Mandatory Reporting Laws and Identification of Child Abuse and Neglect: Consideration of Differential Maltreatment Types, and a Cross-jurisdictional Analysis of Child Sexual Abuse Reports

Mandatory reporting laws have been created in many jurisdictions as a way of identifying cases of severe child maltreatment on the basis that cases will otherwise remain hidden. These laws usually apply to all four maltreatment types. Other jurisdictions have narrower approaches supplemented by differential response systems, and others still have chosen not to enact mandatory reporting laws for any type of maltreatment. In scholarly research and normative debates about mandatory reporting laws and their effects, the four major forms of child maltreatment—physical abuse, sexual abuse, emotional abuse, and neglect—are often grouped together as if they are homogenous in nature, cause, and consequence. Yet, the heterogeneity of maltreatment types, and different reporting practices regarding them, must be acknowledged and explored when considering what legal and policy frameworks are best suited to identify and respond to cases. A related question which is often conjectured upon but seldom empirically explored, is whether reporting laws make a difference in case identification. This article first considers different types of child abuse and neglect, before exploring the nature and operation of mandatory reporting laws in different contexts. It then posits a differentiation thesis, arguing that different patterns of reporting between both reporter groups and maltreatment types must be acknowledged and analysed, and should inform discussions and assessments of optimal approaches in law, policy and practice. Finally, to contribute to the evidence base required to inform discussion, this article conducts an empirical cross-jurisdictional comparison of the reporting and identification of child sexual abuse in jurisdictions with and without 461 mandatory reporting, and concludes that mandatory reporting laws appear to be associated with better case identification.


Introduction
Child maltreatment presents challenges for societies in the domains of public health, human rights, criminal justice, social justice, community welfare and economics.Policy discussions and research concerning the identification and response to maltreatment often treat the four classical forms-physical abuse, sexual abuse, emotional abuse, and neglect-together as if they are homogenous.Yet, different forms of maltreatment have different characteristics, contexts and consequences.Sexual abuse, and severe physical battering of an infant, for example, are very different to mild "neglect" of a teenaged child produced only by poverty (such as an inability to provide some kinds of medical treatment), which in many societies would not be classed as neglect.Heterogeneity across and even within types of maltreatment, suggests some legal and policy responses to detect and respond to cases may be especially well suited to some of its forms.
Societies worldwide continue to engage with the question of how best to identify and respond to cases of child maltreatment.There are good reasons to do so, both as a matter of principle and pragmatics.In principle, early case identification and service provision are justified because the majority of child maltreatment is inflicted on helpless infants and very young children who are vulnerable and dependent on their parents and caregivers in multiple ways [1].As such, maltreatment occurs in a context of extreme power imbalance.Neonates and infants in particular are uniquely vulnerable to violence and neglect, being preverbal, immobile, and completely dependent on the caregiver for all human needs.Other child victims also occupy an unequal position, in physical, cognitive, psychological, emotional, legal and economic terms.While not all maltreated children suffer in the same way, many victims are marginalised and disadvantaged, with maltreatment causing both substantial suffering and health consequences which can endure through the lifespan and compromise functioning in both childhood and adulthood [2,3].Pragmatically, societies need to be seen as good global citizens.The United Nations Convention on the Rights of the Child 1989 requires States parties to take all appropriate legislative, administrative, social and educational measures to protect children from all forms of abuse and exploitation while in the care of parents and caregivers (article 19).Moreover, there are self-serving societal interests in developing better child maltreatment detection and response measures, since it offers the prospect of reducing cost to society caused by placing children in out-of-home care, lost productivity resulting from victims' reduced functional capacity, other costs produced by victims' injuries and compromised health, and costs of social welfare [2][3][4].
Mandatory reporting laws have been created in many jurisdictions in different forms, primarily as a way of identifying existing cases of severe maltreatment which are likely to otherwise remain hidden, so that the child's victimisation can cease, health rehabilitation and other support can be delivered (both to the child and her or his parents and family), the child's safety can be promoted, and, where necessary, perpetrators can be held accountable [5][6][7][8].These laws are always intended to be only one part of a public health response to child maltreatment, primarily forming a tertiary response to supplement secondary and primary prevention [9].The laws continue to be adopted as a case identification strategy, with recent enactments in jurisdictions as diverse as Saudi Arabia [10] and Ireland (the Children First Bill 2014).This article will first consider some of the key characteristics of different types of child abuse and neglect, and will outline the nature and operation of mandatory reporting laws.It posits a differentiation thesis, arguing that different patterns of reporting between both reporter groups and maltreatment types must be acknowledged and analysed, and should inform discussions and assessments of optimal approaches in law, policy and practice.To add to the evidence base about the effects of mandatory reporting on report numbers and case identification of a specific type of maltreatment, it then conducts an empirical cross-jurisdictional comparison of the reporting and identification of child sexual abuse in jurisdictions with and without mandatory reporting to explore whether mandatory reporting laws appear to be associated with better case identification.

