents examined that actively encourages workers to proactively share wherever appropriate (2015, p. 31).

Some international literature goes further, concluding that ‘rational bureaucratic lines of thinking’ that focus on reducing ‘human variability’ are not adequate to address barriers to information sharing, because information sharing is a complex exercise that cannot be ‘perfected’ (Thompson 2013, pp. 190, 7). For example, Gil-Garcia and Sayogo’s national survey of criminal justice and public health government managers in the United States found that:

Political and policy factors in the form of regulations or formal agreements about the initiative, existing legislation that made the initiative possible, and legislators supporting the initiative are not found to be statistically significant for the success of inter-organizational information sharing initiatives (2016, p. 579).

In sum, legal and policy factors enabling information sharing are recognised as important, but limited in relevant literature, and these conclusions have resulted in recognition by some international governments such as the UK Home Office, that adequate guidance is important to supplement legislation (Home Office (UK) 2014).

Technological (or operational) factors

Barriers

Technology and IT systems have been cited as a hindrance to effective information sharing in both Australian and UK evaluations of multi-agency collaborations (Home Office (UK) 2013; Keeley et al. 2015; State of Victoria 2016a). In particular, the RCFV found that: ‘almost all submissions and witnesses who gave evidence about IT in the context of family violence acknowledged that the current arrangements present major barriers to information sharing’ (State of Victoria 2016a, p. 176). In New Zealand, a recent consultation regarding Approved Information Sharing Agreements between agencies found:

The feedback from agencies was overwhelmingly that the barriers to information sharing were operational. These included issues such as a lack of interoperability between IT systems, security concerns, cost, and differing priorities between agencies (Privacy Commissioner (NZ) 2017, p. 4).

The types of technological issues identified by literature include incompatible databases or multiple IT systems, difficulties storing and accessing databases, inability for automation, difficulties tracking individuals whose information has been shared and those who are sharing or accessing that information, and difficulties identifying which agencies may hold information (Behavioural Insights Unit (Vic) 2017; Home Office (UK) 2013; Keeley et al. 2015; State of Victoria 2016a). These issues extend to processes of collection (HM Inspectorate of Probation et al. 2017); integration of existing systems (Yang & Maxwell 2011); security and storage (Home Office (UK) 2013; Stanley & Humphreys 2014); and recordkeeping (HM Inspectorate of Probation et al. 2017).

Previous research and literature underscore the importance of adequate security for recording and storing information in the context of family violence because unsecured information can put victim/survivors at risk and undermine the confidence of victim/survivors in reporting family violence and sharing their personal information (Adams & Lee-Jones 2016; Stanley & Humphreys 2014; Taylor et al. 2015). For example, in the Australian Capital Territory:

The Death Review found that within some agencies, poor record keeping is also contributing to the inability to share information accurately and in a timely manner. The ACT Government acknowledges the importance of accurate and timely record keeping and record management. The Government also acknowledges that this is particularly important in relation to family violence as it affects an agency’s ability to share accurate information and manage risks facing victims of family violence (ACT Government 2016, p. 10).

Further, Drinkwater and colleagues (2017) study on documentation of family violence in electronic patient records in the UK found that clinician’s concerns about security of records resulted in clinicians using ad hoc workarounds, such as emailing colleagues to communicate sensitive information, rather than using the electronic patient record. Although this study related only to electronic hospital records, it highlighted the ways in which technology can create concerns about privacy for victim/survivors of family violence more generally, particularly where there is a risk of perpetrators accessing that information (Drinkwater et al. 2017).

Enablers

To ensure information systems and technology enable effective information sharing, recent evaluations highlight the need for technology to be relevant and up to date and reflect the needs of the users accessing the information, including, for example, providing space for contextual information (ACT Government 2016; Behavioural Insights Unit (Vic) 2017; Keeley et al. 2015). Staff also need to be trained to use IT and data systems, including records management (Keeley et al. 2015).

