Bridging the data gaps for family, domestic and sexual violence

In February 2013, the ABS published Defining the data challenge for family, domestic and sexual violence, Australia, 2013 (cat. no. 4529.0). Defining the data challenge provides a systematic and rigorous framework for family, domestic and sexual violence, through six measurable ‘elements’. These elements are designed to organise information needs in a coherent way that supports a set of statistics about family, domestic and sexual violence. The framework delineates important concepts, and organises them into a logical structure that shows the key relationships, processes and flows that exist between the following six elements.

1. Context – The environmental and psycho-social factors that influence community and individual attitudes, and otherwise provide the context for the occurrence and experience of family, domestic and sexual violence.
2. Risk – The actual and perceived risk of family, domestic or sexual violence.
3. Incident/Experience – The occurrence of family domestic or sexual violence, and the way this violence is experienced.
4. Responses – The response to family, domestic and sexual violence by individuals, families, the community, and formal or system responses such as the justice, health and community services sectors.
5. Impacts and Outcomes – The short, medium and long-term impacts and outcomes of family, domestic and sexual violence for victims, perpetrators, families and the broader community and economy.
6. Programs, Research and Evaluation – The response of research and education to family, domestic and sexual violence to inform targeted prevention, intervention, and support services.

For more detailed discussion of these concepts, refer to the information paper Defining the data challenge for family, domestic and sexual violence (cat. no. 4529.0).

One of the key challenges in measuring family, domestic and sexual violence lies in the complexity of the behavioural acts, relationships and situations that can be considered.

Definitions of family, domestic and sexual violence are shaped by the context of enquiry and informed by the strategies, perspectives and agendas of individuals or organisations. Definitions may be based on, for example, specific legal, policy, service provider or research perspectives. The impact that definitions can have on the use, or role of data in regard to family and domestic violence is discussed in ‘The gender debate in domestic violence: The role of data’ (Australian Domestic and Family Violence Clearinghouse (ADFVC), 2013). Similarly ‘The nature and extent of sexual assault and abuse in Australia’ discusses the effect that definitions of sexual assault and abuse can have on data and understandings of sexual violence in its broader form (Australian Centre for the Study of Sexual Assault (ACSSA), 2012).

Defining the data challenge outlines these definitional complexities but does not seek to set a definition. Instead, it provides a common language for family, domestic and sexual violence by outlining the potential behaviours and relationships that should be considered when establishing a measurable definition. It recognises that the dynamics of sexual violence incidents can be different and occur in the context of a wider range of relationships, both known and unknown, between perpetrators and victims. The broad term ‘family, domestic and sexual violence’ is used in this publication as an umbrella term that encompasses all the potential behaviours and relationships outlined in Defining the data challenge. This is a combination of the terms ‘family violence’, ‘domestic violence’ and ‘sexual violence’ (Diagram 3).

From a statistical measurement perspective, some issues are specific to family and domestic events only, while others are relevant to sexual violence. Accordingly the term ‘family and domestic’ is used to encompass violence that occurs in a context where there is a particular kind of known relationship between the victim and the perpetrator. This can be narrowly defined (e.g. intimate partner violence) or broadly defined (e.g. people living in the same house). Sexual violence may, or may not, be a feature of a family and domestic violence event. Similarly, ‘sexual violence’ can be considered an umbrella term that includes all the behaviours that are outlined in Defining the data challenge. More specific terms such as ‘sexual assault’ and ‘sexual abuse’ are used verbatim when referring to existing data reports and data collections.

There are a number of challenges in measuring the extent of sexual violence in Australia. Sexual violence can include a broad range of behaviours; however, under an offence-based definition incidents of sexual violence are narrowly defined as sexual assault that are classified as criminal acts by law. As a result, a reported incident may not be considered to be ‘sexual assault at law’, and, consequently may not be recorded. Individual perceptions of what constitutes sexual violence can also result in a person not reporting an incident/event as sexual violence/assault. This can present problems in attempting to collect and classify information related to a person’s perception of the incident.

There are a range of organisations involved in the provision of services to those affected by family, domestic and sexual violence, and subsequent collection of information in this field (e.g. justice, health and human service organisations). Often the responses to those who have experienced family and domestic violence events differ from responses to those that have experienced sexual violence. These organisations tend to be independent entities and as such are not required to share or compare information across sectors or jurisdictions. Accordingly data collection is organisation specific and results in little commonality on data procedures and definitions.

The distribution of information across sectors (police, justice, health and emergency, social and human services) and barriers to reporting and access to appropriate services make it difficult to measure the true extent of family, domestic and sexual violence. Currently, information about these forms of violence is held in a variety of survey and administrative by-product datasets. The following is a summary of existing data sources in Australia (for more information see Appendix 3). The sources included here are a selective list of national data collections that currently exist. For a full list of available data sources please consult the Directory of Family and Domestic Violence Statistics (cat. no. 4533.0).

• The National Community Attitudes to Violence against Women Survey (NCAS) monitors awareness and attitudes understood to influence the perpetration of violence, women’s victimisation and responses to violence, as well as the responses of family and community members, and professionals who witness violence, or to whom women turn for help (VicHealth, 2010). Attitudes are also an indicator of broader social norms (the behavioural expectations of society) as they pertain to violence against women. It is important to monitor social norms as they have been found to influence the behaviour of individuals (VicHealth, 2010; Flood and Pease, 2007). NCAS gauges attitudes toward physical, sexual and psychological abuse as well as property damage (ABS, 2011a). The primary purpose of the survey is to gauge the attitudes of the whole community toward violence (for more information see Appendix 3).
• The Personal Safety Survey (PSS) is an in depth measure of individual experience of a number of different types of violence and perceptions of personal safety. It provides information about men's and women's experience of violence in the last 12 months by different types of male and female perpetrators (e.g. current/previous partner, other known man or woman, and stranger) since the age of 15. Aspects of family, domestic and sexual violence captured by the PSS are physical assault or threat, sexual assault or threat and emotional abuse as well as harassment and stalking (ABS, 2011a). The 2012 data release will enable some comparisons to be made to 2005 data (ABS, 2006) (for more information see Appendix 3).
• The Crime Victimisation Survey (CVS) also provides a measure of personal experience of violence and provides an indicator of prevalence, as well as some information about victim and incident characteristics. It is run annually and provides a measure of change in the number of victims and victimisation rates over time. Information about the relationship of the offender to the victim at the time of the most recent incident of physical assault, threatened assault and robbery is published where data are of sufficient quality. While CVS includes some data items of interest, its usefulness for measuring family, domestic and sexual violence is limited as it does not collect specific information about these topics (ABS, 2011a) (for more information see Appendix 3).
• Administrative by-product data can be used as a measure of family, domestic and sexual violence experiences that have come to the attention of organisations through the delivery of services. One example is Recorded Crime – Victims, Australia (cat. no. 4510.0) which is produced using police data. The aspects of family, domestic and sexual violence are limited to behaviours that are outlined by law and are focused on actual or threatened physical acts. The information is limited to events that are in breach of relevant state and territory criminal law (for more information see Appendix 3).
   

Together these data sources contribute to the current evidence base for family, domestic and sexual violence in Australia. The limitation of these datasets is that there is no easy way to identify if the violence experienced was in fact a family or domestic related event. A proxy measure is often used to determine if violence includes a family or domestic element by analysing the relationship between perpetrator and victim. This, however, is limited by the relationship types included in the existing administrative data systems or survey questions. Further, definitions of violence used in some of these sources are limited to actual and threatened physical behaviours; broader definitions that include emotional, physical, psychological and financial behaviours are not included in all data sources.