Key Characteristics of Different Types of Child Abuse and Neglect
To clarify the conduct with which this discussion is concerned, it is first necessary to explain the broad conceptual nature of the discrete types of child maltreatment.Adopting the basis of a previous model [11], I will use the following general definitions of each type, which draw on the work of leading scholars, and which are generally consistent with frameworks of mandatory reporting laws as well as rights and obligations in criminal law and civil law:  physical abuse includes acts of physical assault by parents or caregivers which result in death or serious physical harm, or which present an imminent risk of doing so; it excludes lawful corporal punishment [12];  sexual abuse includes acts not only of penetrative abuse, but also acts of masturbation, oral sex, fondling, voyeurism, exposure to sexual acts, exposure to or involvement in pornography and other forms of commercial sexual exploitation, all of which are acts done to sexually gratify the abuser; it is usually inflicted by an adult, but is often and can be inflicted by another, usually older child, where the victim is not developmentally capable of understanding the acts or is not able to provide true consent [13];  psychological or emotional abuse exists when the relationship between the parent or caregiver and the child is characterized by pervasive or persistent acts or omissions which result in serious emotional harm or present an imminent risk of doing so [14,15];  neglect is constituted by omissions by parents or caregivers to provide the basic necessities of life such as food, shelter, clothing, supervision and medical care, which result in serious harm or present an imminent risk of doing so [16].
Importantly, each definition can be seen to include a qualification of severity or "seriousness" of harm caused by or likely to eventuate from the abuse or neglect.Generally, while there are some exceptions, in the context of policy discussions about mandatory reporting laws it must first be recognised that not all manifestations of harm are treated by the laws as "abuse" or "neglect", and nor were they intended to.Sexual abuse is an exception to this, as all sexual abuse is seen as seen as having such seriousness as to merit some kind of response.For the other three maltreatment types, the reporting duties are targeted at instances of abuse or neglect which have already been of sufficient severity to cause serious harm, or to involve acts which have already been committed which may not yet have caused such harm but present an imminent risk or likelihood of causing such harm.They are not targeted at trivial incidents or conditions in children's lives which could simply be seen as less than ideal.
The youngest children are most often victimised, apart from sexual abuse [1].Based on detailed statistics from industrialised Western nations, and reflecting the increased vulnerability of younger children, those aged under 4 collectively experience one third of all maltreatment, and three quarters of all victims are under 12 [1,7].There may be patterns of increased incidence of some kinds of maltreatment at transitional ages later in childhood when oppositional behaviour increases.Sexual abuse, which disproportionately affects girls, more often commences around age 9-10 [17,18].
There is heterogeneity across and within types of maltreatment, and in their consequences for individuals.Neglect, for example, has several classes: medical, nutritional, emotional, educational, and supervisory [16].All of these can have different extents and consequences.Neglect causes more fatalities than any other maltreatment type, followed by physical abuse [19].Severe neglect in infancy compromises brain growth and development and functioning through the lifespan [20].Yet, mild or even moderate neglect of a much older child is less likely to produce significant harm.Neonates (those in the first month of life) and infants and very young children are most vulnerable to fatality and very serious physical harm.Emotional abuse also takes many forms [15], and has a range of severity.Physical abuse ranges from acts causing or capable of causing death and permanently compromised functioning, to corporal punishment which moves beyond lawfully permissible discipline.Parents and caregivers inflict the vast majority of physical abuse, emotional abuse and neglect [19,21].Sexual abuse is less often inflicted by a parent although the NIS-4 still identified biological parents as the perpetrator in 36% of all cases [21], and an Australian study estimated that 1.1% of all children were sexually abused by a parent [22].Sexual abuse is usually inflicted by a person known to the child, who is often a member of the child's extended family [21][22][23][24].It can range from an isolated incident of external touching, to persistent penetrative abuse taking place regularly for years.Some children can suffer more severe consequences from sexual abuse even though the acts endured are apparently less chronic than those endured by others [25].

Consequences and Costs of Serious Child Abuse and Neglect
Different kinds of maltreatment are more likely to present different kinds of consequences for the child.Even within types, and even where the acts are of the same kind, duration and chronicity, not all individuals will experience the same type or extent of physical, social and behavioural consequences [2,25].Yet, overall, in a collective sense the range of severe consequences and the size of the costs of child maltreatment are well established and are vast.The adverse physical and mental health, behavioural, educational, social, and economic consequences of abuse and neglect are often extremely substantial [26,27].Studies have identified these consequences for distinct maltreatment types: physical abuse [12,28], sexual abuse [25,27,29], emotional abuse [30,31] and neglect [20,32,33].Maltreatment causes substantial and enduring economic costs to children who experience maltreatment [3], and to communities [4,34].Maltreatment types often co-exist, and poly-victimisation-where a child suffers several types of victimization-is especially damaging [35].Some effects are specific to or are more typical for some kinds of maltreatment, such as failure to thrive, impaired brain development, and fatality [20].However, a range of effects common to multiple kinds of maltreatment include: impaired social, emotional and behavioural development; reduced reading ability and perceptual reasoning; depression; anxiety; post-traumatic stress disorder; low self-image; physical injuries; alcohol and drug use; aggression; delinquency; long-term deficits in educational achievement; and adverse effects on employment and economic status [25][26][27][28][29][30][31][32][33].All these consequences also affect the child's dignity and capacity to develop fully as a human being; their core capabilities are compromised [36].

A Response to a Hidden Phenomenon
Mandatory reporting laws were first created to respond to the problem presented by the fact that severe child maltreatment is a largely hidden phenomenon taking place in the family sphere, which even when it comes to the attention of professionals tends to be ignored.Abuse and neglect is most frequently inflicted on infants who are pre-verbal, and other young children who cannot resist, represent themselves, resolve the situation or disclose the experience [7].In the groundbreaking article by Kempe et al. which exposed child physical abuse and doctors' failure to respond appropriately to it, severe physical abuse (head injuries and fractures) was identified as being mostly inflicted on children aged under three years of age [37].Sexual abuse is also usually undisclosed by the child and the perpetrator [38,39].Overall, children rarely disclose their own suffering, and because acts of significant maltreatment often constitute criminal conduct, or may activate child welfare proceedings of some kind, those who inflict abuse or neglect are also unlikely to reveal it to authorities which can help the child [6,7].The real annual incidence of child abuse and neglect far exceeds the number of cases which are brought to the attention of welfare agencies [21,40].