Shared databases or other ways to systemise tools for collecting information are frequently cited as key enablers for information sharing in different contexts (Doyle 2015; Home Office (UK) 2014; Taylor et al. 2015). This reflects findings that consistent recording and communication of information assists ‘all referral agencies and the multi-agency team to convey clear and sufficient information about cases’ (Home Office (UK) 2014, p. 11) and minimise duplication of services (Taylor et al. 2015)

Literature consistently notes however that technological factors are not usually defining enablers (or barriers) for information sharing in practice. In evaluating Chapter 16A, Keeley et al. ultimately concluded ‘The study found that in no case did technology create a fundamental barrier to information sharing (or conversely provide a solution to problems around information sharing)’ (Keeley et al. 2015, pp. 7-8). Similarly, in reviewing multi-agency collaborations in the UK, the Home Office concluded that cultural and organisational barriers were more important, and therefore ‘simply using a shared tool would not overcome these barriers’ (Home Office (UK) 2014, p. 11).

Organisational factors

Organisational factors are frequently cited as the most significant barriers and enablers for effective information sharing. These factors are not consistently categorised in the literature. However the RCFV summarised the following key organisational barriers and enablers based on Keeley et al.’s identification of these factors in the following extracted table (State of Victoria 2016a, p. 175).

Noting this table, and Keeley and colleagues’ (2015) concise summary of these barriers and enablers in their evaluation, the following section of this literature review limits itself to a brief overview of the most substantive organisational factors discussed in recent literature. Some of these overlap with, or can be addressed by, related legal and policy, and technological factors identified above.

Barriers

Much of the research on information sharing in the family violence context shows that ambiguity about what, how and when information is shared and how risk is appraised inhibits the effectiveness of such schemes (Cleaver et al. 2019; Cornford 2019). Cornford (2019) attributes this confusion to competing institutional logics that frame information sharing as either a (socio-technical) design problem, an (information) governance problem or as a (organisational) culture change problem. Each of these institutional lenses hold often conflicting positions about legitimate decision-makers in organisations as well as how they frame and define the problem and solution of information sharing (Cornford 2019). The ways in which these different institutional logics create tensions and ambiguity with regard to effective information sharing are discussed in detail below.

Different professional cultures and values

In relation to child safety information sharing in NSW, Keeley et al. concluded that ‘the two main reasons for the lack of information sharing were: risk-averse organisations [and] organisational or professional cultures which did not value holistic interventions’ (Keeley et al. 2015, p. 7). Similarly, Adams and Lee-Jones found that although effective information sharing relies on the confidence of staff in exercising judgement (and therefore mentoring, training and other confidence-building exercises enable information sharing), ‘these endeavours may be inhibited, to significant but varying degrees, by deep, and possibly unresolvable, differences in the aims and values of some agencies’ (2016, p. 18). Taylor et al. (2017) conclude this is because the functioning of integrated systems relies on trust and shared standards and values, while, according to Adams and Lee-Jones (2016), earlier literature has found that legislation does not necessarily result in increased communication; rather agencies also need to agree on the objectives and policy basis for sharing information.

In particular, as noted earlier in this literature review, ‘sharing of information can be problematic for professionals whose ethical conduct is driven by their professional bodies’ high expectations of maintaining client confidentiality’(Keeley et al. 2015; Taylor et al. 2015, p. 14). This may mean, for example, social care professionals are more likely to share information than health professionals due to different focuses on the family/community rather than the individual and consequently health professionals may prefer to share information with other health professionals (Adams & Lee-Jones 2016; Home Office (UK) 2013; Keeley et al. 2015). There may also be cultural barriers between public and private sector bodies (Adams & Lee-Jones 2016). As such, recent key literature indicates that legislation or policy on its own will not achieve effective information sharing where cultural barriers exist (Adams & Lee-Jones 2016; Price-Robertson 2012).

