The Australian Institute of Health and Welfare has been responsible for collecting the national child protection data since the early 1990’s. The first national data published by the Institute were 1990-91 data which were published in 1993 and the latest data published this year are the 1998-99 data. For a number of years prior to 1990-91, national data on child protection was collected by WELSTAT (The Standardisation of Social Welfare Statistics Project).

The three areas of the child protection system for which national data are currently collected are:

• child protection notifications, investigations and substantiations;
• children on care and protection orders; and
• children in out-of-home care (collected since 1995-96).

There are no data at the national level on children and their families who are referred to or who access other services for protective reasons.

This paper analyses the last 9 years of national data on child proteciton notifications, investigations and substantiations to see what it can tell us about child protection in Australia. Trends in children on care and protection orders and children in out-of-home care will also be examined in a separate report to be released by the Institute later this year.

It is argued that the national child protection data reflect the policies and practices of community service departments in relation to child protection, and the impact of signficant policy changes that have occurred over the last decade is evident in the data on trends.

Child protection is the responsibility of the community services department in each State and Territory. Children who come into contact with the community services department for protective reasons include those:

• who have been or are being abused or neglected; and/or
• whose parents cannot provide adequate care or protection.

The community services department provides assistance to these children and their families through the provision of, or referral to, a wide range of services. Some of these services are targeted specifically at children in need of protection (and their families), whereas others are available to a wider section of the population and attempt to deal with a broad range of issues or problems.

The processes used to protect children are broadly similar across jurisdictions, although they are based on different legislation and policies. Figure 1 illustrates a simplified version of the main child protection processes and these are outlined in more detail below. The shaded boxes in the table represent those items for which data are collected at the national level.

Children who are seen to be in need of protection can come into contact with the community services department through a number of avenues. These include reports of concerns about a child made by someone in the community, by a professional mandated to report suspected abuse and neglect, or by an organisation that has contact with the family or child. The child, his or her parent(s) or another relative may also contact the department seeking assistance. These reports may relate to abuse and neglect or to broader family concerns such as economic problems or social isolation. There are no national data on the total number of reports made to community services departments relating to concerns about children.

Reports to the department are assessed to determine if the matter they relate to should be dealt with by the community services department or referred to another agency. Those reports that are appropriate for community services departments are further assessed to determine if any further action is required.

Reports requiring further action will generally be classified as either a family support issue or a child protection notification, although how reports are classified varies somewhat across jurisdictions. A range of factors is taken into account by departmental officers in deciding whether a report will be classified as a child protection notification. Those reports classified as a family support issue will be further assessed and may be referred to family support services. Child protection notifications are dealt with through a separate process.

After a report has been classified as a notification, it is assessed by the community services department to determine whether it requires an investigation; whether it should be dealt with by other means, such as referral to other organisations or to family support services; or whether no further protective action is necessary or possible. An investigation is the process whereby the community services department obtains more detailed information about a child who is the subject of a notification and makes an assessment of the degree of harm or risk of harm for the child. After an investigation is completed, a notification is will either be ‘substantiated’ or ‘not substantiated’.

A notification will be substantiated where it is concluded after investigation that the child has been, is being or is likely to be abused or neglected or otherwise harmed. States and Territories differ somewhat in what they actually substantiate. In some jurisdictions the substantiation process will focus on an act or an incident of child abuse or neglect, while in others the focus is on the harm or risk of harm to the child.

In Tasmania the category ‘child at risk’ is also used. This refers to situations where the notification is not substantiated, but where there are reasonable grounds for suspecting the possibility of previous or future abuse or neglect and it is considered that continued departmental involvement is warranted.

Prior to 1998-99 child protection notifications, investigations and substantiations were referred to as notifications, investigations and substantiations of child abuse and neglect. The new terms were adopted in recognition of the fact that what was being reported on nationally in the ‘child abuse and neglect’ data collection, actually included a broad range of incidents that would not be classified as ‘child abuse and neglect’ in all jurisdictions.

At any point in the child protection process the community services department has the authority to apply to the relevant Court to place the child on a care and protection order. Recourse to the Court is usually a last resort and is used in situations where supervision and counselling are resisted by the family, where other avenues for the resolution of the situation have been exhausted, or where removal of a child from home into out-of-home care requires legal authorisation. In some jurisdictions, for example, all children who are placed in out-of-home care must be on an order of some kind.

Children can also be placed on a care and protection order and/or in out-of-home care outside of the substantiation process, for example in situations where family conflict is such that ‘time out’ is needed, or where the parents are ill or unable to care for the child.