Nature of the Duty
A mandatory reporting law is a statutory duty imposed on members of a specified occupational group or groups to report suspected cases of designated types of child maltreatment to child welfare agencies.In some jurisdictions the duty is imposed on all citizens [5].Where it is placed only on occupational members, the designated occupational groups are those whose members frequently encounter children in the course of their work: common examples are police, teachers, doctors and nurses.The core premise is to oblige designated people who are well-placed to detect cases of severe child abuse and neglect to report known and suspected cases to government welfare agencies.This enables the taking of measures to ensure the child is safe, the maltreatment stops, rehabilitation can be provided, and the needs of the child and the family can be identified and supported.By using the expertise of persons outside the family who encounter the child, the hidden nature of child maltreatment can be overcome.The laws also enable reports to be made by nonmandated reporters (whether family members, friends, neighbours, or nonmandated professionals), and it is important to note that nonmandated reporters in most jurisdictions make almost half of all reports [1].
Mandatory reporting laws can be seen as one part of an overall public health response to child maltreatment which should include some measure of primary prevention at the population level, arguably the highest focus on secondary prevention of high-risk subsets of the population, and a required tertiary response to identify and assist those who have already endured severe maltreatment.Being primarily concerned with identifying cases of maltreatment that have already occurred, reporting laws are an aspect of tertiary response.In their capacity to also identify cases of serious maltreatment before they have occurred, and to prevent escalation of cases, they are also a measure of secondary response.
Because of its hidden nature, the first primary objective of these laws is to identify cases of serious child abuse and neglect.For example, if in the course of their work a doctor, a police officer or a teacher encounters a three year old child who has suffered severe intentional physical injury, or injuries suggesting sexual abuse, or severe neglect, the legal obligation requires the professional to report their knowledge or reasonable suspicion that the child has been abused and has suffered harm to a government child welfare agency so that the agency can assess the child's situation to determine what protective and supportive actions need to be taken.Importantly, and unlike a softer policy-based obligation, the legislation provides the reporter with protections: their identity as the reporter is confidential, and they cannot be liable in any civil, criminal or administrative proceeding for any consequences of the report [5].Reporters are not expected to always be correct; the indicators of abuse can be mimicked by other health conditions, accidental injuries and childhood adversities [11].
The laws need to be accompanied by sufficient training, as reporters need to know the scope of their duty, the indicators of child abuse and neglect, what they should and should not report, how to make a useful report, and to whom to report.Child welfare agencies are equipped with staff to receive reports and assess them, and where necessary to investigate them and determine the appropriate course of action as set out in child protection legislation.These agencies are not required by statute to investigate every report and data shows that large proportions of reports are screened out at intake [1,7].Generally, child welfare statutes dictate that the least intrusive course of action is preferred [41].Consistent with this, methods of differential response are increasingly used, especially for reports of neglect and emotional abuse; this is discussed in more detail below.However, where necessary, cases of severe maltreatment where a child cannot be protected within the home may result in more formal action including court orders which may extend to removal of the child from the family home for the child's protection.Such removal is often only temporary with the child then being returned to the family home, but in some cases it may be necessary to be of a more extended duration, and sometimes may result in permanent placement with other caregivers.The cost component of out-of-home care (foster care, kinship care and other forms e.g.residential care) constitutes by far the greatest economic burden for child protection systems, dwarfing investigation costs [8].

A Spectrum of Approaches
The laws are not uniform across jurisdictions, which means it is neither possible nor correct to make absolute statements about their scope and therefore their consequences.Rather, they can be designed to be broad or narrow.There are two major dimensions of difference (both between countries, and within countries): first, in which types of abuse and neglect must be reported; second, which persons have to report.Accordingly, there is a spectrum of approaches to mandatory reporting laws.
At one end of the spectrum, a law might require reports only of sexual abuse, and limit the reporter groups to teachers, doctors, nurses, midwives and police.This is the position in Western Australia [42].Or, like the first reporting laws in the USA, it might only require medical practitioners to report serious physical abuse [5,43].Further along the spectrum, the law might require reports of physical abuse and sexual abuse, as in Victoria and the Australian Capital Territory [5].At the other end of the spectrum, a jurisdiction might design its law more broadly, requiring reports of all four forms of child abuse and neglect, and perhaps even others such as situations of a child being exposed to domestic violence, where the significant harm dimension is also present; this is the position in several Australian jurisdictions including New South Wales and Tasmania [5].Regarding the second major dimension of variance, at various points on the spectrum, there may be narrower or broader ranges of reporter groups.Saudi Arabia's legislation, for example, applies only to health practitioners [10].The Australian jurisdictions of Queensland, Victoria and Western Australia limit the reporting duty to a smaller range of occupations.In contrast, most jurisdictions in the USA, Canada and Australia have an extensive range of reporter groups, and some states apply the duty to all citizens [5].
Many jurisdictions now have relatively broad laws, requiring a range of occupational groups to report a number of types of child maltreatment.However, it is important to recall that these laws are generally not designed to require reports of any and all manifestations of harm or "maltreatment"; rather, for the most part they are directed at serious forms of harm and maltreatment of children [5,7].Some jurisdictions have a more preventative orientation and so have a lower threshold for the activation of the reporting duty; these jurisdictions would need to devote a higher level of resourcing to receive and respond to the corresponding higher number of reports.Problems can arise where laws are poorly drafted, and reporters are not adequately trained [7].

The Differentiation Thesis
The differentiation thesis proposed here argues that the different patterns of reporting between reporter groups and maltreatment types must be acknowledged and analysed, and careful analysis (including empirical analysis) should inform discussions and assessments of optimal approaches in law, policy and practice.It can readily be perceived that a broader legislative reporting duty will likely lead to differential levels of reporting, and different levels of case identification, than a narrower version.However, reporting practices differ substantially across reporter groups (both mandated and non-mandated), and by maltreatment type, and over time.This differentiation again means that all-encompassing statements about "the effect" of mandatory reporting laws on reporting cannot be made.For example, neglect and emotional abuse, and exposure to domestic violence, are far more frequently reported by mandated reporters and by all reporters than are sexual abuse and physical abuse [7].Furthermore, the reporting practice of one group of reporters, regarding one kind of maltreatment, can substantially skew total statistics on reporting, and hence can have a disproportionate impact on child protection systems.A good example of this was identified in New South Wales, where reports by police of children's exposure to domestic violence in one year accounted for one quarter of all reports by all mandated reporters of all types of child maltreatment [7].That single group of reports outnumbered all reports by all mandated reporters combined of physical abuse and sexual abuse combined.For multiple reasons, this particular example of reporting practice has not continued, demonstrating that reporting practices and trends change over time, influenced by changes to the law, reporter training, and other influential variables.Researchers, policy-makers, opinion leaders and practice leaders must remain mindful of the heterogeneity of both maltreatment types, and of reporting practices regarding them, when considering issues concerning mandatory reporting and other methods of case identification, and other child and family welfare responses.

