Child Abuse & Neglect 48 (2015) 208–220
Contents lists available at ScienceDirect
Child Abuse & Neglect
Research article
Burden attributable to child maltreatment in Australia Sophie E. Moore a , James G. Scott b,c , Alize J. Ferrari a,d,e , Ryan Mills f,g , Michael P. Dunne h , Holly E. Erskine a,d,e , Karen M. Devries i , Louisa Degenhardt e,j,k , Theo Vos e , Harvey A. Whiteford a,d,e , Molly McCarthy l , Rosana E. Norman m,n,∗ a
School of Public Health, University of Queensland, Herston, QLD, Australia Metro North Mental Health, Royal Brisbane and Women’s Hospital, Herston, QLD 4029 Australia c The University of Queensland Centre for Clinical Research, Herston, QLD 4029 Australia d Queensland Centre for Mental Health Research, Wacol, QLD, Australia e Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA, USA f Department of Paediatrics, Logan Hospital, Metro South Hospital and Health Service, QLD, Australia g School of Medicine, University of Queensland, Herston, QLD, Australia h Children and Youth Research Centre, School of Public Health and Social Work, Queensland University of Technology, QLD, Australia i Gender Violence and Health Centre, London School of Hygiene and Tropical Medicine, London, United Kingdom j University of New South Wales, National Drug and Alcohol Research Centre, Sydney, NSW, Australia k Melbourne School of Population and Global Health, University of Melbourne, Melbourne VIC 3010, Australia l School of Criminology and Criminal Justice, Griffith University, Australia m Institute of Health and Biomedical Innovation, Queensland University of Technology, QLD, Australia n School of Public Health and Social Work, Queensland University of Technology, QLD, Australia b
a r t i c l e
i n f o
Article history: Received 5 February 2015 Received in revised form 1 May 2015 Accepted 6 May 2015 Available online 6 June 2015 Keywords: Child maltreatment Australia Child sexual abuse Child physical abuse Child emotional abuse Child neglect Depressive disorders Anxiety disorders Intentional self-harm
a b s t r a c t Child maltreatment is a complex phenomenon, with four main types (childhood sexual abuse, physical abuse, emotional abuse, and neglect) highly interrelated. All types of maltreatment have been linked to adverse health consequences and exposure to multiple forms of maltreatment increases risk. In Australia to date, only burden attributable to childhood sexual abuse has been estimated. This study synthesized the national evidence and quantified the burden attributable to the four main types of child maltreatment. Meta-analyses, based on quality-effects models, generated pooled prevalence estimates for each maltreatment type. Exposure to child maltreatment was examined as a risk factor for depressive disorders, anxiety disorders and intentional self-harm using counterfactual estimation and comparative risk assessment methods. Adjustments were made for co-occurrence of multiple forms of child maltreatment. Overall, an estimated 23.5% of self-harm, 20.9% of anxiety disorders and 15.7% of depressive disorders burden in males; and 33.0% of self-harm, 30.6% of anxiety disorders and 22.8% of depressive disorders burden in females was attributable to child maltreatment. Child maltreatment was estimated to cause 1.4% (95% uncertainty interval 0.4–2.3%) of all disability-adjusted life years (DALYs) in males, and 2.4% (0.7–4.1%) of all DALYs in females in Australia in 2010. Child maltreatment contributes to a substantial proportion of burden from depressive and anxiety disorders and intentional self-harm in Australia. This study demonstrates the importance of including all forms of child maltreatment as risk factors in future burden of disease studies. © 2015 Elsevier Ltd. All rights reserved.
∗ Corresponding author at: Institute of Health and Biomedical Innovation, School of Public Health and Social Work, Queensland University of Technology, 60 Musk Ave, Kelvin Grove, QLD 4059, Australia. http://dx.doi.org/10.1016/j.chiabu.2015.05.006 0145-2134/© 2015 Elsevier Ltd. All rights reserved.