Differences between jurisdictions

The national child protection data are extracted from the administrative systems of the State and Territory community services departments according to definitions and counting rules agreed to by the departments and the Institute. As each jurisdiction has its own legislation, policies and practices, the child protection data they provide reflect some important variations in the what jurisdictions do and in how they count child protection matters. The data from different jurisdictions therefore should not be directly compared.

The differences between jurisdictions are apparent in the variation in the rates of children aged 0-16 years in child protection notifications, investigations and substantiations (Figure 2). In relation to notifications in 1998-99, the rate of children notified ranged from 25.3 per 1,000 children in South Australia to 5.1 per 1,000 in Western Australia and 3.5 per 1,000 in Tasmania.

There was also a considerable range in the rate of children in investigations across jurisdictions, ranging from 15.8 per 1,000 children in the Australian Capital Territory to 3.1 per 1,000 in Tasmania. The range in the rate of children in substantiations was not as great, though there was still considerable variation. The rate of children in substantiations ranged from 6.3 per 1,000 children in Victoria to 1.1 per 1,000 children in Tasmania.

Rather than indicating differences in the number of children in need of protection across jurisdictions, it is likely that most of the variation across jurisdictions is due to differences in what States and Territories do in the child protection area. One of the main differences between jurisdictions is the extent to which reports to departments are responded to outside the formal child protection system.

For example in Western Australia and Tasmania who have the lowest rates of children in notifificantions, investigations and substantiations, a high proportion of reports to the department receive a differential response and are not counted in the national child protection data. This response does not involve an investigation, but may involve ongoing assessment or the provision of family support services. Only those reports that relate to child maltreatment are counted as a notification and included in the national data.

While most other jurisdictions have also introduced non-investigatory options for responses to reports, they respond to a higher proportion of reports through the formal child protection system. Victoria for example, which has relatively high rates of notifications, investigations and substantiations, appears to have both a broad intake and to substantiate a broader range of cases than Western Australia or Tasmania.

This means that some cases that are classified as a substantiation in Victoria would be classified as a family support issue in Western Australia. An example of this would be in situations where a parent has an drug problem and requires some kind of additional assistance, but where maltreatment of the child has not occurred. Depending on the particular circumstances of the case, this case is more likely to be counted as a notification in Victoria, while in Western Australia it would likely to be classified as a family support issue. The type of service provided to the family in both jurisdictions, however, is likely to be very similar.

There are other differences in what is classified as a child protection notification and what is substantiated that are also worth noting.

• Some jurisdictions, such as New South Wales include abuse by a stranger as a child protection matter while most others include only abuse or neglect within the family or where the parents are unwilling or unable to protect the child.
• What is substantiated varies: some jurisdictions substantiate harm, some the risk of harm, and others an action or an incident.

More detail on differences between jurisdictions is provided in the report ‘Comparability of child protection data’ (AIHW 1999b).

National trends

The Institute has national data on the number of child protection investigations and substantiations for the years from 1990-91 to 1998-99, and on the number of notifications for the years from 1995-96 to 1998-99. While data on the number of notifications are only available for a limited period, they show a trend of increasing numbers of notifications between 1995-96 and 1998-99 (Figure 3). Across Australia there were 91,734 notifications in 1995-96, while in 1998-99 there were 103,980 notifications, an increase of 13%. It is likely that the number of notifications also increased in in the first half of the 1990’s, as there were large increases in the number of investigations over this period.

Possible reasons for the increase in the number of notifications include:

• increases in the number of children who require a child protection response for example due to an increase in the incidence of child abuse and neglect or inadequate parenting causing harm to a child;
• increased reporting by professionals as a result of the mandatory reporting provisions in most jurisdictions;
• increased awareness in the community about child abuse and neglect and the role of community service departments in this area.

It is likely that the number of notifications also increased in the first half of the 1990’s as there were large increases in the number of investigations in this period. The number of investigations increased by 55% from 49,721 in 1990-91 to 76,954 in 1994-95. From 1994-95 to 1998-99, however, the number of investigations decreased by 24% to reach 58,717 in 1998-99.

The national trend in substantiations during the 1990’s was similar to the trend in investigations, that is one of increasing numbers of substantiations up to 1994-95, followed by a period of decreasing numbers. There were 20,868 substantiations in 1990-91, increasing by 47% to a peak of 30,615 in 1994-95, and then decreasing by 17% to 25,447 in 1998-99.

These national trends, however, mask the different trends that have occurred in each State and Territory over the last decade. Increases in numbers of children in the child protection system in one jurisdiction can cancel out decreases that occur in another, so that what has occurred in each jurisdiction can vary significantly from the national trends.