Observations on Mandatory Reporting and Different Forms and Extents of Maltreatment
It is not the purpose of this article to formulate a systematic proposal for how a child protection system should treat various forms and extents of maltreatment.Such a complex task would require much more detailed treatment and in any event will always be a controversial exercise and one which of necessity also involves political and economic choices.Yet, it is possible to make some observations about current developments in different forms and extents of maltreatment and mandatory reporting laws, and about the motivations behind them.
A range of evidence indicates the reporting laws and associated infrastructure and education will identify more cases, although there can be an added element of systems burden depending on the nature and scope of the laws, differential reporting patterns, and efficacy of implementation [6][7][8]43,44].Reports made by mandated professional reporters account for the majority of all substantiated cases ( [6,7]; [45], Tables 3-9).Jurisdictions continue to explore the use of mandatory reporting laws to identify cases, overcome the gaze aversion that can otherwise characterise professional practice [10,37], and implement the laws in various forms.Ireland has forthcoming legislation (the Children First Bill 2014) which will apply to sexual abuse, and to physical assault, neglect and other ill-treatment which seriously affects or is likely to seriously affect the child's health, development or welfare.Saudi Arabia recently introduced mandatory reporting laws which apply to health practitioners only [10] with new enactments potentially forthcoming for educational professionals [46].In 2014, in Australia, Queensland passed legislation which restricts the reporting duty to physical and sexual abuse, placing a new higher emphasis on differential response for other forms of maltreatment.
In every jurisdiction where they have been enacted, except one, mandatory reporting laws apply to both physical and sexual abuse.The exception is Western Australia, where only sexual abuse must be reported.The focus on sexual abuse reflects its serious consequences, its clandestine nature, its criminality, and the well-established difficulty and delay in children's disclosure.The focus on serious physical abuse reflects the genesis of the first reporting laws, the serious nature of physical abuse and its potential consequences especially for neonates, infants and young children, its co-existence with emotional harm, and the likelihood of its continuance.Most reporting laws also require the reporting of serious neglect and emotional abuse, recognizing the serious consequences which can flow from such acts and omissions, and the need to provide appropriate support to the child and family.
However, in several jurisdictions, including some in Australia, mandatory reporting laws do not extend to neglect or emotional abuse, no matter how severe.This indicates that some jurisdictions have concluded for various reasons that this strategy of case identification is not optimal or necessary.This article does not seek to evaluate all aspects of this context, but while it seems reasonable to not require reports of milder forms of neglect (which are not required anywhere in Australia in any event), it is difficult to justify a complete removal of neglect from mandatory reporting laws unless evidence indicates reporting practice is intolerable, irremediable, and on balance unhelpful.Neglect can involve intentional failure to provide medical care, intentional withholding of nutrition, failure by caregivers to engage with assistance and refusal or inability to change.Such cases can result in serious harm and even death; far more children die of neglect than any other kind of maltreatment [19].Instances of this wilful criminal conduct can result in convictions for manslaughter or murder [47].Reports of serious neglect can therefore, in theory, save lives.Hence, variance of circumstances and levels of harm within the category of neglect suggests that while some cases may suit a particular strategy, other cases require a different approach.
Similarly, several jurisdictions' mandatory reporting laws do not require reports of emotional abuse; examples include Victoria, the Australian Capital Territory, and Western Australia [5].Some commentators have suggested that more formal child protection responses may not be appropriate for emotional abuse, due to its particular characteristics which distinguish it from other maltreatment types [48,49].Glaser has developed a detailed typology of such harmful acts and omissions [15].Somewhat like neglect, emotional abuse can be of lesser or greater extents, and can cause either low or high degrees of harm.This again may suggest that a range of responses are required depending on the circumstances.
There does appear to be a persuasive argument that at least some instances of some types of maltreatment should not fall within the remit of mandatory reporting laws, although they may well warrant referral to welfare services.For example, where a teenaged child is experiencing non-intentional mild or even moderate neglect (such as inadequate clothing) caused only by poverty, but otherwise displays no indicators of serious harm, and lives in a loving caring family, there seems to be no compelling reason to require any report or formal engagement with child protection agencies.Instead, what is required is an offer of appropriate assistance to the child's family.While such a case would not activate the reporting duty in jurisdictions where the reporting duty is only triggered by suspicion of significant harm, in those jurisdictions where the reporting duty is not clearly confined to cases of serious harm it is conceivable that a welfare and service-oriented response is appropriate.