These data sources are effective at measuring different aspects of family, domestic and sexual violence. However, the difference between information sources provides only a partial picture of the true extent of family, domestic and sexual violence in Australia. The rest of this publication will explore ways to improve the evidence base through better utilisation of existing data captured by organisations.

Research questions and definitions used in data collection are influenced by the context of inquiry and theoretical approaches. The ABS coordinated a priority setting exercise that aimed to align over 100 research and policy questions identified in Defining the data challenge to the National Plan Indicators (see Appendix 1 for methodology). The outcome of this process was the identification of nine refined priority research questions, as follows:

• What is the prevalence and incidence of family, domestic and sexual violence in Australia?
• Who is involved in family, domestic or sexual violence events?
• Why do/don’t those who experience violence disclose events of family, domestic and sexual violence events?
• What services and support are needed for those who experience family, domestic and sexual violence?
• Which formal interventions reduce the occurrence of family, domestic and sexual violence events?
• What are the impacts and outcomes of family, domestic and sexual violence?
• What are the impacts and outcomes of engagement with services, programs and support?
• What are community attitudes to family, domestic and sexual violence?
• What are the impacts of public policy approaches and educational campaigns targeting family, domestic or sexual violence?
   

From a data perspective there are a number of specific data items that are required to be collected for each of the priority questions (see Transforming data into statistical information for more information).

In order to develop a flexible evidence base, data collection should include information that describes key characteristics of people who have experienced family, domestic and sexual violence. Groups that are of particular interest (perpetrators, victims etc.) are not explicitly identified in the high priority questions. However, from a data perspective, information should be collected in ways that allow data to be disaggregated to identify these groups. Data collection techniques should attempt to address the unique challenges these groups present due to individual circumstances and living arrangements (e.g. in care, institutions or remote areas).

Defining the data challenge outlined the specific population groups of interest as:

• children;
• young women;
• pregnant women and women with children;
• Aboriginal and Torres Strait Islander people;
• Culturally and Linguistically Diverse (CALD) communities;
• people with disabilities;
• people living in rural and remote areas;
• older people; and
• Lesbian, Gay, Bisexual, Transgender and Intersex (LGBTI) people.
   

In addition, data should be collected in ways that allow for information to be compared over time. Societal views and attitudes about family, domestic and sexual violence are subject to gradual change over time. As attitudes change, policy and legislation develop that reflect these shifts. Information is required to measure these movements. The ability to assess change over time is also important for measuring the progress of the National Plan and it is expected that this will occur as a consequence of creating a strong evidence base. Consistent collection of data that informs these questions will assist in establishing a baseline measure and also facilitate comparisons over time.

State and territory government structures determine which organisations are responsible for family, domestic and sexual violence responses. These vary significantly across jurisdictions. The initial environment scan revealed that there were a number of key organisations involved in responses to, and subsequent collection of, administrative information for family, domestic and sexual violence events. Each organisation is separately responsible for collecting information about the work or service they provide (see Appendix 3 for more detailed information). The environmental scan highlighted a number of issues for family, domestic and sexual violence data. The complex service arrangements for responding to family, domestic and/or sexual violence shape the information environment and are illustrated in Diagram 4.

* Victim support services are provided in some states and territories specifically to victims of crime, this includes events that are not family, domestic and sexual violence events. These services often provide referrals to existing family, domestic and sexual violence services where appropriate. 
** Family, domestic and sexual violence services include counselling, disability assistance, family services, health care providers, mental health, Centrelink, immigration, housing, homelessness and drug and alcohol services (both housing and homelessness services provide data to AIHW).

The key issues uncovered include, but are not limited to the following:

• police and human service information is collected and reported at the state level but is not collated by a central organisation for release at a national level for family, domestic and sexual violence information;
• police protective orders and associated breaches are not well captured in existing national police and courts collections;
• referral pathways and take up of services are not captured in all states and territories, nor at the national level;
• an understanding of provision of human services and when to intervene is not available;
• specialist court information is often not represented in national publications;
• civil court matters (where most applications for protective orders are heard) are not reported in existing national collections; and
• a number of agencies (including hospitals, emergency services and child protection) collect administrative data in a manner that is not conducive to transforming the data into statistical information for purposes other than that for which they are collected.
   

There are a number of national data collections that contain information about family, domestic and sexual violence, some survey based and others, administrative data. For the latter, administrative information is collected in each state and territory, which then contributes to larger collections that are reported nationally. The environment scan identified a number of key national surveys which have the potential to be augmented to include family, domestic and sexual violence information. These include the following:

• Personal Safety Survey;
• National Community Attitudes towards Violence against Women Survey;
• Crime Victimisation Survey;
• Recorded Crime, Victims – Australia;
• National Hospital Morbidity Database;
• National Non-Admitted Patient Emergency Department Care Database;
• Child Protection National Minimum Data Set;
• Specialist Homelessness Services Collection;
• National Coronial Information System;
• National Homicide Monitoring Program; and
• Community Legal Services Information System (See Appendix 3 for detailed information about these sources).
   

The lack of connectivity between organisations has a number of implications as highlighted by the key findings identified through the environmental scan:

• there is no consistent data definition or identification method for family, domestic and sexual violence;
• from a data perspective, there are a number of entry points, therefore increasing the likelihood that a person may be counted more than once for the same event;
• a large number of organisations are involved in providing services and collecting information about those that have experienced family, domestic and sexual violence;
• information is collected differently across organisations, and sometimes within organisations;
• IT systems and infrastructure used between organisations differ;
• while some standards and classifications are used, these are often disparate and organisation specific; and
• key organisations collate and report on service provision, however, not from a family, domestic and sexual violence perspective.
   

A number of initiatives identified as part of the environment scan are working to overcome these complexities. These include:

• An example of states and territories working together was the Australian Domestic and Family Violence Death Review (ADFVDR) Network project.⁶ This network was established in 2011 following a review of family, domestic and sexual violence related deaths in Australia. Representatives of the ADFVDR Network include the Domestic Violence Death Review Team (New South Wales), Domestic and Family Violence Death Review Unit (Queensland), Domestic and Family Violence Death Review (South Australia) and Victorian Systemic Review of Family Violence Deaths (Victoria). The ADFVDR is an example of agencies working across state and territory boundaries to improve knowledge and identify the context and circumstances in which domestic and family violence deaths occur (Coroners Court of Victoria, 2012).
• There are also examples of local agencies working together to share information and IT infrastructure to improve and co-ordinate their response (Appendix 4). For example, Tasmania’s ‘Safe at Home’ program includes the Victim Safety Response Team, Police Prosecutions, Family Violence Counselling and Support Service, Court Support and Liaison Service, Child Protection and Special Needs Liaison Service. These local agencies meet weekly to ensure coordinated case management. They also use the Integrated Case Coordination Management System (ICCMS) database to link data from the Police Family Violence Management System (FVMS) with data from the Department of Justice databases (Department of Justice Tasmania, 2009).
• Risk assessments are another initiative used in some states and territories to support integrated approaches to family and domestic violence. Risk assessments can enable uniform screening and subsequent identification of family and domestic violence and facilitate the management of risk through integrated case management.⁷ This understanding of risk assessment incorporates a screening process and while this is a feature of the examples found in the Australian context this is not necessarily a feature of all tools.⁸ There are a number of screening and assessment tools available and often the choice of tool is informed by a broader framework (Robinson and Moloney, 2010), such as the Family Safety Framework used in South Australia. In some states (e.g. Victoria, South Australia, Western Australia and Tasmania), risk assessments are used by multiple agencies that incorporate screening and risk management (Appendix 5). These tools are not standardised or comparable between jurisdictions and consequently there is currently no standard national tool for assessing or screening family and domestic violence events.
• A number of NGOs across Queensland are providing domestic and family violence service information to a central database managed by the Centre for Domestic and Family Violence Research (CDFVR).⁹ This information is collated and released publicly (in accordance with data privacy and confidentiality considerations) to provide a profile of clients and their support needs. There are some limitations with this information; data are collected over a two week period twice a year and data provision by organisations is optional. However, this initiative demonstrates that collection of information that informs purposes other than service provision, allocation or budgets, funding applications or tracking performance is possible. To expand this model to a national scale will require significant effort, commitment and buy-in at all levels, including NGOs and the wider community.