Trends in child protection also need to be interpreted carefully because changes in legislation or policies can have a direct and immediate effect on the numbers of notifications, investigations or substantiations reported. Large increases or decreases in the numbers of children in the child protection system are generally due to such changes rather than to changes in the number of children in the community who are in need of child protection response.

In particular, in most jurisdictions policies have been introduced over the last decade that allow for non-investigative responses to reports of concerns about children that are made to community service departments. These policies have been introduced at different times in different jurisidictions, but in all cases they have lead to substantial decreases in the number of investigations and substantiations. Other policy changes can also impact on the numbers in different ways.

Notifications in New South Wales increased in the period between 1995-96 and 1998-99 by 2,583 or 9% (Figure 4). The number of investigations increased by 62% in the early 1990’s from 20,646 in 1990-91 to 33,448 in 1994-95. Over the next four years the number of investigations decreased by 41% to 19,798 in 1998-99. The number of substantiations peaked one year earlier rising by 30% from 11,611 in 1990-91 to 15,128 in 1993-94 and then falling by 50% to 7,540 in 1998-99.

The introduction of new child protection policies in July 1996 is likely to have been the main reason for the considerable fall in the number of investigations and substantiations that occurred after this date. Prior to July 1996 there was a policy to investigate all notifications made to the community service department. From July 1996 onwards, notifications were initially assessed to determine whether they related to child abuse or neglect or to some other concern about children. Only those relating to child abuse and neglect were investigated (AIHW 1998). The impact of this change is evident in the data available after 1995-96 when the proportion of notifications that were invesgated decreased considerably.

At the same time changes were made to the concept of a substantiation. Before the introduction of the new policies, substantiation of a notification did not necessarily mean that child abuse and neglect had occurred, but rather that information about the notification was confirmed. This was changed so that a notification would only be substantiated where there was child abuse or neglect or where the child was regarded as being at risk of abuse or neglect. The fairly substantial fall in the number of substantiations after July 1996 is related to this change in policy.

There was a greater increase in the number of notifications in Victoria than in New South Wales between 1995-96 and 1998-99. The number of notifications increased from 29,914 to 34,679, or by 16% (Figure 5).

The trend for investigations and substantiations over the 1990’s was one of increasing numbers up to and including 1994-95, followed by a period of relative stability. The number of investigations increased from 8,397 in 1990-91 to 15,618 or by 85%. The number then fluctuated between 14,000 and 15,000 up until 1997-98 and fell to 13,708 in 1998-99. Substantiations followed a similar pattern increasing by over 200% from 2,427 in 1990-91 to 7,326 in 1994-95, and then fluctuating around 6,000 to 7,000 in the period to 1998-99.

Amendments to the Child and Young Persons Act 1989 are likely to have contributed to the increase in investigations and substantiations in the early 1990’s. In 1992 the Victorian Police ceased their involvement in receiving notifications and the Department of Human Services was given statuatory responsibility to protect all children and young people aged under 17 years from child abuse and neglect (Angus and Hall 1996). In addition, the introduction of mandatory reporting, which was phased in from late 1993 to 1994, may have lead to increasing numbers of investigations.

There were a number of initiatives introduced in Victoria in 1998-99 which are likely to lead to decreases in the number of notifications, investigations and substantiations. These changes were introduced in response to concerns that families in need were being subjected to unnecessary, or unnecessarily intrusive, investigation. The new initiatives aim to improve the appropriateness of responses to child protection notifications by basing decision making and practice on a broad perspective and assessment framework, with more flexible response options which maximise the potential for developing partnership with families and collaboration with other agencies (AIHW 2000). The impact of these new initiatives are likely to be seen on the data over the next few years.

In Queensland the trend for notifications was also one of increasing numbers. Notifications increased from 15,362 in 1995-96 to 18,721 in 1998-99, an increase of 22% (Figure 6). The trend for investigations was one period of decrease in the early 1990’s and then increases in numbers each year up until 1997-98. The number of investigations fell by 25% from 11,390 in 1990-91 to 8,496 in 1992-93, and then increased by 79% to 15,245 in 1997-98. The number fell slightly to 15,010 in 1998-99.

The number of substantiations followed a smilar pattern to the number of investigations, but the overall trend was one of increasing numbers. In 1990-91 there were 3,500 substantiations. The number of substantiations fell to 2,743 in 1992-93, then increased to 4,000 by 1994-95, fell to 3,332 in 1995-96 before increasing again to 6,373 reach in 1998-99.