Differential Response
Differential response is the concept that a different kind of response to formal investigation can more efficiently and justifiably treat some types of situation which do not involve serious harm and which have a different type of needed response [50].The focus is on situational assessment and provision of services to the child's caregivers and the child.While it varies in implementation, the central premise of such a system is that it can operate in tandem with a more traditional investigative response, so that in cases of only mild or moderate harm or risk of harm, most often characterised by family need, child welfare agencies can respond by the provision of support rather than by investigating the situation to determine formally whether maltreatment has occurred [50].The idea is that this softer, less confronting approach more flexibly accommodates qualitatively different situations, heightens parental engagement with support services, avoids stigma, and is faster and cheaper [50,51].Children's situations can be redirected from the differential pathway to the formal investigation path, and vice versa.
Some jurisdictions have legislatively enabled direct referrals of such lesser situations-that is, of need, rather than harm to the child-to be made to community welfare agencies, even where the mandatory reporting duty is only activated by a case of suspected significant harm.Examples of this direct referral by a professional reporter to a differential response strategy in Australia exist in New South Wales, Victoria, and Tasmania.Another way in which lower-level cases can be referred to such agencies for the purpose of assessment of need and delivery of services, rather than for investigation, is by the intake agency forwarding reports by mandated reporters of suspected maltreatment to these agencies.Differential response is also widely used in the USA [51], where, largely for historical reasons related to tied federal grants, some jurisdictions may have traditionally more preventative approaches and so do not as clearly limit the reporting duty to cases of significant harm [5,43].
Differential response has in fact been facilitated by many child protection statutes for years, but may not have been implemented appropriately [7,41].It is a strategy which is intuitively appealing and appears to offer much practical promise.However, its success is now being vigorously debated [50,[52][53][54][55].It has been asserted that such systems must be shown by rigorous evidence to be successful [56] (and not only by measures of parental satisfaction), should not compromise the child's safety, should be supported by a capacity to compel parental compliance where necessary (noting that parental engagement is voluntary), and must not be used by politicians to withdraw net funding from the child protection and child welfare endeavour [53][54][55].The release in June 2014 of a recent evaluation of a controlled trial in Illinois found that despite being given a higher level of services and direct financial support, of the group assigned to differential response, less than half of the parents engaged with services to completion, and higher rates of children were involved in subsequent reports of maltreatment [57,58].Governments seem to find differential response increasingly appealing, but caution should be exercised before concluding that it is a successful option and a rigorous evidence base should be developed to ascertain the extent of its success and the conditions required for it to function.If adopted, it must not result in reduced investment in child and family welfare.
As with all public health measures, it is essential that legal and practical responses to various types and extents of maltreatment be supported by a rigorous evidence base; approaches must be monitored to enable assessment of effectiveness and identify areas where improvement or change is needed [9].This applies equally to elements of mandatory reporting laws, other approaches to identification of maltreatment, and differential response.Generating such evidence presents logistical, political and methodological challenges, which explains why aspects of this context are generally under-researched.The next section of this article contributes to the evidence base regarding one dimension of a key question in this field.

A Comparison of Jurisdictions With and Without Legislative Mandatory Reporting of Child Sexual Abuse: Does Mandatory Reporting Appear to Identify More Cases?
Questions arise about the impact of mandatory reporting legislation on the number of reports, and the actual identification of child abuse cases.Some have argued on various grounds, including empirical grounds, that mandatory reporting laws are overall a useful social policy response to child maltreatment, and especially so for certain classes of maltreatment [5,7,8].Recent government inquiries in Australia, in New South Wales, Victoria and Queensland, have all supported the continuance of mandatory reporting legislation [59][60][61], and even its expansion [60].This is not to say that improvements cannot be made; there are specific subsets of reporting which have been found to be problematic, and research should be conducted to identify areas of ineffective reporting so that they may be improved [7].
Others have asserted that mandatory reporting is not an effective measure for case identification [62].However, when made, this assertion is not founded by a sufficiently detailed analysis of child protection reporting data, either as a whole, or by distinguishing between reports and outcomes of different kinds of abuse and neglect, by different reporter groups.As well, such an assertion is often based on an overall increase in numbers of reports in one jurisdiction without adequate discrimination between different jurisdictions, reporter groups, or maltreatment types.This absence of analysis based on evidence renders the assertion nugatory.Reporting patterns and outcomes differ markedly across different maltreatment types; for example, neglect is generally by far the highest number of reports, followed by emotional abuse (which can include exposure of children to domestic violence).Neglect and emotional abuse combined generally account for around two thirds of all reports [7,40].In contrast, reports of physical abuse and sexual abuse represent a much smaller proportion of all reports [7].The assertion also fails to recognise that reports by mandated reporters generally only account for around 50%-60% of all reports, the rest being made by nonmandated reporters such as family members and neighbours [1,5,19].As well, report patterns over time indicate that report numbers do not constantly rise, and in fact may remain stable over periods of several years, and can even decline substantially [40].
A central question arises, regarding the impact of mandatory reporting within a jurisdiction on numbers of reports and numbers of substantiated, or confirmed, reports (as well as of other reports which may not be confirmed but which are useful).To accommodate the differentiation thesis proposed here, this question must be analysed in a highly nuanced fashion.That is, because maltreatment types are different, and because reporter groups and practices are different, this question needs to be posed for each reporter group, for each maltreatment type.Ideally, this question needs to be the subject of rigorous analysis, as a robust evidence base-both quantitative and qualitative-should inform policy debates about optimal responses to detect cases of child abuse and neglect, and should form the basis of systemic improvements [8].Quantitatively, for example, the kinds of questions explored by Drake et al. [8] should be asked to generate evidence within jurisdictions about whether a mandatory reporting law results in a higher number of reports by mandated reporters, and to what extent.Does a mandatory reporting duty result in a higher number of substantiated or confirmed cases from reports by that reporter group, and to what extent?Do these patterns continue, stabilise, or decline, or does the trend vary?If there is a higher number of reports, do they involve such a higher number of children (at least in unconfirmed cases) and an extra burden on the child protection system, children and families that it is intolerable?As part of a public health approach which requires ongoing monitoring and assessment of systems, all these and other questions should be asked, and investigated, within jurisdictions which have enacted mandatory reporting laws.
A related question often confronts policymakers in a jurisdiction which does not have a mandatory reporting law.Would the introduction of a reporting law for a specific type of maltreatment improve case detection?What other effects, such as an increase in report numbers, might be produced?While no method of measurement in any discipline is perfect, there are several ways in which such questions may be explored to produce useful results.One way of doing so is to empirically compare government data over the same time period on numbers and outcomes of reports for a specific type of child maltreatment in two comparable jurisdictions, only one of which has mandatory reporting legislation.This analysis conducts such a comparison regarding the reporting of child sexual abuse.