The difficulty of identifying family, domestic and sexual events in existing datasets and an absence of co-ordinated data collection at the sector, state and territory or national level are key challenges in creating nationally consistent data.

⁶ For more information about the Australian Domestic and Family Violence Death Review see http://www.coronerscourt.vic.gov.au/resources/54bbc2f9-bb23-45c0-9672-16c6bd1a0e0f/vsrfvd+first+report+-+final+version.pdf
⁷ For a more detailed discussion of the assessment of risk assessments refer to http://www.safeathome.tas.gov.au/__data/assets/pdf_file/0011/142310/RAST_Report_Analysis_of_Risk_Assessment_Screening_Tool.pdf
⁸ For a detailed discussion of the differences between screening and risk assessment refer to Robinson, E and Moloney, L 2010, Family violence: Towards a holistic approach to screening and risk assessment in family support services, Australian Family Relationships Clearinghouse, Melbourne.
⁹ For more information about the Queensland Centre for Domestic and Family Violence Research refer to http://www.noviolence.com.au/

The table below highlights that the current family, domestic and sexual violence sources are able to partially provide answers to the high priority questions identified.

It is not unexpected that existing data are unable to fully answer the high priority questions; the primary purpose of these collections is not for family, domestic and sexual violence information. However, this does not preclude the use of these datasets for other statistical purposes. In fact making use of existing data for other purposes is a cost effective way of building an evidence base.

A number of information units can be collected about who is involved and in what events they are involved, as illustrated in Diagram 5. It is important to note that a number of data items, such as location, age, gender, etc. would need to be collected to provide information about the units outlined below.

A range of statistical information can be created from the same data providing that data are collected in information units that allow for the generation of different combinations. Frameworks (such as Defining the data challenge) often identify counting units of interest for a particular topic. As people are central to these issues, the most commonly used counting unit is a person. To support the National Plan and to improve the evidence base for family, domestic and sexual violence there were three key information units identified for counting; person, transaction and event.

These three units are influenced by a range of contextual factors that relate to systemic, community or individual characteristics as well as the formation and effect of attitudes (Diagram 6). The context element, illustrated in the blue box, reflects that the information included occurs outside of a family, domestic and sexual violence event. It comprises information about the factors that shape understandings of family, domestic and sexual violence including aspects that influence community and individual attitudes, changes to policy and education campaigns. While many of the components in this element are not ideally suited to statistical measurement, a number of sources of information may be consulted to build an understanding. It is expected that a large proportion of this information could be collected through survey data.

The grey box includes information that is likely to be extracted from administrative by-product data and is specific to an event of family, domestic and sexual violence. The ‘event’ unit consists of information about the incident, such as date, time and location as well as information about the other two units: person and transaction.

Person level information is collected by a number of organisations and the type of information collected is determined by the function of the organisation (e.g. courts focus on defendants and services against sexual violence focus on victims). Likewise transaction information is collected by a range of organisations. Formal transactions involve a person reporting to or engaging with services provided by formal systems that focus on detection (criminal justice system) and support (health and community services). While formal transaction data are captured as a by-product of administrative processes informal transactions are actions that do not, by their nature, involve reporting to a formal organisation. Information is required about informal responses to family, domestic and sexual violence, including why people do not report or engage with services, however this information is best collected by a survey.

It is important to note that Diagram 6 does not outline the data requirements in detail. This information will be included in the forthcoming National Data Collection Reporting Framework. It does, however, provide a succinct way to further assess the utility of current data collections for family domestic and sexual violence.

Information units (identified in the previous section) are designed to underpin a flexible evidence base, and understanding how the existing data sources support these units is crucial. Using information units as a basis for analysis, the existing data sources were further assessed to determine the amount of information currently available.

Table 2 highlights that, although there are gaps in the information required to strengthen the evidence base, there are a number of areas where improvements in data collection could yield large gains.

At present the data required to support a flexible evidence base for family, domestic and sexual violence are not available. Information does exist but is not currently an adequate measure of the extent of violence against women and their children. It is acknowledged that the consequence of a focus on administrative data in this paper is the improvement of reported events of family, domestic and sexual violence only. Surveys collect information about unrecorded/unreported events and surveys such as the PSS and NCAS are currently in place to measure these factors.

The analysis undertaken identified two types of data gaps - the first was where information is not collected at the national level, and the second was where information currently collected is not sufficient to support family, domestic and sexual violence information. Regardless of the type of data gap, the key to strengthening the current family, domestic and sexual violence evidence base is to transform the data collected in current administrative systems.

The augmentation of existing datasets is the key to making substantial gains in improving the evidence base and transforming data into statistical information that is fit for a number of different purposes.

The environment scan identified a number of datasets that are currently turned into statistical information about specific areas of interest, such as the NHMD which holds a range of data about hospitals in Australia. However, the purpose of this database is not specific to, or allows easy identification of, family, domestic and sexual violence events. Transforming datasets like these, that are currently inadequate (but adequate for their current purposes) into adequate data for family, domestic and sexual violence information (outlined in Diagram 7), is a key challenge. Any efforts to improve organisations statistical information should complement the primary purpose of data collection, and not be at the expense of core business functions.

Transforming inadequate data into adequate data is crucial in strengthening the existing family, domestic and sexual violence evidence base. The actions outlined in the following chapter will help address the deficiencies in existing datasets and improve the overall utility of administrative data.

The three themes do not operate in isolation; it is expected that enhancements in each theme will enable further improvements to other themes and enhance the quality and value of the Australian statistical system. Enhancements in theme one are also expected to have benefits in increasing the utility of existing data sources, especially legislative amendments that enable agencies to collect and share information.

While most if not all organisations collect administrative data to support service provision, allocate budgets, apply for funding and track performance, often there are cost implications for adapting the data to enable the production of statistical information. The following themes can be used as a guide to inform future resourcing decisions.

Many data collections have the potential to provide important statistical information that forms part of the evidence base for family, domestic and sexual violence, if appropriate statistical infrastructure is in place. The National Statistical Service (NSS) describes statistical infrastructure as statistical standards, policies and tools that are shared to maximise the value of investment, reduce provider load and support integrated statistics. Enhancements to these areas provide the foundation for this theme, in particular actions that encourage the:

• use of data standards and definitions that enable consistent identification of family, domestic and sexual events;
• application of data quality assessments to encourage the production of accessible information; and
• consideration of amendments to legislation that enable the use of data for statistical purposes.
   

There are a number of ways to transform existing data into statistical information. Some suggested actions are outlined in Table 3.

The existing information for family, domestic and sexual violence can be sourced from many collections including a number that are not currently publicly available. Those available provide detailed information on various aspects of family, domestic and sexual violence but there are a number of data gaps due to coverage limitations and methodological differences. Through improving data collection practices, populations of interest will become more visible in existing datasets. However sub population groups which are not accessed in traditional survey collection methods and administrative data will require targeted strategies that complement traditional methods. It is advised that identification of suitable data holdings for those in unique living circumstances (e.g. care facilities or institutions) be investigated in future work.

