Delineating disproportionality and disparity of Asian-Canadian versus White-Canadian families in the child welfare system

ObjectiveThis paper builds upon the analyses presented in three companion papers using data from the 2003 and 2008 cycles of the Canadian Incidence Study of Reported Child Abuse and Neglect (CIS-2003 and CIS-2008) and the Ontario Child Abuse and Neglect Data System (OCANDS) to examine disproportionality and disparity of child welfare involvement for Asian-Canadian children involved in the child welfare system.MethodsThis study used CIS-2008 data adjusted by Census child population data to examine rates per 1000 and three disparity indexes (population-based disparity index [PDI], decision-based disparity index [DDI], and maltreatment-based disparity index [MDI]) to determine the representation of child maltreatment investigations for Asian-Canadian versus White-Canadian children involved in the child welfare system. Logistic regression analyses were conducted to determine the odds of case closure for substantiated child maltreatment investigations, and whether Asian ethnicity remained significant while controlling for child demographics and household composition, case characteristics, and clinical concerns.ResultsThis study found that Asian-Canadian children were underrepresented in the child welfare system compared to White-Canadian children (13.9 per 1000 Asian children in the Canadian population vs. 36.1 per 1000 White children in the Canadian population). Child welfare involvement for Asian-Canadian children are almost 2 times more likely to close after an investigation than White-Canadian children. The three disparity indexes (PDI, DDI, MDI) showed substantially different results with respect to the representation of child maltreatment investigations involving Asian-Canadian versus White-Canadian children for physical abuse, sexual abuse, neglect, emotional maltreatment, and exposure to domestic violence.ConclusionsDisproportionality and disparity are complex phenomena. The variation in results derived from different methods of calculating representation suggests the need for greater clarity and consistency in the definitions and methodology in examining racial disparity in child welfare research. Some methodological considerations for future child welfare research with Asian-Canadian populations were discussed.     

Determinants of maltreatment substantiation in a sample of infants involved with the child welfare system

Children under one year of age are highly vulnerable to child maltreatment, which can lead to serious immediate and long-term physical and mental health consequences. Child welfare workers encounter unique challenges when assessing allegations of maltreatment involving infants. This study identifies correlates of maltreatment substantiation in a sample of 793 infants less than one year of age investigated by child welfare. It is based on data from the 2003 Canadian Incidence Study of Reported Child Abuse and Neglect (CIS-2003), a national study of reports of child maltreatment. The analysis consisted of two stages. First, bivariate analyses were conducted using unadjusted odds ratios and chi-square tests on variables known to be associated with substantiation. In the second stage, a logistic regression was performed to examine these associations in a multivariate model.Multivariate findings suggest that certain vulnerabilities present in the primary caregiver, such as substance abuse and few social supports, as well as the child's positive toxicology at birth are associated with substantiated infant maltreatment. Exposure to domestic violence, which is reported mostly by the police, and physical harm stemming from maltreatment are strongly associated with substantiation, whereas reports of physical abuse and neglect have an inverse relationship. The authors discuss these results and implications for policy and programs.     

Describing the context and nature of emotional maltreatment reports in children

The circumstances leading to reports of emotional maltreatment are potentially unique from those of other abuse types. This study utilized data from the National Child Abuse and Neglect Data System (NCANDS), which is a federally sponsored effort requiring the submission of case-level data on all children who were subject to maltreatment investigations by state child welfare agencies. The central aims of this study were to examine case-level characteristics of sole occurrences of emotional maltreatment Child Protective Service (CPS) reports in an effort to address how caregiver and child risk correlates affect the substantiation of these allegations. A cluster analysis was performed to explore how the risk assessment factors grouped together; secondarily, the cluster groupings were examined in a logistic regression model to measure factors associated with substantiation of these abuse reports. Four unique clusters emerged, with one cluster absent any risk assessment factors entirely and another cluster solely composed of domestic violence. Caregiver risk factors with the highest odds of predicting report substantiation included caregiver substance abuse, domestic violence, and caregiver learning disabilities.     