A Cross-Jurisdictional Comparison Between Ireland and the Australian State of Victoria: Does a Jurisdiction With Mandatory Reporting Detect More Cases of Child Sexual Abuse Than One Without it?
A quantitative comparison across two jurisdictions of numbers of reports of child sexual abuse, and the outcomes of these reports, can add to an understanding of the association of a mandatory reporting duty (and its associated infrastructural measures) on reporting practice and case identification.This analysis compares the jurisdictions of Ireland and the Australian State of Victoria.
These two jurisdictions provide a useful and legitimate comparison for several reasons, with Victoria having several features making it the best Australian comparator jurisdiction. 1First, based on the best available data, both jurisdictions in the year 2010 had comparable child populations.In the year 2010, child population data was not available for Ireland [63].However, census data for the Republic of Ireland from the year 2011 was accessed and analysed to reveal a population of children aged 0-17 inclusive of 1,148,687 [64].This is the same population number cited in the HSE Report [65].Australian census data from 2011 was accessed and analysed to reveal a population in Victoria of children aged 0-16 inclusive of 1,152,251 ( [66], Table 8).A cut-off of 16 years of age in Victoria was used for data analysis purposes because the mandatory reporting duty only applies to children aged under 17, and the data accessed and analysed were restricted to this population group.Hence, these data indicate almost identical populations, and year-on-year birth patterns indicate it is reasonable to assume the populations for the year 2010 were similarly close.
Second, the two jurisdictions are in different countries but have similar demographic characteristics.Ireland is a wealthy industrialised nation classed as very high on the Human Development Index [67], and Victoria is the second largest State in the nation of Australia, which also has a very high HDI [67].Both jurisdictions have compulsory education systems and well-established health, education and police services.Both jurisdictions also have government child welfare agencies that have existed for decades [41,68].In addition, both jurisdictions have witnessed numerous government inquiries into child abuse, including child sexual abuse, in recent years and it is not unreasonable to proceed on the 1 Population size, population density, ethnicity, geographical size, duration and consistency of mandatory reporting duty across professions, stability of child welfare approach, and availability of data.basis that at the general societal level there is a similar consciousness of the phenomenon of child sexual abuse.Both Ireland and Victoria are predominantly Anglo-Saxon jurisdictions with cultural characteristics that are not sufficiently dissimilar to suggest there would be distinct differences in the incidence of child sexual abuse or in general public awareness of it; for example, neither is a society in which evidence indicates a lower incidence of child sexual abuse for various cultural and social reasons [69].There are no other ethnic differences between the two jurisdictions sufficient to warrant a hypothesis that either would have a markedly higher prevalence of child sexual abuse.Some Indigenous communities in Australia have been found to have higher than normal rates of child sexual abuse but these are relatively isolated instances in small communities, are not generalizable, and have occurred in places beyond Victoria.Finally, it is well-established that child sexual abuse affects girls disproportionately, but there are no distinctive gender imbalances at the population level in either jurisdiction such that different incidence and detection patterns could be expected.
Third, population studies indicate that it is plausible to proceed on the basis that the prevalence and incidence of child sexual abuse in Ireland is not substantially different to that in Victoria, or Australia.In Ireland, McGee et al. found that one in five women (20%) reported experiencing contact sexual abuse in childhood and a further one in ten (10%) reported non-contact abuse [70].Penetrative sexual abuse was reported by 5.6% of the female participants.One in six men (16.2%) reported contact sexual abuse in childhood and a further one in fourteen (7.4%) reported non-contact abuse.Penetrative sexual abuse was reported by 2.7% of the male participants.These findings are similar to those produced in Victoria [71], and other Australian studies [18,22,72,73].
Fourth, one key differentiating variable is the presence in one jurisdiction, Victoria, of a legislative duty to report suspected child sexual abuse, imposed on police, teachers, doctors and nurses; together with whatever training about this duty is delivered to these reporters.This duty had existed since 4 November 1993 for doctors, nurses and police [74], and since 18 July 1994 for teachers [75], and it could be expected that over time, a reasonably well-developed awareness of the duty and the nature and consequences of child sexual abuse had crystallised in these professions.The duty in Victoria is actually quite narrow relative to most mandatory reporting duties, since it applies to a small range of reporter groups, and applies only to cases in which not only is the child's sexual abuse suspected, but the reporter must also suspect the child does not have a parent who is able and willing to protect the child [5].In contrast, Ireland had no legislative reporting duty, although some occupations had policy-based duties to report under the policy framework entitled Children First: National Guidelines for the Protection and Welfare of Children [76].These guidelines were subsequently crystallised in 2011 under the new policy package named Children First: National Guidance for the Protection and Welfare of Children [77].
The data presented and analysed are from the calendar year 2010.This year was selected because it is recent, and Ireland stopped collecting and publishing data on confirmed reports after this year [65].The Victorian data were accessed from the government's child protection department as part of a broader study.The Ireland data are presented in a government publication ( [63], p. 39, Table 17; p. 41, Figure 8).Table 1 presents data concerning the number of reports of suspected child sexual abuse, and the number of these reports that were substantiated or confirmed after assessment by government child protection agencies, with breakdowns by mandated/nonmandated reporter groups in Victoria.These data show several results.First, almost double the number of reports was made in the jurisdiction with legislative mandatory reporting (Victoria) compared with the jurisdiction without it (Ireland).Second, the data show that a substantial proportion of reports in Victoria are made by mandated reporters.They made 3,113 reports (53% of all reports of sexual abuse), compared with 2,757 made by all other reporters (47%).By comparison, Ireland received 2,962 reports from all sources, with a breakdown by profession not able to be made.Third, the higher volume of reports in Victoria resulted in almost five times as many sexually abused children being identified.There were 989 substantiated cases in Victoria compared with 209 in Ireland; a ratio of 4.73 times the amount, and numerically a difference in one year alone of 780.Fourth, a substantial number of confirmed cases are found as a result of reports made by mandated reporters.In Victoria, 536 substantiated cases were identified as a result of a report by a mandated reporter. 2 An interpretation of these results supports several important conclusions.First, it can reasonably be concluded that the higher number of reports made is associated with the reporting duty.This has implications for child protection systems because there must be adequately resourced intake and response systems so that the volume of reports and substantiated cases can be dealt with appropriately.However, it should be noted that in Victoria, exactly half of these reports were not investigated at all, and instead were screened out or referred to another service or agency.Accordingly, a substantial proportion of these reports resulted in little strain on the child protection system.It is also prudent to recall that sexual abuse reports are consistently the lowest proportion of any kind of maltreatment in Australia, the USA and Canada [7], although strangely this pattern was not replicated in the Irish data [63].As well, it must be recalled that many reports that are officially "unsubstantiated" still will involve a level of harm to the child, or some other kind of harm, and require the provision of services 2 Of the 3113 reports made by mandated reporters (involving 3039 children), 1838 were investigated and of these, 536 were substantiated.
to the child and family; there are frequently not substantial differences between substantiated and unsubstantiated decisions [8,78,79].Nevertheless, any jurisdiction which has mandatory reporting, or which is considering introducing it, must be prepared to invest sufficiently in the necessary infrastructure, personnel, training and service provision required to deal with the expected higher number of reports.This investment is nevertheless worthwhile and should result in long-term returns.
Second, a substantial proportion of reports are made by mandated reporters, accounting for 53% of all reports in Victoria.This indicates that overall there is a substantial amount of compliance with the reporting duty; it is possible that some suspected cases were not reported, but it cannot be concluded without further research that there is complete or substantial noncompliance.In comparison, reports made by the rest of the Victorian population accounted for less than half of all reports.In Ireland, without a breakdown of reporting by professional groups, and further research, it is not possible to draw conclusions about different reporting practices of different groups.However, if Irish police, teachers, and medical practitioners made a similar proportion of all reports as were made by their Victorian counterparts, they would have made approximately half the number of reports.
Third, and arguably most significantly, there was a substantially higher number and proportion of sexually abused children identified in the jurisdiction with mandatory reporting.Victoria identified 4.73 times the amount of children than did Ireland in the same year as a result of all reports of suspected cases to child protection agencies.Over 700 more victims of child sexual abuse were identified in the course of a single year.If the primary unit of analysis in this context is the identification of cases of actual child sexual abuse (and in a small minority of cases, imminent risk of sexual abuse), then this outcome suggests the strategy of legislative mandatory reporting, along with its accompanying reporter training, and the cultural, attitudinal and behavioural changes it can help engender, is by far a superior strategy.
Fourth, a substantial number of the confirmed cases were found as a result of reports by mandated reporters.In Victoria, 536 of the 989 substantiated reports were made by mandated reporters (54%).Even in the unlikely event that all of Ireland's 209 confirmed cases were a result of reports by the same professional groups as are mandated in Victoria, the number of detected cases is still exceeded by a factor of 2.5.