The fall in both investigations and substantiations in 1992-93 occurred following the introduction of a new framework for managing child protection matters introduced in 1992. The three tenets of this framework were minimal intervention by the department, on-going assessment of the child’s protective needs and participation by families in decision making processes (Elliott and Sultmann 1998).

Earlier this year the Child Protection Act 1999 was proclaimed. A significant feature of the Act is the concept of protecting children and young people from harm, rather than from child abuse and neglect thus bringing the legislation into line with current practice. The legislation has provided the impetus for broad-based reform of the child protection system (Testro 1998). In particular funding has been provided to increase the number of child protection workers as well as for Indigenous child protection services. These reforms may have impacted on numbers in 1998-99.

The trends in Western Australia reflect the major new policy direction taken in the mid 1990’s. Under ‘New Directions’, which was phased in during 1995 and 1996, reports of concerns about children and their families were distinguished from concerns about maltreatment of children. Where there was no indication of maltreatment, reports were dealt with separately and not classified as a notification. A pilot of the new arrangements indicated that a significant proportion of reports would be classified as child concern reports rather than a child maltreatment allegation that is counted as a notification in the national data. A new definition of maltreatment that focused on harm to a child rather than an action was also introduced.

After the introductions of ‘New Directions’ the number of notifications declined from 3,748 in 1995-96 to 2,099 in 1996-97, a decrease of 44%, although a significant decrease in the number of notifications would have occurred one year earlier (Figure 7). The number of notifications increased in 1997-98 and 1998-99. The fall in the number of investigations was more dramatic. Between 1993-94 and 1996-97, the number of investigations decreased by 4,676 or by over 70%. The number of investigations has also increased since then.

There was also a substantial decrease in the number of substantiations between 1993-94 and 1996-97 after the introduction of the new policies. Substantiations decreased from 1, 830 in 1993-94 to 945 in 1996-97, or by 48%. The number of substantiations then increased by 29% to reach 1,215 in 1998-99.

The number of notifications rose fairly steeply in South Australia between 1995-96 and 1998-99, increasing from 8,895 to 13,132, or by 48% (Figure 8). One of the factors contributing to this increase was a change in the processes used to count notifications. In 1996-97 the range of reports included in notifications was broadened to include a new category called ‘notifier concerns’. These are reports made by mandated professionals, but where insufficient information is provided to undertake an investigation. This change would therefore not have impacted on the number of investigations or substantiations.

The number of investigations rose during the 1990’s and peaked at 7,323 in 1996-97. The number then fell by 34% to 4,841 in 1997-98, and increased slightly to 5,198 in 1998-99. The number of substantiations peaked two years earlier at 2,547 in 1994-95 and then fell by 19% to 2,072 in 1996-97. It has remained relatively steady since then.

A new child protection system was introduced in South Australia in 1997. Its features include a central intake system to classify all reports and a differential response system based on risk that also provides the option of a non-investigative response. Those involved in cases assessed as low risk but with significant need are invited to attend a family meeting (AIHW 1999). It is likely that these changes were responsible for the marked fall in investigations that occurred between 1996-97 and 1997-98.

The trends in Tasmania were both unique and dramatic. The number of notifications dropped markedly over the period in which data were collected decreasing from 2,933 in 1995-96 to 653 in 1998-99, a decrease of 78% (Figure 9). The number of investigations increased gradually in the early 1990’s but then dropped by some 68% between 1996-97 and 1997-98. Substantiations in Tasmania started to fall gradually after 1991-92, but they also fell sharply between 1996-97 and 1997-98 from 223 to 135.

The fairly dramatic decreases in the number of notifications, investigations and substantiations in 1997-98 occurred after Tasmania introduced a model in July 1997 that was similar to the Western Australian model. Under the new system reports to the department are screened by senior staff and where there is no indication of maltreatment, they are classified as a child and family concern report. Training for the new system began in early 1997 and this is likely to also have had some impact on the count of notifications in 1996-97. In relation to substantiations, it appears that Tasmania is currently only reporting the most serious forms of harm or risk, though it is not clear why the number of substantiations has dropped so much.

In the period between 1995-96 to 1997-98 the number of notifications in the Australian Capital Territory decreased by 22% from 1,437 to 1,125 (Figure 10). The number then increased by 21% to 1,358 in 1998-99. The number of investigations fluctuated over the 1990’s. The lowest point was in 1990-91 when there were 778 investigations, and the highest point was in 1993-94 when there were 1,208 investigations. Similar to the fairly large increase in notifications over the period, the number of investigations increased by some 31% between 1997-98 and 1998-99.