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
209
Introduction Child maltreatment is a serious public health problem in Australia and worldwide, adversely affecting short- and longterm physical and mental health (Chen et al., 2010; Norman et al., 2012), social development and functioning (Alink, Cicchetti, Kim, & Rogosch, 2012; Kim & Cicchetti, 2003). Maltreatment of children is commonly divided into four main types: childhood sexual abuse (CSA), childhood physical abuse (CPA), childhood emotional abuse (CEA), and childhood neglect (CN) (Butchart, Phinney Harvey, Kahane, Mian, & Furniss, 2006). The definition of child maltreatment varies within Australian studies and globally, along with the definition of what constitutes a child (under 15, 16 or 18 years of age). These sources of variation, as well as other methodological factors such as the mode of data collection and type of sample assessed, lead to differences in the reported prevalence (Andrews, Corry, Slade, Issakidis, & Swanston, 2004). For instance, meta-analyses of CSA prevalence around the world have made use of estimates ranging from 8% to 31% in females and 3% to 17% in males; with pooled self-reported prevalence estimated higher in females (approximately 18% for females and 8% for males) (Barth, Bermetz, Heim, Trelle, & Tonia, 2013; Pereda, Guilera, Forns, & Gómez-Benito, 2009; Stoltenborgh, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). The prevalence of other forms of child maltreatment has not been as extensively investigated; however, available literature suggests that every year 4–16% of children are physically abused and 10% emotionally abused or neglected in high income countries (Gilbert et al., 2009). In recent meta-analyses of worldwide prevalence, the overall estimate was 23% for studies using self-report measures of CPA, with no apparent gender differences, (Stoltenborgh, Bakermans-Kranenburg, van IJzendoorn, & Alink, 2013) and 36% for studies using self-report measures of CEA (Stoltenborgh, Bakermans-Kranenburg, Alink, & van IJzendoorn, 2012). Regarding CN, estimated global prevalence based on a modest number of studies was 16% for physical neglect and 18% for emotional neglect with no apparent gender differences (Stoltenborgh, BakermansKranenburg, & van IJzendoorn, 2013). Currently there are no Australia-wide studies of the prevalence of child maltreatment. National Child Protection data exist, but these likely under-represent the true number of children experiencing maltreatment as they only capture cases where maltreatment has been reported to authorities, it can be substantiated, and the risk of harm to the child is deemed to be sufficiently high to justify intervention (Australian Institute of Health and Welfare, 2014). Empirical evidence has linked all forms of child maltreatment with adverse mental and physical health outcomes (Chen et al., 2010; Norman et al., 2012). Experiencing at least one form of child maltreatment may double the risk of developing mental health problems such as depressive and anxiety disorders, and increase the risk of self-harm and suicidal ideation and attempts (Afifi et al., 2008; Chen et al., 2010; Devries et al., 2014; Norman et al., 2012). The four types of child maltreatment are highly interrelated (Dong et al., 2004; Finkelhor, 2008; Finkelhor, Ormrod, & Turner, 2007; Jirapramukpitak, Prince, & Harpham, 2005; Mullen, Martin, Anderson, Romans, & Herbison, 1996; Teicher, Samson, Polcari, & McGreenery, 2006). In a cross-sectional study of the effect of multiple types of maltreatment in children and adolescents in Viet Nam, substantial proportions of participants reported exposure to multiple abuse (21% reported two types; 15% three types, and 6% four types of maltreatment) (Nguyen, Dunne, & Le, 2010). Similar levels of multi-type maltreatment have been found in Malaysia (Choo, Dunne, Marret, Fleming, & Wong, 2011). Exposure to multiple forms of maltreatment increases the child’s risk of developing later mental health problems (Edwards, Holden, Felitti, & Anda, 2003; Holt, Finkelhor, & Kantor, 2007; Rikhye et al., 2008; Schneider, Baumrind, & Kimerling, 2007; Teicher et al., 2006; Turner, Finkelhor, & Ormrod, 2006; Zoroglu et al., 2003). In Australia, as in most countries, there has not been a comprehensive assessment of the health consequences of child maltreatment at the national level. CSA was the only form of maltreatment included as a risk factor in the Australian Burden of Disease 2003 study (Begg et al., 2008) and in the Global Burden of Disease 2010 study (GBD 2010) (Lim et al., 2012). The omission of other types of child maltreatment limits our understanding of how these complex phenomena relate to each other, and how co-occurrence of multiple types may influence the overall burden. This is also an important gap in the international literature (Fang et al., 2015). In addition, to date, studies estimating the economic burden of child maltreatment (Fang, Brown, Florence, & Mercy, 2012; Fang et al., 2015) have not adjusted for co-occurrence of maltreatment types. The present analysis aims to derive estimates of the burden of depressive disorders, anxiety disorders and intentional self-harm attributable to CSA, CPA, CEA and CN in Australia adjusting for co-occurrence of multiple forms of child maltreatment. It is hoped that this will lead to quantification of the burden attributable to all forms of child maltreatment in future iterations of the Australian and global burden of disease studies and inform estimation of the economic impact of child maltreatment.