Disentangling substantiation: The influence of race, income, and risk on the substantiation decision in child welfare

Understanding the source of disparities found at decision-making points along the child welfare pathway is essential to understanding and addressing the overrepresentation of African American children. Although research has documented the existence of disparities, it has been less successful in identifying the explanatory factors behind them. Critiques of research examining these disparities have suggested that poverty is likely a stronger explanatory factor than race, yet analyses that include measures of poverty using data from child welfare systems have largely not been conducted. This study uses data from the Texas child welfare system to identify the factors contributing to disparities at the substantiation decision. Given the relationship between poverty and child maltreatment, the analyses control for the effect of family income, as well as other factors related to maltreatment, to better understand the effect of race on this decision-making point. Findings indicate that when family income is controlled, race is not a significant factor in the substantiation decision. However, when also controlling for caseworker perceptions of risk, race emerges as the stronger explanatory factor. This suggests not only an important relationship between race, income, and risk assessment, but also that disproportionality in the child welfare system is a complex phenomenon that cannot be explained by a single factor. These results further demonstrate that the effect of racial bias on decision-making remains an important consideration in understanding the overrepresentation of African American children.     

The Canadian Human Rights Tribunal on First Nations Child Welfare: Why if Canada wins, equality and justice lose

Repeated reports indicate that First Nations children on reserve receive less child welfare funding than other children in Canada despite the fact that First Nations children have higher child welfare needs. After the Government of Canada failed to implement two joint solutions to address the inequality, First Nations organizations in Canada filed a human rights complaint alleging that the Government of Canada is discriminating against First Nations children on the basis of race and national ethnic origin. This historic case is now before the Canadian Human Rights Tribunal and marks the first time that Canada has been held to account before a legal body for its current treatment of First Nations children and their families. This opinion article presents the facts leading up to the filing of the human rights case, the grass roots advocacy and legal processes after the complaint was filed, and the implications for: First Nations children, individuals from minority groups, and the moral fabric of the country if the Government of Canada wins the case.     

An exploratory study of drug-exposed infants: case substantiation and subsequent child maltreatment

This study explores factors related to drug-exposed infants' case substantiation and subsequent child maltreatment. Child protective services computerized administrative data (from January 1998 to October 2001) were obtained from an urban Nevada county. The data included 457 drug-exposed infant cases. Chi-square, t-test, one-way ANOVA, and logistic regression were used to analyze the data. Results indicate that: (1) drug-exposed infant case substantiation was related to the type of drug exposure and the unit to which the case was assigned, but not to the mother's ethnicity; and (2) subsequent maltreatment among drug-exposed infants was related to the mother's age and prior parental alcohol abuse, but not to the type of drug exposure, nor to the initial drug-exposed infant status of case substantiation. Implications for child welfare practice and research are discussed.     

Socioemotional Problems Among Young Children in Out-of-Home Care: A Comparison of Kinship and Foster Care Placement

This study examined socioemotional problems among children age 0 to 5 years in formal kinship and foster care across a northeastern state. Findings revealed that more than one third of children in both types of care showed socioemotional problems. More children were placed in kinship care when they lived in rural settings, were biracial, or spent time in the neonatal intensive care (NICU) after birth. Unique child and maltreatment characteristics predicted socioemotional problems scores for each placement type. Among children in kinship care, being biracial, spending time in the NICU, and being referred to child welfare for either caregiver substance abuse or neglect were related to socioemotional problems. When children were in foster care, spending time in the NICU, being referred to child welfare for child neglect, or being referred for intimate partner violence were all related to socioemotional problems. The results have implications for child welfare policy and practice, especially in light of increased efforts to place children with kin.     