The fall in notifications and investigations in both 1996-97 and 1997-98 is linked to changes to the way that notifications were counted. From July 1997 child concern reports were separated out from child protection notifications. Before the formal introduction of this policy, informal moves were made to separate out child concern reports, impacting on the 1996-97 data.

The number of substantiations also fluctuated over the decade, ranging from 247 in 1990-91 to 495 in 1993-94. There was a small increase in the number of substantiations between 1997-98 and 1998-99.

A new child protection act was passed by the Australian Capital Territory government in 2000. Changes to child protection policies include changes to the way report are screened to include an initial danger assessment screening device followed, where necessary, by a full danger assessment. These new policies may impact on numbers over the next few years.

The numbers of notifications and investigations each year are almost the same in the Northern Territory due to a policy to investigate all child protection notifications (Figure 11). There were two periods of increase in the number of investigations during the 1990’s. In the years from 1991-92 to 1994-95, investigations increased from 427 to 773. The number then fell for the next two years to reach 473 in 1996-97, and increased to 684 in 1997-98. Data are not available for the full 1998-99 financial year as the Territory introduced a new recording system.

The trend in substantiations was very similar to the trend in investigations, with two periods of increase. Substantiations increased by 67% from 226 in 1991-92 to 377 in 1994-95, and then fell over the next two years. There was also a large increase in the number of substantiations between 1996-97 and 1997-98 when they rose by over 40% from 240 to 343.

Indigenous children

Differences between jurisdictions

There was a wide range in rates of Indigenous children in notifications, investigation and substantiations across jurisdictions. In relation to notifications, this rate ranged from 1.6 per 1,000 in Tasmania to 113.2 per 1,000 children in Victoria. For substantiations, this rate ranged from 1.1 per 1,000 in Tasmania to 26.8 per 1,000 in South Australia. Some of the difference in rates across jurisidictions may be due to differences in the number of Indigenous children who require a child protection response. Much of the difference, however, is likely to be due to policy differences described earlier as well as to differences in as well as differences in the way that Indigenous status is identified. There were particular problems with the quality of Indigenous data provided by Victoria for 1998-99, so that the 1997-98 data is shown here.

In all jurisdictions except Tasmania, the rate of Indigenous children in notifications, investigations and substantiations was substantially higher than the rate for other children (Table 1). The ratio of the rate of Indigenous children to the rate of other children in notifications, investigations and substantiations provides a summary measure of the extent to which Indigenous children are over-represented. This rate ratio also shows that as the level of intervention by the community services department increases (that is moving through the child protection process from notification to substantiation) so does the extent to which Indigenous children are over-represented.

1. New South Wales was not able to provide data for 1996-97.
2. Queensland could not provide data on the number of substantiations for 1993-94.
3. Victoria could not provide data on Indigenous children in 1998-99.
4. The Northern Territory could not provide full year data for 1998-99.

Angus G & Hall G 1996. Child abuse and neglect Australia 1994-95. Child Welfare Series No. 16. Canberra AIHW.

Angus G & Wilkinson K 1993. Child abuse and neglect Australia 1990-91. Child Welfare Series No. 2. Canberra AGPS.

Angus G, Wilkinson K & Zabar P 1994. Child abuse and neglect Australia 1991-92. Child Welfare Series No. 5. Canberra AGPS.

Angus G & Woodward S 1995. Child abuse and neglect Australia 1993-94. Child Welfare Series No. 13. Canberra AGPS.

Angus G & Zabar P 1995. Child abuse and neglect Australia 1992-93. Child Welfare Series No. 9. Canberra AGPS.

Australian Institute of Health and Welfare (AIHW) 1998. Child protection Australia 1996-97. Canberra: AIHW.

Australian Institute of Health and Welfare (AIHW) 1999a. Child protection Australia
1997-98. Canberra: AIHW.

Australian Institute of Health and Welfare 1999b. Comparability of child protection data. Canberra: AIHW.

Australian Institute of Health and Welfare (AIHW) 2000. Child protection Australia 1998-99. Canberra: AIHW.

Broadbent A & Bentley R 1997. Child abuse and neglect Australia 1995-96. Canberra: AIHW.

Elliott A and Sultmann C 1998. Principles and processes for child protection decision making, Queensland’s case management framework. Children Australia Volume 23, No.4, 1998.

Robinson 1997. The Court’s Position. Paper presented at Australasian Conference on Child abuse and neglect, Adelaide 20-23 October 1993.

Testro, Paul 1998. Editorial. Children Australia Volume 23, No.4, 1998.