A Cross-Jurisdictional Comparison Between Ireland and Australia: Does a Jurisdiction With Mandatory Reporting Detect More Cases of Child Sexual Abuse Than One Without it?
It might be surmised that for some unknown reason the Victorian data are a statistical artefact.To address the argument or possibility that an unapparent contextual variable makes the two jurisdictions incomparable, or that for some other reason the Victorian data are an outlier, a further analysis of the broader Australian position may be useful.A scan of the data regarding confirmed cases of child sexual abuse from Ireland compared with the entire nation of Australia may indicate whether the Victorian experience is generally consistent with the Australian statistical picture.The number of reports of suspected child sexual abuse made in each Australian jurisdiction, and the breakdown of these (and their outcomes) by reporter group (including mandated reporters), is not publicly available, although a previous analysis concluded that mandated reporters in Australia accounted for approximately 58% of all substantiated reports of all maltreatment [6].Accordingly, this additional analysis is simply of the numbers of children in confirmed cases after all reports have been made to child protection agencies, relative to the national child population.Each of Australia's eight States and Territories have legislative mandatory reporting of child sexual abuse (although not of all other forms), and did so for the year 2010 [42,80].Australian population data accessed from the Australian Bureau of Statistics reveals a child population in June 2011 of 5,074,810 [66]; this compares with the Irish child population of 1,148,687 [64].The Irish data on confirmed cases are reproduced as above.The Australian data was accessed from the annual national publication of aggregate data by the Australian Institute of Health and Welfare ( [40], p. 87).Comparison of the population data from the two countries, and of the numbers of children aged under 18 in confirmed cases of child sexual abuse, is presented in Table 2.These data show two results.First, in the year 2010, the difference in the two nations' child populations means Australia had a child population 4.4 times that of Ireland.Second, in this year, there were 4427 children in confirmed cases of sexual abuse in Australia, compared with 209 in Ireland.
An interpretation of these results supports several conclusions.Based on the difference in the child population, and the similar incidence of child sexual abuse in these two countries, it might be expected that under other broadly similar conditions, the number of children determined after investigation by child protection agencies in Australia to have been sexually abused would be 4.4 times higher than that in Ireland.That is, if Australia's performance is used as a minimal baseline, then the number of cases of child sexual abuse identified in Ireland as a result of all reports made child protection agencies would be around 1,100.This figure is produced by dividing Australia's number of children in substantiated cases (4427) by the population ratio (4.4).Yet, the data show that in Ireland there were only 209 children in confirmed cases.Accordingly, accounting for population difference, the proportional difference between Australia and Ireland in the detection of child sexual abuse is almost five times what could otherwise be expected.This ratio is similar to the 4.73 identified in Victoria, suggesting the Victorian data are not an artefact.
These data show that the core child protection agencies in the two jurisdictions are functioning in significantly different ways.There are incomplete data in some respects; for example, this analysis has not explored the comparative resourcing of child protection departments.It does not require an assertion that Australian child protection agencies are generally functioning in a much more efficient manner than those in Ireland-there are issues of resourcing and implementation in all child protection systems.In sum, in real terms, these data indicate that substantially higher numbers of sexually abused children are having their situations identified in a single year in Australia compared with Ireland.Since Australian jurisdictions have a large range of mandated reporter groups, and as reports from mandated reporters are known to lead to a substantial majority of all identified cases of sexual abuse-as shown by the Victorian finding above of 54%, and by other findings [1,6,7]-it is plausible that the presence of mandatory reporting laws is a significant factor contributing to this difference.