Focus on the first years: Correlates of substantiation of child maltreatment for families with children 0 to 4

The overall purpose of this study was to examine the correlates of substantiation of child neglect or child abuse in children 0–48 months of age in five areas: (1) child characteristics, (2) parent characteristics, (3) home characteristics, (4) social characteristics, and (5) maltreatment characteristics. A second purpose was to compare the results of this study to the current body of child maltreatment research.This study uses an ex-post facto design. A random sample of cases from a large metropolitan area in the southwest was obtained. Families whose cases were substantiated for maltreatment were compared to families whose cases were not substantiated. The exposure of children to a dangerous environment and poor parental skill were predictive of the substantiation in the multivariate analysis. Parents with fewer parenting skills, such as the inability to effectively communicate, lack of impulse control and use of harsh discipline, were predictive of child maltreatment in the multivariate analysis. The bivariate analysis revealed additional 22 variables as correlates of the substantiation within the 5 factors.     

Child maltreatment dynamics among immigrant and U.S. born Latino children: Findings from the National Survey of Child and Adolescent Well-being (NSCAW)

Latino children are the largest and fastest growing ethnic minority group in the United States, and the proportion of child maltreatment victims who are of Latino ethnicity has been growing since 2000. However, our knowledge of the characteristics, maltreatment patterns, and risk factors associated with maltreatment among immigrant and U.S. born Latino children and their families has been incomplete. The goals of this study are to establish the national prevalence of immigrant and U.S. born Latino children who come to the attention of child welfare systems in the United States; to expand our knowledge regarding the role of nativity in child maltreatment patterns among Latinos; and to determine if disparities in child maltreatment patterns and risks exist among Latino families using data from the National Survey of Child and Adolescent Well-being. Despite considerable risks, data indicate that immigrant Latino children are slightly underrepresented among children who present to child welfare systems when compared to the general population. Moreover, no significant differences were observed between immigrant children and U.S. born children with regard to substantiation patterns. However, significant differences emerged between the two groups in risk factors and maltreatment type, warranting further investigation.     

Child welfare supervised children's participation in center-based early care and education

Research suggests that early care and education (ECE) services, particularly center-based ECE, may help prevent child maltreatment and also mitigate some of the negative developmental outcomes associated with child maltreatment. There is also preliminary evidence to suggest that ECE could reduce the likelihood that maltreatment allegations will be substantiated by child welfare authorities and/or result in children being placed in out-of-home care. However, little is known about rates of ECE participation among children receiving child welfare services, nor the factors that determine ECE participation for this population. Data from the first wave of the National Survey of Child and Adolescent Wellbeing II, a nationally representative sample of children referred to the United States (U.S.) child welfare system (CWS) for suspected maltreatment, were used to measure the frequency with which 0–5 year olds participate in center-based ECE. Additionally, logistic regression analyses explored the effects of maltreatment type, substantiation, and children's living arrangements (i.e., with parents, relatives, or foster parents) on this outcome, controlling for a range of child and family covariates associated with ECE participation in the general population. Results indicate that less than a third of 0–5 year olds receiving child welfare services in the U.S. are participating in center-based ECE. Among the various categories of maltreatment type measured, being reported to the CWS for suspected physical abuse was associated with decreased odds of center-based ECE participation; however, other types of maltreatment, substantiation, and living arrangement were unrelated to center-based ECE participation. These findings suggest that, despite recent efforts by the U.S. federal government to promote ECE participation for CWS-supervised children, the vast majority of young children in the U.S. CWS are not receiving center-based ECE, and physically abused children are particularly disadvantaged when it comes to accessing these services.     

Child safety at reunification: A case-control study of maltreatment recurrence following return home from substitute care

When considering reunification, child welfare caseworkers are faced with the difficult challenge of predicting which caretakers will be able to provide a safe environment for their children once they return home. Unfortunately, although an increasing number of studies have examined maltreatment recurrence during investigation and following case opening, little is known about the factors that predict maltreatment recurrence following reunification. Using a case-control design and information gathered from a child welfare administrative database and client case records, the current study examined the factors that predict short-term (i.e., within 60 days) maltreatment recurrence among 174 families with children returning home from their first stay in substitute care. From a variety of child, caretaker, placement, family environment, and service provision characteristics, seven variables uniquely added to the prediction of maltreatment recurrence: 1) child age, 2) caretaker mental illness, 3) number of placements, 4) type of placement, 5) length of time in placement, 6) number of children in the home at reunification, and 7) the interaction between household structure at reunification and the presence of siblings returned home with the index child. The implications of these findings for child welfare practice and future research are discussed in detail.     