Methods Exposure to child maltreatment was treated as a risk factor for disease and injury using counterfactual estimation and comparative risk assessment methods (Lim et al., 2012). This involved comparing the current local health status with the theoretical-minimum-risk exposure defined as a population not ever having been exposed to child maltreatment. Population attributable fractions (PAFs) were determined by the prevalence of ever having been exposed to these risk factors in the population and the relative risks (RR) of disease occurrence given exposure.
210
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
Risk Factor Definitions CSA was defined as the involvement in sexual activity that a child does not fully comprehend, is unable to give informed consent to, or for which the child is not developmentally prepared (Butchart et al., 2006). As per GBD 2010 (Lim et al., 2012), in this study CSA was defined as ever having experienced any form of CSA and hence would encompass all sub-types including unwanted non-contact, contact abuse, or intercourse (penetrative sexual abuse) in childhood. CPA was defined as the intentional use of physical force against a child that results in (or has a high likelihood of resulting in) harm to the child’s health, survival, development or dignity. This included hitting, beating, kicking, shaking, biting, strangling, scalding, burning, poisoning, and suffocating the child (Butchart et al., 2006). CEA involved both isolated incidents as well as a pattern of failure over time on the part of the parent or caregiver to provide a developmentally appropriate and supportive environment to the child. This included the restriction of movement, patterns of belittling, blaming, threatening, frightening, discriminating against or ridiculing and other non-physical forms of rejection or hostile treatment (Butchart et al., 2006; Hamarman & Bernet, 2000). CN included both isolated incidents as well as a pattern of failure over time on the part of a parent or caregiver to provide for the development and wellbeing of the child (where the parent is in a position to do so) in one or more of the following areas: health, education, emotional development, nutrition, shelter and safe living conditions (Butchart et al., 2006). Prevalence of Exposure We conducted a systematic review in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009) to identify studies that provided data on the prevalence of exposure to CSA, CPA, CEA, or CN in Australia. Three electronic databases (PubMed, EMBASE, and PsycINFO) were searched up to 15 August 2014 using the terms, ‘child*’ along with ‘maltreatment’, ‘sexual abuse’, ‘emotional abuse’, ‘neglect’, ‘physical abuse’ as well as ‘Australia*’ and ‘prevalence’ (Supplementary Data Text S1). This systematic review incorporated studies meeting the following inclusion criteria: (1) the study reported original, empirical published research; (2) the study presented lifetime prevalence data (proportion of the population who have ever experienced one or more of the four sub-categories of child maltreatment in childhood); (3) maltreatment was reported by victims rather than informants (health professionals, teachers, family members); (4) the study was conducted in Australia and was largely representative of the Australian population. We included studies with different definitions of childhood (ranging from under 15 to under 18 years) as well as studies where childhood was not defined (e.g. maltreatment occurs “in childhood”). Studies of child maltreatment conducted in justice system, high risk or clinical samples [for example participants with border-line personality disorder (Goldman, D’Angelo, DeMaso, & Mezzacappa, 1992) or a history of depressive disorders (Gladstone et al., 2004)] were excluded. Where the same data were reported across different publications, the most informative article was selected: for example, studies reporting sex- or age-specific prevalence estimates were selected over those providing combined estimates. If multiple studies were carried out in the same population, the first published study was utilized ensuring sample independence with the inclusion of every participant only once in the subsequent meta-analyses. Reference lists of selected