This evidence base has potential for considerable adaption for improved statistical use through data enhancement techniques such as augmentation, sharing and linkage. There are many options for data linkage, ranging from simple enhancements, such as the addition of small area indicators, through to more complex data matching of multiple sources. However, there are significant issues that must be addressed to make data linkages more viable. These mainly stem from privacy and data access legislative requirements, at both state and territory and Commonwealth levels. Opportunities to address these in ways which improve access while maintaining privacy should be explored further.

Additionally, the quality and comparability issues raised in theme one are important enablers that underpin the ability to achieve data matching. If these can be overcome, data linkage strategies can present an effective way to address many high priority information gaps by fully utilising existing data. These types of data linkage options need to be considered in a whole of government approach to effectively govern data and maximise cross linkage between government sectors (e.g. crime and justice, health, education).

Exploring the potential to use data sharing principles and working towards the goal of data linkage are encouraged as guiding aspirations for building the evidence base for family, domestic and sexual violence.

A wide range of statistical information is being collected across jurisdictions and to get the best value from these assets a holistic and coordinated approach to the production of family, domestic and sexual violence is needed. Table 3 outlines the relationship of the three themes to the key issues and highlights the potential benefits from improvement in these areas.

These three themes form the foundations for a flexible family, domestic and sexual violence evidence base. Each theme is designed to contribute to a stronger evidence base and it is expected that enhancements in the themes will work together and promote better practice in all areas of the creation of statistical information. Progressing enhancements in each of the themes will lead to a collaborative community of government agencies and NGOs working together to build a rich statistical picture for a better informed Australia.

There are a number of suggested actions also included, and this list is indicative of some of the possible actions associated with enhancement in the key areas.

Through extensive consultation Defining the data challenge identified more than 100 relevant research and policy questions. This list was refined to eliminate overlaps and to align with the four national indicators. The outcome of this was 38 questions grouped in five areas as displayed in Table 4.

The five key areas of interest were identified from these questions:

• prevalence and incidence of family, domestic and sexual violence;
• characteristics of victims of family, domestic and sexual violence;
• characteristics of perpetrators of family, domestic and sexual violence;
• responses, and services provided, to those who have experienced family, domestic and sexual violence events; and
• government (including education and policy) and community (including beliefs and attitudes) responses.
   

To determine the priority ranking of the questions, a variety of government agencies and NGOs were consulted to assess the questions in terms of importance, frequency and level of geographic need. The 38 questions were separated into three tiers, with the first tier representing questions of critical importance. The first tier comprised 12 questions related to prevalence and incidence, effective service provider response and support mechanisms for individuals who have experienced family, domestic and sexual violence. An examination of the first tier priority questions highlighted the interrelationships between concepts and data items required to build an evidence base. It also highlighted the primary focus on women and children. The process of refining the questions is summarised in Table 5.

An environmental scan was conducted to identify key data holdings and flows between organisations. A two-stage process was used to minimise resource and time constraints. One state was chosen as a case study, the findings from this state were used to inform desktop research and conversations with other states and territories. The environmental scan commenced in Victoria and discussions were held with:

• Department of Premier and Cabinet;
• Department of Justice (Family Violence & Sexual Assault Unit, Courts and Victim Support Agency);
• Department of Human Services (Children, Youth & Families. Child Protection, Housing & Community Building and Family Violence & Sexual Assault Unit);
• Ambulance Victoria;
• State Coroner;
• Department of Health;
• Community Health providers;
• National Coronial Information System (NCIS) data custodians;
• Australian Institute of Family Studies (AIFS); and
• Australian Institute of Health and Welfare (AIHW).
   

These meetings provided an understanding of the data complexities and data flows within and between organisations. This informed a national review of the data environment. Desktop research and discussions with ABS offices in each state and territory were conducted to locate publicly available information about the core organisations. Existing networks and stakeholder groups were then used to confirm where more agency specific information was required. The detailed findings presented in Understanding the environment and Appendix 3 are based largely on the Victorian case study and supplemented by state and territory specific desktop research.

Police services have the potential to capture unique information pertaining to both victims and perpetrators. The information extracted from police information systems differs across the states and territories according to business practices and legislative and legal frameworks. These legislative differences result in different policing practices which can influence the data that are collected. Regardless, police generally have consistent and standardised data recording practices that are reflective of the similarities of underlying police processes. Diagram 9 presents a schematic representation of police data information pathways.

One key difference between police jurisdictions is the method used to identify family and domestic events. The practice of identification of these events is embedded in relevant legislation and policy. Practice varies from professional judgement through to the use of screening tools as part of risk assessment frameworks (Appendix 5). Regardless of the identification process utilised, each state and territory has a family, domestic and sexual violence identifier in their administrative dataset (in Tasmania a separate system exists to record all incidents that relate to family and domestic violence). However, as there is no uniform process to identify these events across state and territories in the data the attached identifiers are unsuitable for national comparisons.

In most states and territories, police powers have been extended under family and domestic violence legislation to include police - issued protection orders. The duration of these orders can vary across jurisdictions, from 24 hours to 12 months. Some orders are linked to court applications (civil and criminal), while others only serve as a temporary measure when court is out of session (in essence a court summons) (ALRC, 2010). As a result, police issued protection orders are not comparable across states and territories, nor are they an accurate reflection of all family and domestic protection orders as they are not the only orders used in these situations.

Data on all aspects of sexual violence in the broadest sense are not currently collected through police administrative information systems. However, sexual assault data are collected, as these offences are framed by strong legislative and criminal procedures. Diagram 5 depicts sexual assault within a family and domestic context; however, they can also occur in situations where the perpetrator is not known to the victim. The true extent of sexual violence is underestimated as sexual offences are often not reported to police, or indeed disclosed to anyone (ALRC, 2010). There has been significant work invested in the legal framework and response to sexual assault offences across the nation but a range of inconsistencies still exist between jurisdictions (ALRC, 2010).

Regardless of offence type, police responsibility to victims is outlined in relevant charters or codes of victim’s rights, including an offer of a referral to support organisations (Australian Federal Police, 2011). In some state and territories this is through a formal referral pathway such as Supportlink, an electronic referral system that allows officers to request support services for a victim through a single platform.10 This service is able to monitor and track referrals and to report on service take-up and then feed this back to the police to improve practice in the field. In other states and territories this is a manual process which is managed with the support of local service provision organisations. Current police information systems are not designed to measure the outcomes of referrals as these do not relate directly to policing duties. Quantifying referrals and service utilisation across states and territories is an important part of establishing a national dataset but is beyond the scope of police administrative systems.

While this data has the potential to inform our understanding of those involved in police response events, data are still subject to quality issues around input because they are a by-product of policing duties. Further information may be found in Measuring Victims of Crime: A Guide to Using Administrative and Survey data (cat. no. 4500.0.55.001). To understand a victim’s full experience, information about service provision and justice responses is also required. A key limitation of police data are a lack of integration with other justice datasets which is where perpetrator outcomes are often determined.

¹⁰ More information about Supportlink is available at their website http://www.supportlink.com.au/

Court (civil and criminal) data are largely focused on the perpetrator of the violence and are highly reliant on the quality of the information collected by police. The family and domestic violence indicator used by police is not included in the information shared with courts as these information systems are not linked.

Currently, family and domestic violence incidents can be identified through a number of avenues: establishing a familial or domestic relationship between victim and offender; identifying cases that involve family and domestic violence legislation or violence orders; and cases that are presented before specialist family violence courts (not established in every state and territory) or family courts. The drawback of all these approaches is the potential for some family and domestic violence events to be overlooked as some cases will be excluded because relationship to victim is not included or a perpetrator may be charged under different legislation.