Mistrust

Related to different professional cultures and values is the impact of mistrust between agencies on their willingness to share information. Keeley et al. summarised earlier findings on mistrust between bodies that handle personal information, that ‘factors including professional cultures which question the professionalism of others, competition between agencies, a history of problematic collaboration, and personal or professional animosity between individual managers in different organisations’ undermine information sharing (2015, p. 18; see also Price-Robertson 2012). Similar findings have been made in the UK while the Victorian Behavioural Insights Unit recently identified a number of organisational factors such as workers’ approaches to empowering victim/survivors, use of different risk-assessments, and trust and understanding of different entities’ roles as significant factors to consider in implementing information sharing arrangements in the context of family violence (Adams & Lee-Jones 2016; Behavioural Insights Unit (Vic) 2017).

Unclear policies or misunderstandings of authority within organisations

In relation to child safety information, both Keeley et al. (2015) and Adams and Lee-Jones (2016) found that workers tend to be risk-averse where the policies or guidance for sharing information are unclear. Although this confusion overlaps with some of the legal and policy factors discussed earlier, it is also an important organisational factor, as Keeley et al. found in relation to child safety information sharing in NSW:

The level of awareness amongst the workforce of Chapter 16A was high. However, … there are gaps between perceptions and actual legislative and policy constraints, in particular regarding agencies proactively sharing information. … Many workers reported not knowing who to ask for advice about when information should be shared and the process for exchanging information (2015, p. 4).

Similarly, following a Cabinet Directive in New Zealand in 2016 inviting agencies to identify information sharing barriers related to Approved Information Sharing Agreements, the New Zealand Privacy Commissioner noted: ‘It has become evident that many of the perceived barriers to information sharing are based in misunderstanding or uncertainty of the law’, highlighting the need for ‘clear legal guidance’ (Privacy Commissioner (NZ) 2017, p. 5). This confusion can arise from lack of clarity about decision-making responsibilities and may also be a result of lack of knowledge about the roles of other organisations, which directly inhibits the ability of appropriate organisations to identify each other (Keeley et al. 2015).

Different processes between organisations

As the Victorian Behavioural Insights Units notes, currently in Victoria 'each organisation providing services to victim/survivors and perpetrators uses a different knowledge management system' (Behavioural Insights Unit (Vic) 2017, p. 73). Stanley and Humphreys (2014) have likewise commented on the fact that risk assessment data collected by police compared to specialist women’s family violence organisations differed dramatically. This difference may reflect victim/survivors’ levels of comfort disclosing personal information to different entities. However, citing similar findings by Lips et al. (2011), Keeley et al. (2015) argue that recordkeeping may differ significantly between agencies and therefore different processes around information sharing can undermine the effectiveness of sharing information.

Even with specific legislation under Chapter 16A permitting information sharing in NSW, Keeley et al. (2015) noticed differences between agencies who interpreted the provisions strictly, and therefore implemented formal processes for requesting information, and agencies that were more collaborative in their approaches. Different processes and information needs can consequently make organisations more reluctant to share their information (Keeley et al. 2015).

Enablers

There is a growing body of work on barriers and enablers to effective interagency collaboration in the context of family violence (Herbet & Bromfield 2017; Joseph et al. 2019; Macvean, Humphreys & Healey 2018; O’Leary et al. 2018; Savic et al. 2017; Zijlstra et al. 2018). Much of the research literature indicates that cultivating positive interagency relationships is a necessary pre-cursor to effective interagency work (Savic et al. 2017). For example, workers in family and sexual violence sectors reported in an Australian workforce survey that having time to build interagency relationships and including interagency collaboration into service agreements would help them collaborate more effectively (Cortis et al. 2018).