Questions of Systems Burden and Net Widening
Some have argued against mandatory reporting on the basis of net widening and systems burden [62,81].Net widening is the claim that an ever increasing number of children are captured by the resulting reports produced by a legislative mandatory reporting obligation.The claim was rejected by the Wood Inquiry in New South Wales after extensive statistical analysis ( [61], p. 170).Systems burden refers to the extra time, personnel and financial impositions placed on government child protection agency intake and response services.When made, these assertions do not distinguish between types of maltreatment, are not made with detailed examination of reports by maltreatment type, and do not distinguish between reports by mandated reports as opposed to nonmandated reporters.The claims should be assessed with reference to evidence.The strength or weakness of the assertions require analysis from multiple perspectives, one of which must involve a detailed child population analysis undertaken alongside an analysis of reports by type of maltreatment.
Within the context of child sexual abuse, an analysis of the reporting data in Victoria and Ireland, situated within the context of the child populations in these jurisdictions, generates information which contributes to an assessment of this claim for one maltreatment type.The population data from June 2011 is used for illustrative purposes as Ireland's data from 2010 is not available; this has been calculated to make only marginal difference to the 2010 rates shown below.Table 3 presents data concerning the number of reports, and the number of children involved in these reports, and generates a contextual rate of children involved in reports.These data show that in Ireland, reports made by all reporters in 2010 involved 2962 children, which equates to one in every 387 children.In Victoria, reports made by all reporters involved 5445 children, equating to one in every 211 children.In Victoria, mandated reporters made 3,113 reports in total, involving 3039 distinct children, equating to one in every 379 children.The rate of children involved in reports by mandated reporters in Victoria (1 in 379) is very close to the rate of children in all reports in Ireland (1 in 387).For context, a primary school of 370-400 students would represent a school of reasonable size.On this basis, these data indicate that in each jurisdiction one child in a school of such a size is being reported as the suspected victim of child sexual abuse per annum.Further, recalling that one half of all reports in Victoria did not result in investigation, it is strongly arguable that, at least for child sexual abuse, the child protection net is neither being cast unreasonably wide, nor is it causing intolerable systems burden.

Conclusions
Mandatory reporting laws are one component of a society's response to situations of significant child abuse and neglect.The laws can be designed in different forms, thus having broader or narrower potential to identify cases of certain types of maltreatment.As with all aspects of any public health response to a complex problem, a rigorous evidence base must be generated to inform assessments of a particular approach and its consequences, effectiveness, and areas for improvement or change.The different forms and extents of child abuse and neglect present different challenges for child welfare systems, and this heterogeneity suggests a range of responsive measures are required to respond to, and prevent escalation of, different cases.Differential response measures, as a part of an essential secondary prevention dimension in a public health approach to child maltreatment, must also be monitored for effectiveness and outcomes.Their adoption may have promise but should not be unquestioningly accepted.
The differentiation thesis proposed in this article posits that the operation of mandatory reporting laws differs across maltreatment types and reporter groups.This differentiation needs to be recognised and it must inform policy debates about the laws and their effects.Nuanced research including quantitative and qualitative work should explore these different phenomena to understand their operation and outcomes, inform enhancements to practice, and even indicate more substantial reform options.Developments continue worldwide in the adoption and implementation of various kinds of mandatory reporting laws, although the field lacks a detailed evidence base about their consequences.Jurisdictions which introduce the laws should carefully monitor their implementation.
This article conducted empirical analysis of the reporting over a single year of child sexual abuse in two comparable jurisdictions, only one of which had mandatory reporting.It made four significant findings.In the jurisdiction with mandatory reporting, double the number of reports were made (with 53% of these made by mandated reporters); a substantially higher number of sexually abused children were identified (the proportional difference was 4.73; the numerical difference was over 700); 54% of confirmed cases were identified as a result of reports by mandated reporters (2.5 times the entire amount identified by all reporters in the other jurisdiction); and additional substantial systems burden and net widening was not apparent.
Further research into the different reporting outcomes associated with different legal and policy approaches needs to be undertaken, for different kinds of maltreatment.In addition, research should be conducted which compares multiple jurisdictions over longer periods, and studies of the effects of introducing a mandatory reporting duty on reporting practices and outcomes should be done to offer further valuable insights.However, the findings of this study indicate that mandatory reporting of child sexual abuse produces substantial and superior outcomes in identifying children who have been abused, compared with an approach which does not include mandatory reporting.

Table 1 .
Number of reports of suspected child sexual abuse, and number of substantiated reports, in Victoria and Ireland (2010).
* Proportional differences are calculated by dividing the larger total by the smaller total.# n.a.not applicable.

Table 2 .
Number of children aged 0-17 in confirmed cases of child sexual abuse, Australia and Ireland, 2010.
* Proportional differences are calculated by dividing larger total by the smaller total.

Table 3 .
Rates of children in reports of child sexual abuse, 2010.