Early Childhood Maltreatment: Substantiation related to Family Risk and Intervention Factors

Child maltreatment interventions had been targeted on intervening parental factors among risk families. However, limited knowledge was known about possible interaction effects between certain family risk factors and such intervention factors. The present study used existing Child Protective Service (CPS) data to explore interaction effects between two family risk variables (child maltreatment pattern and domestic violence) and three common intervention variables (quality of parent–child connection, caregiver knowledge and caregiver skill) on child maltreatment substantiation. Findings showed good quality of parent–child connection, caregiver knowledge, and caregiver skill significantly predicted lower risk of child maltreatment. However, the significant effects were found only among families of minor child maltreatment pattern and were greater for non-domestic violence families. In all, findings of this study suggested that different intervention emphasis might be needed to work with families of more severe risk issues.Szu-Yu Chen and Maria Scannapieco are affiliated with Center for Child Welfare, School of Social Work, University of Texas at Arlington, Arlington     

Directions for developmental screening in child welfare based on the ages and stages questionnaires

Many young children investigated for maltreatment have developmental problems qualifying them for early intervention services, yet only a portion of these children receive such services. To address this gap, all children ages 0-3 with child welfare (CW) substantiated maltreatment in Pennsylvania are screened for developmental and socioemotional problems using the Ages and Stages Questionnaires (ASQ). This study views screening results for over 500 children to address whether children's substantiation status, living situation, and administering worker as CW or early intervention (EI) predicts screening rates. Bivariate and logistic regression analyses were used. Results showed that 22% of children scored in the problem-range of at least one developmental area and 18% scored in the problem-range of social-emotional concerns warranting EI referral. Results of bivariate analyses showed that children who spent time in the NICU were more likely to have developmental concerns. Socioemotional concerns were related to child race, foster care living situation, child as the subject of the referral, and physical neglect. Older children and children whose referral involved lacking basic needs showed both developmental and socioemotional concerns. There were no significant differences in screening results of children with substantiated versus unsubstantiated maltreatment. Children whose screening was conducted by EI were more than three times as likely to show developmental concerns compared with those screened by CW. Child welfare workers need more support when conducting developmental screening, and policies that limit screening to children with substantiated maltreatment or to children in out-of-home care should be reconsidered.     

Disentangling over-representation of parents with disabilities in the child welfare system: Exploring child maltreatment risk factors of parents with disabilities

The study explores the risk factors for child maltreatment and self-reported child maltreatment among a population-based sample of parents with disabilities. Drawing on a nationally-representative, population-based data file that oversampled people of color, income-adjusted odds ratio tests were conducted to establish population differences among parents with and without limitations in activities of daily living (ADLs). Results suggest that parents with disabilities were more likely to report many of the risk factors associated with child maltreatment than parents without disabilities, including witnessing interpersonal violence as a child; experiencing violence, neglect or a foster care stay as a child; mood or substance use disorders; and engaging in or receiving interpersonal violence as an adult. Before controlling for income, parents with disabilities had only a negligibly higher rate of engaging in violence against their children. After controlling for income, parents with disabilities were 2.5 times more likely to engage in violence against their children. Parents with disabilities who did engage in violence against their children had greater amounts of some of the child maltreatment risk factors in comparison to parents with disabilities who did not engage in violence, particularly their own childhood experiences of maltreatment, witnessing of interpersonal violence as a child, childhood stays in foster care, and experiences with interpersonal violence as an adult. Findings add to the understanding of the risk factors for child maltreatment that are related to the collateral effects of having a disability, and through the use of income-adjusted data, help disentangle why parents with disabilities are over-represented in the child welfare system. The findings highlight the need for the child welfare system to increase its disability competence in working with both children and parents with disabilities.     