A key enabler of integrated responses to family violence is the development of a shared understanding of the problem the collaboration aims to address (Laing, Heward-Belle & Toivonen 2018; Macvean, Humphreys & Healey 2018; O’Leary et al. 2018). Previous research and literature show that information sharing is hampered by diverse interprofessional discourses on family violence and the lack of a strong and shared problem definition (Laing, Heward-Belle & Toivonen 2018). Similarly, work by Savic and colleagues (2017) showed that developing a shared professional language improves referral and information sharing processes. These findings echo earlier work on interagency collaboration which highlights the difference between interagency and multi-agency approaches noting that:

The danger with multi-agency and multi-professional approaches is that practitioners work in parallel but in isolation. There is, therefore, a risk that a shared purpose and explicit partnership will not be clarified and vulnerable people may slip through the net (Sully 2008, p. 11).

Trust and management of mistrust

According to Keeley et al., research prior to 2012 ‘consistently identifies shared understandings and trust, or at least management of mistrust, as among the most important determinants of whether staff from different organisations are prepared to share information’ (Keeley et al. 2015, pp. 17, 87; see also Lips et al. 2011). This is because trust allows for more collaboration and tacit information sharing, and ‘eases the need for control’ by each agency (Lips et al. 2011, p. 256). For example, the Victorian Behavioural Insights Unit (2017, p. 39) observed workers in their study ‘were more willing to share information with other services if they knew the person requesting the information', which indicates the value of creating inter-organisational relationships (see also Home Office (UK) 2014). Consequently, building trust is recognised as a key enabler of information sharing but may only be able to be built over time (Keeley et al. 2015). Reflecting the underlying tensions between appropriate information sharing and confidentiality, technological factors such as data security can also enhance trust between organisations, and therefore facilitate information sharing (Keeley et al. 2015). Co-location (discussed further below) has also been identified by some limited UK reports as effective mechanisms for building trust (Home Office (UK) 2013, 2014).

Creating cultures of information sharing

In order to create shared understandings between organisations, recent Australian work indicates that mixed training sessions and providing feedback are effective ways to encourage organisations’ relationships because they enable workers to recognise how their approaches and uses of information may vary (Behavioural Insights Unit (Vic) 2017; Glanfield 2016). Internationally, the UK Home Office found in relation to multi-agency collaborations that:

Some commented that bringing practitioners together improved standards, because of the scrutiny between professional responses that followed. In some cases, this was felt to have fostered greater confidence to share information (Home Office (UK) 2014, p. 9).

In reviewing the implementation of its non-legislative information sharing arrangements, the South Australian Ombudsman concluded:

The size or budget of the organisation appeared not to significantly determine successful implementation. Rather, successful implementation depended more on leadership and commitment to implement the ISG, culture and systems for quality improvement, and capacity to manage competing demands (2013, p. 4).

Some intra-organisational strategies have been identified in the literature to assist in building these cultures. In relation to protection orders, Taylor et al. (2015, p. 45) concluded that ‘an institutional culture of cooperation and engagement with associated services in enforcement is necessary for effective responses.’ Summarising previous literature, Keeley et al. (2015) argue that organisational structures linking agencies and collaborative cultures fostered by organisational leaders are also important. These factors enable information sharing because understanding how information is used by other organisations appears to encourage workers’ willingness to share information (Stanley & Humphreys 2014).

Other factors that have been identified to support collaborative cultures include:

• improving different organisations’ understandings of how and when information will be used by other organisations, and standardising the scope of consent that different organisations seek (Adams & Lee-Jones 2016; Behavioural Insights Unit (Vic) 2017)
• knowing each other’s schedules to improve the timeliness of sharing information (Behavioural Insights Unit (Vic) 2017); and
• implementing protocols and memoranda of understanding (HM Inspectorate of Probation et al. 2017; Taylor et al. 2015)

Workforce training

The need for training and support to understand legislative schemes is frequently identified as a key factor in building organisational cultures and workforce confidence to share information appropriately under relevant legislation (ACT Government 2016; Family Safety Victoria 2017a; Glanfield 2016; Keeley et al. 2015; Taylor et al. 2017). For example, in reviewing the implementation of its non-legislative information sharing guidelines the South Australian Ombudsman concluded:

Implementation progress reports and consultation with those applying the ISG indicate that organisations with a sound staff induction and training culture are able to develop organisational procedures and complete staff induction without significant investment or difficulty. A key positive influence on implementation is strong direction and support from leadership and the commitment and energy of (in most cases) one individual to drive the initiative. Staff induction is a challenge in some larger and more diverse organisations (2013, p. 3).