Sexual assault is identifiable in court data by the charges laid. Understanding if a sexual assault occurs in a family and domestic violence relationship is more complicated, as information collected about the relationship to the victim is not always reliable.

Court data has the potential to provide a good understanding of the outcomes of family, domestic and sexual violence in the justice system. However, court data are limited to only those events where sufficient evidence exists to enable a case to proceed through the judicial process. As a result, care should be taken when interpreting these statistics.

The role of the Coroner is to investigate all reportable deaths in the respective states or territories, in an inquisitorial manner. This role involves establishing the identity of the deceased, and the date, cause and circumstances of death. A coronial investigation may also aim to prevent similar deaths in the future. In an effort to understand the underlying causes of a person’s death, the coroner considers a wider range of information as they are not bound by the same rules of evidence procedures as other court jurisdictions. The coronial process also usually occurs following criminal and civil procedures and is often reliant on the conclusion of these procedures before a coronial inquiry begins. As a result, it may be some time after a death has occurred before findings from these inquiries are released (Coroners Court of Victoria, 2011).

Some information collected by state and territory coroners is uploaded to the National Coronial Information System (NCIS). Like courts, coronial information does not record the relationship between parties, making family and domestic related deaths difficult to identify.¹¹ To understand deaths that are the result of family and domestic violence, a number of states and territories have implemented family violence death reviews and are demonstrating an ongoing commitment to understanding and identifying these events through the establishment of networks like the ADFVDR (see Data environment – key issues for more information).

¹¹ Extra data fields to capture the relationship between victim and perpetrator for assault related fatalities are to be added to the NCIS data collection in 2013-14.

All governments have the responsibility to ensure human services are provided in their jurisdiction. Not all states and territories use the term Human Services (such as Queensland’s Department of Communities, Child Safety and Disability Services), for the purposes of this report, such services will be collectively referred to as Human Services. Human Services work with a number of key organisations, such as service providers, police, courts, corrections and Departments of Justice.

Often Human Service departments are not responsible for the direct provision of family, domestic and sexual violence services. Instead, they provide funding for other organisations to deliver services such as crisis accommodation, counselling and other supports. While some administrative information is collected as a by-product of Human Service provision, this is usually voluntary. Depending upon the service, information about the reason for attending a service may be collected if offered by the client only. The importance of not distressing clients in times of need is a concern that is paramount in Human Service provision. As such, data collection is often viewed as secondary to client needs (Ruuskanen and Kauko, 2008).

Support services include a large number of organisations that provide services in the community including community health, counselling and legal services as well as centres specifically for those affected by family, domestic and/or sexual violence. Nationwide there are vast numbers of individual service providers and it was not feasible for this project to comprehensively describe them all. However, federal government funding arrangements (in part) for a number of these services results in similar processes of information collection and data flows across organisations and jurisdictions.

Support services are often provided by NGOs that capture information about family, domestic and sexual violence through their day-to-day service delivery. Information collected by these service providers is held in a variety of formats, such as paper-based client files or electronic documentation. The primary purposes of collecting this information are to inform service provision, allocate budgets, apply for funding and track performance. Identifying whether service provision is attributable to family, domestic or sexual violence can be challenging. Unless a victim approaches a specific family, domestic or sexual violence service provider, the true reason for seeking support may not be revealed immediately. Case notes may contain information pertaining to the victim's experience of family, domestic or sexual violence. However these types of records are subject to stringent privacy, confidentiality and ethical policies. As a result, these organisations hold discrete, local collections of data about those that access services.

All levels of government have committed to the National Affordable Housing Agreement (NAHA) and the National Partnership Agreement on Homelessness (NPAH). The NAHA provides the framework for all governments to achieve access for all Australians to affordable, safe and sustainable housing. This agreement provides direction for a range of social assistance measures, including social housing, other housing assistance (to renters and those purchasing their homes), and support to people experiencing homelessness.

Social housing and homelessness services report that they deal with a substantial number of clients who also experience family, domestic and sexual violence. These organisations not only collect information about their clients and the reason they need assistance, but are also to directly link it to family and domestic violence. It is unclear if this is possible for sexual violence.

Social housing is provided by the government and community sectors, primarily to those on low to moderate incomes. This information includes administrative data collected by jurisdictions and community housing organisations. Socio demographic data as well as performance indicator data are collected to support the broader policy initiatives outlined in the agreements. Data about family, domestic and/or sexual violence is collected on application for social housing and recorded within the greatest need categories, namely, ‘their life or safety was at risk in their accommodation’. This category is defined as: ‘domestic violence or subject to sexual/emotional abuse; or subject to child abuse; or at risk of violence or who feared for their safety in the home environment.’ The information in the greatest need category published in Housing assistance in Australia (AIHW, 2012a) does not include this category.

Specialist homelessness organisations provide emergency and medium-term accommodation and a range of other support services to people who are homeless or at risk of homelessness (AIHW, 2012b). The Specialist Homelessness Services Collection (SHSC) is an important source of information about people who are homeless or at risk of homelessness due to domestic or family violence. It collects information about people who are displaced because of violence, but does not capture data about people who have utilised alternative accommodation or who do not seek assistance. Information about people who receive specialist homelessness services is published in AIHW releases based on SHSC data (e.g. AIHW 2012a).

Departments responsible for child protection work closely with other organisations that provide support services. The primary function of child protection units is to protect children who are at risk of, or experiencing abuse, neglect or other types of harm.

The processes for notification, investigation and substantiation of child abuse, neglect or harm are similar in each state and territory (AIHW, 2013). The child is the primary client; however child protection workers often work with the family, and refer them to appropriate services that are provided in the community. Information collected is usually associated with the child and while family and domestic violence is taken into account, there is no flag in the national data to identify these cases. The information collected is aggregated and published in Child Protection Australia (AIHW, 2013).

Ambulance services and hospitals collect and store patient data relating to service provision. There is some transfer of data between ambulance and hospitals but the responsibilities for data collection and storage remain separate.

Ambulance services are usually the first to respond to an emergency. In most states and territories, paramedics record patient information electronically at the point of care. Victoria, New South Wales, Queensland, Tasmania and the ACT record clinical and operational data for all emergency incidents on a tablet PC, via a clinical information system called VACIS. This system requires paramedics to record a cause for all trauma patients, and this system does include a case nature of 'assault' as well as an option to specify an assault of a sexual nature. In addition, paramedics can enter free text notes to describe the specific details of the case; however the completion of this field is not mandatory. The data collected are then integrated into each states and territories individual database. 12

Mandatory reporting practices for ambulance officers in regards to family and domestic violence vary in each state and territory (Australian Institute of Criminology (AIC), 2011; Northern Territory of Australia, 2012). In the states and territories where mandatory reporting is in place, the attendance of police at the scene, or transporting of the person to hospital, results in the responsibility for reporting being transferred to that organisation.

Pre-hospital clinical information systems do not generally contain a 'family and domestic violence' or ‘sexual violence’ indicator. While it may be possible to use database filters to identify these patients, it is unlikely that all family and domestic violence patients will be identified in ambulance services data holdings.

Hospital information is collected by emergency stay and hospital admission in each jurisdiction via a Patient Administration System (PAS). The PAS used is at the discretion of hospitals and, as a result, the systems infrastructure and recording on these systems may differ between and within jurisdictions. However, the AIHW undertakes the collection and reporting of national hospital data under the auspices of the Australian Health Ministers’ Advisory Council, National Health Information Agreement. As a result, national data reported is in accordance with the National Minimum Dataset (NMDS) definitions.