studies were screened for any other relevant studies. In addition to the systematic review of electronic databases, we surveyed Australian official statistics websites to identify country-level data for prevalence of exposure. Two relevant reports and data sources (Australian Bureau of Statistics, 2005; Australian Institute of Health and Welfare, 2014) were identified and reviewed. The full-text of studies that appeared to meet the inclusion criteria was retrieved for closer examination and assessed for eligibility. A standardized data extraction sheet was developed, and data retrieved included publication details, methodological characteristics such as sex, age of respondents, sample size, sample type, state in which the study was conducted, definition of childhood, exposure measures, type of abuse and prevalence reported. Quality of studies was assessed using a tool for assessing risk of bias in prevalence studies (Hoy et al., 2012) and converted to a proportional quality score (the total quality score divided by the maximum score possible) (Supplementary Data Text S1 and Supplementary Table S1). In the Child Protection Australia 2012–13 report, unit-record level data has been made available for analysis for the first time, providing unique counts of children receiving child protection services in each jurisdiction (Australian Institute of Health and Welfare, 2014). Substantiated cases are based on the type of abuse that was considered most severe or most likely to place the child at risk (the ‘primary’ type of maltreatment). Although official data could be used to estimate prevalence of substantiated child maltreatment in Australia for the four maltreatment categories (presented in Supplementary Table S1), they were not included in subsequent meta-analyses of prevalence as they are not based on retrospective self-reports and not fully representative of all cases of child maltreatment. The Personal Safety Survey presented CPA and CSA prevalence estimates that could be included in the meta-analysis (Australian Bureau of Statistics, 2005). Meta-analysis of Prevalence Some studies presented prevalence of maltreatment for both sexes combined. These estimates were converted to sex specific estimates in a 2-step process. First a female-to-male ratio was derived from studies identified in the systematic
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
211
Table 1 Results of meta-analysis of prevalencea of child maltreatment in Australia. Type of maltreatment Childhood sexual abuse primary analysis Any childhood sexual abuseb Maleb Femaleb Childhood sexual abuse subgroup analyses Age of respondents (children
Contents lists available at ScienceDirect
Child Abuse & Neglect
Research article
Burden attributable to child maltreatment in Australia Sophie E. Moore a , James G. Scott b,c , Alize J. Ferrari a,d,e , Ryan Mills f,g , Michael P. Dunne h , Holly E. Erskine a,d,e , Karen M. Devries i , Louisa Degenhardt e,j,k , Theo Vos e , Harvey A. Whiteford a,d,e , Molly McCarthy l , Rosana E. Norman m,n,∗ a
School of Public Health, University of Queensland, Herston, QLD, Australia Metro North Mental Health, Royal Brisbane and Women’s Hospital, Herston, QLD 4029 Australia c The University of Queensland Centre for Clinical Research, Herston, QLD 4029 Australia d Queensland Centre for Mental Health Research, Wacol, QLD, Australia e Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA, USA f Department of Paediatrics, Logan Hospital, Metro South Hospital and Health Service, QLD, Australia g School of Medicine, University of Queensland, Herston, QLD, Australia h Children and Youth Research Centre, School of Public Health and Social Work, Queensland University of Technology, QLD, Australia i Gender Violence and Health Centre, London School of Hygiene and Tropical Medicine, London, United Kingdom j University of New South Wales, National Drug and Alcohol Research Centre, Sydney, NSW, Australia k Melbourne School of Population and Global Health, University of Melbourne, Melbourne VIC 3010, Australia l School of Criminology and Criminal Justice, Griffith University, Australia m Institute of Health and Biomedical Innovation, Queensland University of Technology, QLD, Australia n School of Public Health and Social Work, Queensland University of Technology, QLD, Australia b