The case study by Keeley et al. similarly concluded that:

The legislative support provided through the provisions of Chapter 16A was seen as a significant enabler for information exchange, especially as this was accompanied by a high-profile rollout and significant investment in training (2015, p. 3).

But even so, practitioners needed refreshers on the legislation, particularly where they did not share information frequently or where there was high staff turnover (Keeley et al. 2015).

Specific types of training identified as necessary by the literature include training on:

• conflicts relating to consent and how to discuss information sharing with victim/survivors (Behavioural Insights Unit (Vic) 2017)
• proactively sharing information, particularly for early intervention (Keeley et al. 2015)
• ongoing training, including in practice settings (not just online) (Keeley et al. 2015)
• risk assessment, to facilitate timely information sharing (HM Inspectorate of Probation et al. 2017)
• record keeping (ACT Government 2016; HM Inspectorate of Probation et al. 2017); and
• multiagency training (Adams & Lee-Jones 2016; Behavioural Insights Unit (Vic) 2017)

In addition to providing training, recent Australian and international literature indicate that workforces also need to be adequately resourced more generally to cope with the administrative demands of information sharing arrangements (Behavioural Insights Unit (Vic) 2017; Home Office (UK) 2014; Jones 2016; Keeley et al. 2015).

Co-location

As briefly noted earlier, some UK studies of multi-agency collaborations identify co-location of organisations as a strategy for building trust and facilitating information sharing (HM Inspectorate of Probation et al. 2017; Home Office (UK) 2014). However, this is disputed in other literature, and the benefits identified frequently overlap with discussion of broader implications of multi-agency coordination models. For instance, according to Adams and Lee-Jones’ examination of the NSW child safety information legislation:

Some evidence canvassed in this report shows that creating central ‘hubs’ of information that are controlled by a central agency is not the most effective mechanism for sharing information and can lead to delays and other problems. Regimes that allow front-line professionals in government agencies and non-government organisations to exercise their judgement and to share information laterally appear to be a more effective approach (2016, p. 3).

They also note criticisms of the model by an earlier NSW government inquiry, and that ‘New South Wales has since amended the relevant legislation, enabling the direct exchange of information between prescribed bodies’ (Adams & Lee-Jones 2016, pp. 58-9). UK studies have similarly recognised that other collaborative strategies such as conference calls can create the same benefits as co-location (HM Inspectorate of Probation et al. 2017). As such, co-location appears to be of limited significance as an organisational factor in enabling information sharing.

Regulatory challenges

Role definition

Many of the legislative schemes identified in Appendix Three prescribe specific entities that are permitted to share information for family violence risk assessment or protection purposes at an organisational level, although some prescribe professions such as the ‘nurse’ or ‘principal officer’ of prescribed organisations (e.g. s28B of the Children and Community Services Act 2004 (WA) and Part 5A of the Family Violence Protection Act 2008 (Vic)).

Although there is a lack of literature on best practice in prescribing roles and responsibilities for family violence information sharing schemes, work from the UK highlights the need to include organisations from the following fields in family violence multi-agency arrangements: criminal justice, health, education, voluntary sector, welfare agencies, housing and children’s services (Home Office (UK) 2014; Robbins et al. 2014; Steel, Blakeborough & Nicholas 2011). Safety Action Meetings in NSW similarly rely on the involvement of representatives from victim support, health, community, education, accommodation and corrective services (Trimboli 2017).