Emergency datasets usually contain de-identified demographic, administrative and clinical data detailing patient presentations at public hospital emergency services. Data on most emergency department presentations in public hospitals are compiled in the National Non-admitted Patient Emergency Department Care Database (NAPEDCD). There is no national requirement to report NMDS data items related to family, domestic and sexual violence. Family, domestic and sexual violence data are sensitive and detailed patient information may be held in separate electronic or hard copy files. Often these files are accessible only to the direct patient care team and not released for administrative reporting. The quality and consistency of this locally held information is unknown.

Information about attendance at hospital for an inpatient admission is collected during the hospital stay by various administrative and health personnel involved in the care of the patient. Information collected includes detailed de-identified demographic, administrative and clinical information (disease and procedures) related to the hospital admission. The clinical information is entered into the PAS after the patient is discharged from hospital. Clinical coders assign International Statistical Classification of Diseases and Related Health Problem (ICD-10-AM) and Australian Classification of Health Interventions (ACHI) codes to indicate diseases and procedures that the treating clinician/s has documented during the hospital admission. ICD-10-AM codes exist for the assignment of injuries and maltreatment of adults and children including an indication of the perpetrator. There is also an Australian Coding Standard (ACS: 1909) which guides the assignment of disease codes in these circumstances. While codes exist to classify such conditions, definitive documentation in the clinical notes may not always be present. Currently, there is no national system or requirement in the admitted patient care data specifications to flag family and domestic violence. Once reported to the AIHW data on all hospital admissions are compiled in the National Hospital Morbidity Database (AIHW, 2012c).

Both AIHW datasets contain information that is reported for each hospitalisation, not patient. This means that an individual may appear in the data more than once. At present it is not possible (national level) to identify repeat hospitalisations without the introduction of data linkage procedures.

Data type: Survey
Frequency: Irregular
Custodian: ABS

Until recently, the PSS had been conducted on an ad hoc basis; the next release of the PSS presents results from the 2012 survey. The survey collects information from a sample of households across Australia via personal face to face interviews, conducted with one randomly selected person aged 18 years or over who is a usual resident of the selected household. The sample design and collection methodology have been specifically developed to obtain reliable and valid detailed data about a person’s experience of violence, including partner violence. The PSS provides information about the prevalence of men's and women's experience of violence (i.e. physical or sexual assault or threat) since the age of 15 and in the last 12 months by different types of male and female perpetrators (including current partner, previous partner, boyfriend/girlfriend or date, other known man or woman, and stranger). Information is also collected about men's and women's experience of stalking, sexual harassment, abuse before the age of 15 and general feelings of safety. More detailed information is provided for the most recent incident for each of the different types of violence by a male/female perpetrator, this includes information such as when it happened, location, whether injured, actions taken (help sought, contact with police) and consequences of violence (time off work, fear or anxiety). It also provides detailed information about men's and women's experience of current and/or previous partner violence such as frequency, actions taken (help sought, contact with police) and consequences (children witnessing, temporary/final separations from partner, fear or anxiety). Additional information on emotional abuse by a current and/or male/female previous partner is also collected.

Data type: Survey
Frequency: Irregular
Custodian: VicHealth

The 2013 NCAS is currently being conducted, and is the third in a series of attitudinal surveys (previously conducted in 1995 and 2009) designed to gauge community awareness of, and attitudes toward violence against women. It also monitors changes in attitudes over time and build an understanding of the factors contributing to the shaping of these attitudes. Two further series are planned at four year intervals. The survey includes modules on attitudes towards domestic violence, sexual assault and sexual harassment. Key themes addressed include: how the community defines violence against women; community perceptions of who commits, and who is affected by violence; how common and serious the phenomena is; whether violence can be justified or excused; and the responses of women and others when violence occurs. Respondents are also asked about their behavioural intentions on witnessing domestic violence and their knowledge of sources of assistance. The survey collects information about a survey respondent, who may or may not have experienced family, domestic or sexual violence, and gathers information about their attitudes toward interpersonal violence (VicHeath, 2010). This survey is used to understand community attitudes towards violence against women and tracking changes in these across time, but is not designed to identify or measure the prevalence of, or incidence of family, domestic or sexual violence. NCAS does not ask respondents about their experiences of violence. This is because its primary concern is with gauging the attitudes of the whole community toward violence.

Data type: Survey
Frequency: Annual
Custodian: ABS

This survey estimates the extent of victimisation experienced by Australians for both personal and household crimes including whether or not incidents were reported to police. The types of violence included in the survey are physical assault, threatened assault and sexual assault. It includes information about whether victims reported these incidents to police, characteristics of their most recent incident, and people’s perceptions about whether alcohol or another substance contributed to their most recent incident of physical and threatened assault. (ABS, 2013a). Understanding if these types of assaults occur in a family and domestic situation is not straightforward, as relationship to offender is used as a proxy measure. While the survey provides information about violence in Australia and reasons for reporting/not reporting, it is not designed to produce information about family, domestic or sexual violence as understood in Defining the data challenge. As a result caution needs to be exercised when utilising this information in a family, domestic and sexual violence context.

Data type: Administrative
Frequency: Annual
Custodian: ABS

This publication presents statistics on incidents of victimisation for selected offence types that came to the attention of police and were recorded by them over a calendar year. Recorded crime statistics are the result of policing activities where incidents come to police attention and a subsequent decision-making process is carried out by police in accordance with criminal law and organisational business practices. As such, they are subject to different legislation, rules of operation and procedures in different jurisdictions. Victims are counted once for each type of offence that is reported to police. Therefore, a person reporting a crime with multiple offences in the same incident may either be counted multiple times, or may be counted only once, depending on the types of offences committed during the incident. State and territory recorded crime statistics about victims are published for the following variables: sex, age, relationship of offender to victim, Aboriginal and Torres Strait Islander status, location of offence and the use of weapons for selected offences (ABS, 2012).

Data type: Administrative
Frequency: Annual
Custodian: AIHW

The NHMD is a compilation of admitted patient morbidity data collection systems in all Australian public and private hospitals. It provides administrative data based on the National Minimum Data Set (NMDS) for Admitted Patient Care and has the potential to provide information about victims of family, domestic and sexual violence. There is no uniform flag to identify events related to family, domestic and sexual violence in this dataset; however a set of related codes including the external cause code of the injury may be used to identify cases in the dataset. A limitation of the data collection is that a record is created for each hospitalisation, not patient. Data are published by the AIHW in Australian Hospital Statistics (AIHW, 2012c).

Data type: Administrative
Frequency: Annual
Custodian: AIHW

The NAPEDCD contains data collected from the state and territory health authorities about non-admitted patients treated in most public hospital emergency departments. The statistical counting unit for the collection is a patient presentation. This dataset collects demographic, triage and waiting time information. The dataset does not contain clinical information about the presenting patient (for example diagnosis or type of care provided) and it is not possible to identify events related to family, domestic and sexual violence. A limitation of the data collection is that a record is created for each patient presentation, not patient. Data are published by the AIHW in Australian Hospital Statistics (AIHW, 2012c) and Australian Hospital Statistics: emergency department care (AIHW, 2012d). In addition, emergency data (waiting times, time in emergency department and presentations) are reported on My Hospitals and under the National Healthcare Agreement.

Data type: Administrative
Frequency: Annual
Custodian: AIHW

The AIHW manages the national collection of child protection administrative information from Australian states and territories (who maintain responsibility for collection of the information). While each state and territory has different legislation, policies and practices surrounding statutory child protection processes and information collected are similar. However, AIHW acknowledges that definitions and thresholds, and therefore data, are not always comparable. Departments receive notifications or reports that are investigated for sufficient evidence. If sufficient, the notification is 'substantiated'. There is a range of actions a department may take when a report is substantiated, such as make an application to the court for a care and protection order or placing the child into out-of-home care. The information collected is focused on the child with limited information collected about perpetrators. In addition, there is no formal definition or flag for family, domestic and sexual violence. Data are published by the AIHW in Child Protection Australia (AIHW, 2012a). However family and domestic violence information is not currently included in this publication.