Community perceptions and foster care placement: A multi-level analysis

Signs of physical and social disorder in a community have the potential to influence the perceptions about community, which could, in turn, impact the decision making processes involved in child maltreatment investigations. While extant research has demonstrated a link between community context and child maltreatment rates, there is a dearth of research examining the potential relationship between social conditions in a community and the child welfare system's response to allegations of neglect. This study examined the relationship between child- and community-level characteristics and the decision to place a child in foster care. Two data sets were linked, one containing detailed observations of community conditions (e.g., cigarette and alcohol advertisements) and a second with individual child level information about child welfare investigations. Multilevel models allowed for the simultaneous understanding of child- and community-level characteristics. The results indicated strong associations between individual characteristics and foster care placement. The race or ethnicity of the child was a strong predictor of entry into foster care, and young children were much more likely to be placed in foster care than older children. Furthermore, community factors were also associated with placement into foster care. Specifically, the presence of litter and graffiti and the lack of security signage in a neighborhood appeared to impact the decision making process for foster care.     

Child Maltreatment, Abortion Availability, and Economic Conditions

Child maltreatment is a substantial problem in the U.S. yet has received relatively little attention from economists. This article examines the relationship between abortion availability and economic factors at the time children were conceived and subsequent measures of child maltreatment in the U.S. as well as the influence of contemporaneous economic conditions. Our measures of child maltreatment are state-level rates of child abuse and neglect reports, the fraction of children receiving social services, and child deaths and murders. The results indicate that legalized abortion for each successive cohort led to a decline in total reported incidents of child abuse and neglect of about 10 percent and a negative effect on the fraction of children receiving social services. Child deaths and murders are not related to abortion legalization. Medicaid funding restrictions are associated with an increase in substantiated reports of abuse and an increase in murders by relatives or parents; other post-legalization restrictions are not consistently associated with the various measures of child maltreatment. The effects of welfare benefits, average income, and unemployment rates are mixed.     

Deconstructing disproportionality: views from multiple community stakeholders

While the existence of racial disproportionality has been well documented, the causes of disproportionality are less clear. Studies identifying contributing factors have relied largely on analyses of state and national data sets, which may lack the robust data necessary to fully explain the factors related to this issue. Further, a limitation of existing research is the lack of data from the voice of those in communities affected by disproportionality. This study was designed to develop a deeper understanding of disproportionality from the views of multiple community stakeholders. Using a qualitative approach, data were collected to provide a greater depth of information that can be used alongside existing studies toward developing an enhanced understanding of disproportionality in child welfare.     

Professional judgements and decisions on placement in foster care and reunification in Portugal

This study aims to analyse professional judgements and decisions made in the Portuguese child care system in face of a case vignette of child maltreatment. Using the approach proposed by Benbenishty et al. [(2015). Decision making in child protection: An international comparative study on maltreatment substantiation, risk assessment and interventions recommendations, and the role of professionals’ child welfare attitudes. Child Abuse & Neglect, 49, 63–75. doi:10.1016/j.chiabu.2015.03.015], we assessed the decisions of professionals, such as to place the child in foster care or reunify her with her family, on the basis of a series of judgements (e.g. substantiation of alleged abuse and neglect, risk assessments) that are influenced by the characteristics of the case, the decision-making context, and mother's and child's wishes. We conclude that there are different approaches to the case based on different professionals' attitudes that can be classified in two groups: one more pro-removal and the other anti-removal. These groups presented different risk assessments and intervention recommendations, and their decisions where significantly influenced by the mother's and child's wishes. Furthermore, we have done comparisons with studies made in other countries, concluding that the country context can be an important factor that leads to different outcomes. Implications for both practice and research are presented.     

Comments on how child protective services investigators decide to substantiate mothers for failure-to-protect in sexual abuse cases

This commentary discusses the decisions of child protective service (CPS) investigators to substantiate mothers for failure to- protect (FTP) in child sexual abuse cases. Four areas are identified in which the scientific literature remains inadequate to fully inform child maltreatment researchers, CPS practitioners and child welfare policymakers on models and outcomes for FTP substantiation decisions. Attention is called to the need for more information about how differences in statutory definitions, common guidelines, and professional education and training influence FTP substantiation decisions. Current gaps in the knowledge base on FTP and nonoffending caregivers are highlighted. doi:10.1300/J070v15n04_05.