In relation to prescribing certain positions within organisations involved in sharing child safety information, Adams and Lee-Jones found that ‘regimes that allow front-line professionals in government agencies and non-government organisations to exercise their judgement and to share information laterally appear to be a more effective approach’ (2016, p. 3). Likewise, Gil-Garcia and Sayogo noted that:

Inter-organizational information sharing initiatives often rely on collaborative work involving various managers or personnel from diverse organizations performing different roles at different times. …Considering the complexities of collaboration and the availability of managers' and other personnel’s' time, the existence of formally assigned project managers is crucial to sustain and make the collaboration successful (2016, p. 579).

Similarly, guidance for the Safety Action Meetings (SAM) that form part of the NSW Government’s coordinated response to domestic violence establishes that members:

must be in a senior role with authority to commit to actions, prioritise matters and allocate resources on behalf of their service provider … without having to take decisions or proposals back to their service provider for approval (Safety Action Meeting Manual, p. 14).

This seniority allows actions to be developed and implemented quickly (Trimboli 2017, p. 5).

Taken together these findings suggest prescriptions of information sharing entities should allow both a broad range of organisations and flexible positions within those organisations to share information.

Reporting systems

It does not appear that any of the current state or territory information sharing schemes require prescribed entities to report on the information sharing requests they receive or respond to. However, previous evaluations of the NSW child safety ISS by Keeley e al. (2015) and Cassells et al. (2014) highlight the value of being able to review information sharing practices. Although participants in the case study by Keeley and colleagues acknowledged that such practices would be labour intensive, key Australian and international child safety information sharing scheme evaluations emphasise the importance of collecting systematic data for evaluating the performance of information sharing hubs and outcomes of information sharing and referral practices (Cassells et al. 2014; Home Office (UK) 2014; Keeley et al. 2015). The ACT Government commented that it:

[R]ecognises that accurate and reliable data needs to inform future government decisions on responding to family violence. The ACT Domestic and Family Violence Data Project is laying the foundation for improving data collection in the ACT and ultimately the sharing of this information (2016, p. 10).

Summary - Family violence information sharing schemes

There is a well-established evidence base of legal, technological and organisational factors that influence the effectiveness of information sharing in the context of family violence (Keeley et al. 2015; Yang & Maxwell 2011). In relation to legal or policy factors, regulatory frameworks enable information sharing to the extent that they provide a clear authority for when and how information can be shared. However, it is important that legislation is partnered with adequate training and guidance to be effective and ensure ‘action is taken promptly, particularly for high risk cases of domestic violence’ once information is shared (Taylor et al. 2015, p. 40). This requires consideration of the range of other technological and organisational factors that impact on information sharing in the context of family violence.

Technological factors impacting on effective information sharing include systems of collecting, recording and storing information, which can inhibit or enable effective family violence information sharing either directly or indirectly by undermining confidence in information security. Evidence of these factors led the RCFV to conclude ‘the primary organisations in the family violence system see IT system reform as a priority, as well as a major challenge’ (State of Victoria 2016a, p. 181). Previous research and literature indicate that technological difficulties are not insurmountable and tend not to hold the same weight in terms of barriers to information sharing for experienced workers (Keeley et al. 2015; Yang & Maxwell 2011)

Work on family violence information sharing schemes indicates that organisational factors hold the most weight in terms of effective information sharing (Keeley et al. 2015). Organisational factors such as trust, interagency relationships, shared understandings and cultures of sharing information are key determinants of successful information sharing (Keeley et al. 2015; Yang & Maxwell 2011). These factors inform and impact on each other facilitating and/or inhibiting effective sharing of information, particularly in the context of family violence where tensions about confidentiality and privacy persist.

Finally, it is important to note that introducing information sharing arrangements is not a panacea for effective risk assessment and management in the context of family violence (Jones 2016; Keeley et al. 2015). Information sharing is only one aspect of successful multi-agency collaboration that is necessary to support the safety of family violence victim/survivors (Healey et al. 2013).