Data type: Administrative Census
Frequency: Annual
Custodian: AIHW

The SHSC is an online data collection, storage and validation system that collects information monthly from organisations that are funded under the National Affordable Housing Agreement (NAHA) and the National Partnership Agreement on Homelessness (NPAH). SHSC clients are identified as experiencing domestic and family violence if they reported domestic and family violence as a reason they sought assistance, or they required domestic or family violence assistance during any support period. The information collected is primarily about the victims of family and domestic violence, although there are some details about the secondary victims. However, perpetrators who may be using the service are not identifiable. Data are published by the AIHW in Specialist Homelessness Services (AIHW, 2012b).

Data type: Administrative
Frequency: Ad Hoc
Custodian: Victorian Department of Justice

The NCIS is an internet based data storage and retrieval system that sits outside state and territory Coroners Court information systems. The main purpose of this system is to collect administrative by-product data that provides information about the frequency and nature of deaths reported to Australian and New Zealand Coroners. The information collected is primarily about the deceased, although there may be some details about an offender if relevant to the death. Family and domestic violence is not defined in the database although possible cases of interest can be identified by searching for cases that display family and domestic violence characteristics.¹³ The data held in this system inform the National Homicide Monitoring Program (NHMP) managed by the AIC.

¹³ For more information about NCIS refer to http://www.ncis.org.au

Data type: Administrative
Frequency: Biannual
Custodian: AIC

The NHMP presents information about the nature and context of homicides occurring in Australia in a given time period. To achieve this, information from the NCIS and from offence records from state and territory police is used to create a picture of the incident, victim and offender. To establish if a death occurs as part of a family and domestic event, the relationship between the victim and offender is used. The use of relationship to determine if an event is family and domestic related includes most familial relationships including step-parents/children and extended family members, such as cousins and grandparents (AIC, 2013).

Data type: Administrative
Frequency: Not published
Custodian: AGD

This system collects information about individuals who access Family Violence Prevention Legal Services and Community Legal services. CLSIS captures information about its clients and, whilst it has a family and domestic violence indicator, the system is unable to distinguish if the client is the victim or the perpetrator of the violence. Currently there is no publication or release of this data that is readily available publicly.

There are a number of states that have introduced measures or initiatives that attempt to address the complexity of multiple agencies responding to those that have experienced family, domestic and sexual violence. There are variations between jurisdictions with Tasmania, Western Australia and Australian Capital Territory at relatively advanced stages of coordinating cross agency responses. Others, like New South Wales and Queensland¹⁴ are developing their strategies and have strong examples of localised integrated approaches.¹⁵

¹⁴ For more information about Queensland's strategy and local examples refer to https://www.csyw.qld.gov.au/violence-prevention 
¹⁵ For more information about integrated approaches refer to http://www.alrc.gov.au/publications/29.%20Integrated%20Responses/integrated-responses-context-family-violence#_ftn22

The Victorian Family Violence Database (VFVD) requires a multitude of agencies to share their data holdings with the Department of Justice who then compile and release the information as a consolidated report. Currently it consists of a series of five reports (established in 2000) that provide trend analysis of Victorian cross-sector, whole-of-government and family violence incident reports. The VFVD brings together data from a range of agencies working in the area of family violence with more agencies be included in each addition (currently nine agencies contribute data). These publications are unique to Victoria and are evidence of the need for an integrated approach to data dissemination for family, domestic and sexual violence.

While the VFVD is a comprehensive and useful project, especially in its ability to provide information at the sub-state level it focuses on family and domestic violence only. It excludes all other types (especially if the perpetrator is unknown to the victim) and as a result does not include all types of sexual violence. The VFVD does not receive ongoing funding and as a result the reports are not published on a rolling cycle but in response to government impetus.

‘Safe at Home’ is a multi-agency response to family and domestic violence initiated by the Tasmanian Government in 2004. It is an integrated criminal justice response and service delivery system that places the victim’s safety at the centre of the model and includes a range of services. The Department of Justice is responsible for ensuring that Safe at Home services are able to provide an integrated approach through the facilitation of weekly Integrated Case Coordination (ICC) meetings. The ICC meetings consider family violence cases and attendees include the Victim Safety Response Team, Police Prosecutions, Family Violence Counselling and Support Service, Court Support and Liaison Service, Child Protection and Defendant Health Liaison Service. Together these agencies work to manage risk and safety and ensure that case management is coordinated appropriately.

The Safe at Home Information Management System (SIMS) is used by ICC agencies to record and share case management information. The SIMS database builds on family violence incident reports recorded in the Police Family Violence Management System (FVMS) by undertaking a relational search for victims and offenders and importing incidents into a case record. Phase two of SIMS development will automate the importation of data from other Department of Justice databases including CRIMES and CIS.

The Family Violence Counselling and Support Service FVCSS (Adult Program and Children and Young Persons Program) are delivered by the Department of Health and Human Services. As part of the case management of family violence cases FVCSS workers input information in SIMS. A range of NGOs also respond to family violence victims, often providing longer-term counselling support, women’s refuges and/or legal assistance. The data collected by these organisations are not entered on SIMS, but supplied to the funding body as part of the contractual agreement.

The ACT Family Violence Intervention Program (FVIP) is a proactive, multi-agency approach to family and domestic violence police incidents that proceed to prosecution. The ACT has a pro-arrest, pro-charge policy on domestic and family violence and it is the role of the FVIP to integrate police, prosecution, courts and corrections activities as well as coordinating work with other key agencies in the community.

To support state strategic planning for family and domestic violence the Senior Officers’ Group (SOG) was established in 2006. The SOG membership is composed of agencies involved in responding to family and domestic violence and includes Commonwealth and state government agencies and representation from the community sector. The SOG is responsible for development, implementation and monitoring of across government policy and programs including the WA Strategic Plan for Family and Domestic Violence 2009-2013 and the Family and Domestic Violence Prevention Strategy 2012-2022 (Department for Child Protection (DCP), 2009).

Significant initiatives to support integrated service responses to family and domestic violence that have been led by the SOG include: the Memorandum of Understanding to support information exchange; guidelines for multi-agency case management of high risk family and domestic violence cases; and implementation of ‘Family and Domestic Violence Response Teams’ (FDVRT). The FDVRT is a partnership between WA Police, the Department for Child Protection and Family Support (DCPFS) and community sector agencies. Their role is to provide joint assessment, triage and service responses for people experiencing family and domestic violence where there has been a police call out.

The CRAF is designed to be a tool that works within the Integrated Family Violence System (IFVS) in Victoria and is reliant on Victorian agencies building effective partnerships and networks.¹⁸ The CRAF guides practitioner responses and ensures that a standardised approach is used by all agencies to enable consistent service provision for all victims. The CRAF provides three practice guides tailored to meet the different requirements of practitioners that assist in identifying family violence, preliminary assessment and comprehensive assessment (Department of Human Service (DHS) Victoria, 2007). The Victoria Police Risk Assessment and Risk Management Report (VP FORM L17) is consistent with conducting a preliminary assessment (Victoria Police, 2010). The CRAF is now widely used in Victoria and has been essential in establishing referral pathways that support the IFVS.