a r t i c l e
i n f o
Article history: Received 5 February 2015 Received in revised form 1 May 2015 Accepted 6 May 2015 Available online 6 June 2015 Keywords: Child maltreatment Australia Child sexual abuse Child physical abuse Child emotional abuse Child neglect Depressive disorders Anxiety disorders Intentional self-harm
a b s t r a c t Child maltreatment is a complex phenomenon, with four main types (childhood sexual abuse, physical abuse, emotional abuse, and neglect) highly interrelated. All types of maltreatment have been linked to adverse health consequences and exposure to multiple forms of maltreatment increases risk. In Australia to date, only burden attributable to childhood sexual abuse has been estimated. This study synthesized the national evidence and quantified the burden attributable to the four main types of child maltreatment. Meta-analyses, based on quality-effects models, generated pooled prevalence estimates for each maltreatment type. Exposure to child maltreatment was examined as a risk factor for depressive disorders, anxiety disorders and intentional self-harm using counterfactual estimation and comparative risk assessment methods. Adjustments were made for co-occurrence of multiple forms of child maltreatment. Overall, an estimated 23.5% of self-harm, 20.9% of anxiety disorders and 15.7% of depressive disorders burden in males; and 33.0% of self-harm, 30.6% of anxiety disorders and 22.8% of depressive disorders burden in females was attributable to child maltreatment. Child maltreatment was estimated to cause 1.4% (95% uncertainty interval 0.4–2.3%) of all disability-adjusted life years (DALYs) in males, and 2.4% (0.7–4.1%) of all DALYs in females in Australia in 2010. Child maltreatment contributes to a substantial proportion of burden from depressive and anxiety disorders and intentional self-harm in Australia. This study demonstrates the importance of including all forms of child maltreatment as risk factors in future burden of disease studies. © 2015 Elsevier Ltd. All rights reserved.
∗ Corresponding author at: Institute of Health and Biomedical Innovation, School of Public Health and Social Work, Queensland University of Technology, 60 Musk Ave, Kelvin Grove, QLD 4059, Australia. http://dx.doi.org/10.1016/j.chiabu.2015.05.006 0145-2134/© 2015 Elsevier Ltd. All rights reserved.
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
209
Introduction Child maltreatment is a serious public health problem in Australia and worldwide, adversely affecting short- and longterm physical and mental health (Chen et al., 2010; Norman et al., 2012), social development and functioning (Alink, Cicchetti, Kim, & Rogosch, 2012; Kim & Cicchetti, 2003). Maltreatment of children is commonly divided into four main types: childhood sexual abuse (CSA), childhood physical abuse (CPA), childhood emotional abuse (CEA), and childhood neglect (CN) (Butchart, Phinney Harvey, Kahane, Mian, & Furniss, 2006). The definition of child maltreatment varies within Australian studies and globally, along with the definition of what constitutes a child (under 15, 16 or 18 years of age). These sources of variation, as well as other methodological factors such as the mode of data collection and type of sample assessed, lead to differences in the reported prevalence (Andrews, Corry, Slade, Issakidis, & Swanston, 2004). For instance, meta-analyses of CSA prevalence around the world have made use of estimates ranging from 8% to 31% in females and 3% to 17% in males; with pooled self-reported prevalence estimated higher in females (approximately 18% for females and 8% for males) (Barth, Bermetz, Heim, Trelle, & Tonia, 2013; Pereda, Guilera, Forns, & Gómez-Benito, 2009; Stoltenborgh, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). The prevalence of other forms of child maltreatment has not been as extensively investigated; however, available literature suggests that every year 4–16% of children are physically abused and 10% emotionally abused or neglected in high income countries (Gilbert et al., 2009). In recent meta-analyses of worldwide prevalence, the overall estimate was 23% for studies using self-report measures of CPA, with no apparent gender differences, (Stoltenborgh, Bakermans-Kranenburg, van IJzendoorn, & Alink, 2013) and 36% for studies using self-report measures of CEA (Stoltenborgh, Bakermans-Kranenburg, Alink, & van IJzendoorn, 2012). Regarding CN, estimated global prevalence based on a modest number of studies was 16% for physical neglect and 18% for emotional neglect with no apparent gender differences (Stoltenborgh, BakermansKranenburg, & van IJzendoorn, 2013). Currently there are no Australia-wide studies of the prevalence of child maltreatment. National Child Protection data exist, but these likely under-represent the true number of children experiencing maltreatment as they only capture cases where maltreatment has been reported to authorities, it can be substantiated, and the risk of harm to the child is deemed to be sufficiently high to justify intervention (Australian Institute of Health and Welfare, 2014). Empirical evidence has linked all forms of child maltreatment with adverse mental and physical health outcomes (Chen et al., 2010; Norman et al., 2012). Experiencing at least one form of child maltreatment may double the risk of developing mental health problems such as depressive and anxiety disorders, and increase the risk of self-harm and suicidal ideation and attempts (Afifi et al., 2008; Chen et al., 2010; Devries et al., 2014; Norman et al., 2012). The four types of child maltreatment are highly interrelated (Dong et al., 2004; Finkelhor, 2008; Finkelhor, Ormrod, & Turner, 2007; Jirapramukpitak, Prince, & Harpham, 2005; Mullen, Martin, Anderson, Romans, & Herbison, 1996; Teicher, Samson, Polcari, & McGreenery, 2006). In a cross-sectional study of the effect of multiple types of maltreatment in children and adolescents in Viet Nam, substantial proportions of participants reported exposure to multiple abuse (21% reported two types; 15% three types, and 6% four types of maltreatment) (Nguyen, Dunne, & Le, 2010). Similar levels of multi-type maltreatment have been found in Malaysia (Choo, Dunne, Marret, Fleming, & Wong, 2011). Exposure to multiple forms of maltreatment increases the child’s risk of developing later mental health problems (Edwards, Holden, Felitti, & Anda, 2003; Holt, Finkelhor, & Kantor, 2007; Rikhye et al., 2008; Schneider, Baumrind, & Kimerling, 2007; Teicher et al., 2006; Turner, Finkelhor, & Ormrod, 2006; Zoroglu et al., 2003). In Australia, as in most countries, there has not been a comprehensive assessment of the health consequences of child maltreatment at the national level. CSA was the only form of maltreatment included as a risk factor in the Australian Burden of Disease 2003 study (Begg et al., 2008) and in the Global Burden of Disease 2010 study (GBD 2010) (Lim et al., 2012). The omission of other types of child maltreatment limits our understanding of how these complex phenomena relate to each other, and how co-occurrence of multiple types may influence the overall burden. This is also an important gap in the international literature (Fang et al., 2015). In addition, to date, studies estimating the economic burden of child maltreatment (Fang, Brown, Florence, & Mercy, 2012; Fang et al., 2015) have not adjusted for co-occurrence of maltreatment types. The present analysis aims to derive estimates of the burden of depressive disorders, anxiety disorders and intentional self-harm attributable to CSA, CPA, CEA and CN in Australia adjusting for co-occurrence of multiple forms of child maltreatment. It is hoped that this will lead to quantification of the burden attributable to all forms of child maltreatment in future iterations of the Australian and global burden of disease studies and inform estimation of the economic impact of child maltreatment.
Methods Exposure to child maltreatment was treated as a risk factor for disease and injury using counterfactual estimation and comparative risk assessment methods (Lim et al., 2012). This involved comparing the current local health status with the theoretical-minimum-risk exposure defined as a population not ever having been exposed to child maltreatment. Population attributable fractions (PAFs) were determined by the prevalence of ever having been exposed to these risk factors in the population and the relative risks (RR) of disease occurrence given exposure.