¹⁸For full details of the recommendations, see Australian Law Reform Commission 2010, Family Violence - A National Legal Response (ALRC Report 114), ALRC, Sydney. http://www.alrc.gov.au/publications/29.%20Integrated%20Responses/integrated-responses-context-family-violence

Western Australia has developed the CRARMF based on the Victorian CRAF to provide a standardised approach to identifying, assessing and responding to family and domestic violence for all agencies in WA. It is designed to facilitate a consistent, collaborative and integrated response through establishing minimum standards that guide identification, screening, response, referrals, risk assessment and management (DCP, 2011). The development of Domestic Violence Incident Report (DVIR 1-9) is the WA police approach to identifying family and domestic violence incidents. It was developed at a similar time to the CRARMF and aims to provide information for the purpose of risk assessment. The CRARMF fits into the state wide integrated approach to family and domestic violence as the tool used by agencies to refer individuals to the Coordinated Response Service (Department for Child Protection and Family Support, 2013).

There are two risk assessments used in South Australia for family and domestic violence, the South Australia Police (SAPOL) Risk Assessment Form and the Family Safety Framework’s (FSF) Risk Assessment Form which is based on the former. The FSF form is a practical tool that is used to indicate risk. It is not a risk assessment; instead it is used by agencies to consistently identify clients that should be referred to the Family Safety Meetings (Office for Women, 2013).

As part of the Safe at Home initiative (see National Datasets – Appendix 3 for more information) the Tasmania Police and the Department of Justice developed the RAST which is used by operational police when attending a family violence incident. The RAST has been implemented to ensure consistency in the approach and assessment of offenders in relation to their risk of repeating the violence. The tool is used to classify perpetrators risk of reoffending on a scale so that police can make informed decisions about actions that need to be taken (Tasmanian Institute of Law Enforcement Studies (TILES), 2009).

At the time of publication NSW police were in the process of developing and implementing the PRIT

Australian Bureau of Statistics (ABS) 2013a, Crime Victimisation, Australia, 2011-12, cat. no. 4530.0, ABS, Canberra.

ABS, 2013b, Defining the data challenge for family, domestic and sexual violence, 2013, cat. no. 4529.0, ABS, Canberra.

ABS, 2013c, Criminal Courts, Australia, 2011-12, cat. no. 4513.0, ABS, Canberra.

ABS, 2013e, Recorded Crime - Offenders, 2011-12, cat. no. 4519.0, ABS, Canberra.

ABS, 2012, Recorded Crime - Victims, Australia, 2011, cat. no. 4510.0, ABS, Canberra.

ABS, 2011a, Directory of Family and Domestic Violence Statistics, 2011, cat. no. 4533.0, ABS, Canberra.

ABS, 2011b, Measuring Victims of Crime: A Guide to Using Administrative and Survey data, cat.no. 4500.0.55.001, ABS, Canberra.

ABS, 2009, Information Paper: Conceptual Framework for Family and Domestic Violence, 2009, cat. no. 4529.0, ABS, Canberra.

ABS, 2006, Personal Safety Survey, 2005, cat. no. 4906.0, ABS, Canberra.

ABS, 2003, Sexual Assault Information Development Framework, 2003, cat. no. 4518.0, ABS, Canberra.

Australian Centre for the Study of Sexual Assault (ACSSA) 2012, The nature and extent of sexual assault and abuse in Australia, Australian Institute of Family Studies (AIFS), Melbourne.

Australian Domestic and Family Violence Clearinghouse (ADFVC), The Gender Debate in Domestic Violence: the Role of Data, The University of New South Wales, Sydney.

Australian Federal Police (AFP), 2011, AFP Practical Guide on family violence incidents (ACT Policing). [Online]

Australian Institute of Criminology (AIC) 2013, Homicide in Australia: 2008-09 to 2009-10 National Homicide Monitoring Program annual report, AIC, Canberra.

AIC, 2011, Children’s exposure to domestic violence in Australia, AIC, Canberra.

Australian Institute of Health and Welfare (AIHW), 2013, Child Protection Australia: 2011-12. Child Welfare series no. 55. Cat. no. CWS 43, AIHW, Canberra.

AIHW, 2012a, Housing assistance in Australia, AIHW, Canberra.

AIHW, 2012b, Specialist Homelessness Services 2011-12. Cat. no. HOU 267, AIHW, Canberra.

AIHW, 2012c, Australian Hospital Statistics: 2010-11. Health Services Series no.43. Cat.no. HSE 117, AIHW, Canberra.

AIHW, 2012d, Australian Hospital Statistics 2011-12: Emergency Department Care. Health Services Series no.45. Cat. no. HSE 126, AIHW, Canberra.

Australian Law Reform Commission (ALRC) 2010, Family Violence - A National Legal Response. Final Report, Australian Government, Canberra.

Chung, D 2013, Understanding the Statistics about Male Violence Against Women, White Ribbon Research Series, Sydney.

Coroners Court of Victoria, 2011, Victoria's Coronial Model for Investigating Family Violence Related Death, Coroners Court of Victoria Australia, Melbourne.

Coroners Court of Victoria, 2012, Victorian Systemic Review of Family Violence Deaths, First Report, Coroners Court of Victoria, Melbourne.

Department for Child Protection and Family Support (DCPFS) 2013, Western Australia’s Family and Domestic Violence Prevention Strategy to 2022: Crating safer communities, Perth: Western Australian Government.

Department for Child Protection (DCP) 2011, The Western Australian Family and Domestic Violence Common Risk Assessment and Risk Management Framework, Perth: Western Australian Government.

DCP, 2009, WA Strategic Plan for Family and Domestic Violence 2009-2013, Perth: Western Australian Government.

Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) 2012, The National Implementation Plan for the First Action Plan 2010-2013, Building a Strong Foundation, FaHCSIA, Canberra.

FaHCSIA, 2011, The National Plan to Reduce Violence against Women and their Children: Including the first three-year Action Plan, FaHCSIA, Canberra.

FaHCSIA, 2009a, The Cost of Violence against Women and their Children, March 2009, FaHCSIA, Canberra.

FaHCSIA, 2009b, Time for Action: The National Council’s Plan for Australia to Reduce Violence against Women and their Children, 2009-2021, FaHCSIA, Canberra.

Department of Human Service (DHS) Victoria, 2007, Family Violence Risk Assessment and Risk Management Framework and Practice Guides 1-3, DHS, Melbourne.

Department of Justice Tasmania, 2009, Review of the Integrated Response to Family Violence: Final Report, Successworks, Hobart.

Northern Territory Government, 2012, Child Safety and Wellbeing Plan 2012. Key strategies and actions, NT Government, Darwin.

Northern Territory of Australia, 2012, Domestic and Family Violence Act, NT Government, Darwin.

Office for Women, 2013, Family Safety Framework: Practice Manual. Version 4, Government of South Australia, Adelaide.

Pease, B & Flood, M, 2006, The factors influencing community attitudes to violence against women; a critical review of the literature, Victorian Health Promotion Foundation, Melbourne.

Robinson, E and Moloney, L 2010, Family violence: Towards a holistic approach to screening and risk assessment in family support services, Australian Family Relationships Clearinghouse, Melbourne.

Ruuskanen, E & Aromaa, K 2008, Administrative data collection on domestic violence in Council of Europe member states, Council of Europe, Strasbourg.

Tasmanian Institute of Law Enforcement Studies (TILES) 2009, Analysis of the Tasmania Police Risk Assessment Screening Tool (RAST); Final Report, TILES, Hobart.

VicHeath, 2010, National Survey on Community Attitudes to Violence Against Women 2009. Changing cultures, changing attitudes - preventing violence against women. A summary of findings, VicHealth, Carlton.

Victoria Police, 2010, Code of Practice for the Investigation of Family Violence. 2nd Edition, Victoria Police, Melbourne.