210
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
Risk Factor Definitions CSA was defined as the involvement in sexual activity that a child does not fully comprehend, is unable to give informed consent to, or for which the child is not developmentally prepared (Butchart et al., 2006). As per GBD 2010 (Lim et al., 2012), in this study CSA was defined as ever having experienced any form of CSA and hence would encompass all sub-types including unwanted non-contact, contact abuse, or intercourse (penetrative sexual abuse) in childhood. CPA was defined as the intentional use of physical force against a child that results in (or has a high likelihood of resulting in) harm to the child’s health, survival, development or dignity. This included hitting, beating, kicking, shaking, biting, strangling, scalding, burning, poisoning, and suffocating the child (Butchart et al., 2006). CEA involved both isolated incidents as well as a pattern of failure over time on the part of the parent or caregiver to provide a developmentally appropriate and supportive environment to the child. This included the restriction of movement, patterns of belittling, blaming, threatening, frightening, discriminating against or ridiculing and other non-physical forms of rejection or hostile treatment (Butchart et al., 2006; Hamarman & Bernet, 2000). CN included both isolated incidents as well as a pattern of failure over time on the part of a parent or caregiver to provide for the development and wellbeing of the child (where the parent is in a position to do so) in one or more of the following areas: health, education, emotional development, nutrition, shelter and safe living conditions (Butchart et al., 2006). Prevalence of Exposure We conducted a systematic review in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009) to identify studies that provided data on the prevalence of exposure to CSA, CPA, CEA, or CN in Australia. Three electronic databases (PubMed, EMBASE, and PsycINFO) were searched up to 15 August 2014 using the terms, ‘child*’ along with ‘maltreatment’, ‘sexual abuse’, ‘emotional abuse’, ‘neglect’, ‘physical abuse’ as well as ‘Australia*’ and ‘prevalence’ (Supplementary Data Text S1). This systematic review incorporated studies meeting the following inclusion criteria: (1) the study reported original, empirical published research; (2) the study presented lifetime prevalence data (proportion of the population who have ever experienced one or more of the four sub-categories of child maltreatment in childhood); (3) maltreatment was reported by victims rather than informants (health professionals, teachers, family members); (4) the study was conducted in Australia and was largely representative of the Australian population. We included studies with different definitions of childhood (ranging from under 15 to under 18 years) as well as studies where childhood was not defined (e.g. maltreatment occurs “in childhood”). Studies of child maltreatment conducted in justice system, high risk or clinical samples [for example participants with border-line personality disorder (Goldman, D’Angelo, DeMaso, & Mezzacappa, 1992) or a history of depressive disorders (Gladstone et al., 2004)] were excluded. Where the same data were reported across different publications, the most informative article was selected: for example, studies reporting sex- or age-specific prevalence estimates were selected over those providing combined estimates. If multiple studies were carried out in the same population, the first published study was utilized ensuring sample independence with the inclusion of every participant only once in the subsequent meta-analyses. Reference lists of selected studies were screened for any other relevant studies. In addition to the systematic review of electronic databases, we surveyed Australian official statistics websites to identify country-level data for prevalence of exposure. Two relevant reports and data sources (Australian Bureau of Statistics, 2005; Australian Institute of Health and Welfare, 2014) were identified and reviewed. The full-text of studies that appeared to meet the inclusion criteria was retrieved for closer examination and assessed for eligibility. A standardized data extraction sheet was developed, and data retrieved included publication details, methodological characteristics such as sex, age of respondents, sample size, sample type, state in which the study was conducted, definition of childhood, exposure measures, type of abuse and prevalence reported. Quality of studies was assessed using a tool for assessing risk of bias in prevalence studies (Hoy et al., 2012) and converted to a proportional quality score (the total quality score divided by the maximum score possible) (Supplementary Data Text S1 and Supplementary Table S1). In the Child Protection Australia 2012–13 report, unit-record level data has been made available for analysis for the first time, providing unique counts of children receiving child protection services in each jurisdiction (Australian Institute of Health and Welfare, 2014). Substantiated cases are based on the type of abuse that was considered most severe or most likely to place the child at risk (the ‘primary’ type of maltreatment). Although official data could be used to estimate prevalence of substantiated child maltreatment in Australia for the four maltreatment categories (presented in Supplementary Table S1), they were not included in subsequent meta-analyses of prevalence as they are not based on retrospective self-reports and not fully representative of all cases of child maltreatment. The Personal Safety Survey presented CPA and CSA prevalence estimates that could be included in the meta-analysis (Australian Bureau of Statistics, 2005). Meta-analysis of Prevalence Some studies presented prevalence of maltreatment for both sexes combined. These estimates were converted to sex specific estimates in a 2-step process. First a female-to-male ratio was derived from studies identified in the systematic
S.E. Moore et al. / Child Abuse & Neglect 48 (2015) 208–220
211
Table 1 Results of meta-analysis of prevalencea of child maltreatment in Australia. Type of maltreatment Childhood sexual abuse primary analysis Any childhood sexual abuseb Maleb Femaleb Childhood sexual abuse subgroup analyses Age of respondents (children
