The use of risk assessment to predict recurrent maltreatment: a Classification and Regression Tree Analysis (CART)

Assessing the Predictive Validity of Risk Assessment Tools in Child Health and Well-Being: A Meta-Analysis

BACKGROUND/OBJECTIVES

Violence and harm to children's health and well-being remain pressing global concerns, with over one billion children affected annually. Risk assessment tools are widely used to support early identification and intervention, yet their predictive accuracy remains contested. This study aims to systematically evaluate the predictive validity of internationally used child risk assessment tools and examine whether the tools' characteristics influence their effectiveness.

METHODS

A comprehensive meta-analysis was conducted using 28 studies encompassing 27 tools and a total sample of 136,700 participants. A three-level meta-analytic model was employed to calculate pooled effect sizes (AUC), assess heterogeneity, and test moderation effects of tool type, length, publication year, assessor type, and target population. The publication bias was tested using Egger's regression and funnel plots.

RESULTS

Overall, the tools demonstrated moderate predictive validity (AUC = 0.686). Among the tool types, the structured clinical judgment (SCJ) tools outperformed the actuarial (AUC = 0.662) and consensus-based tools (AUC = 0.580), suggesting greater accuracy in complex decision-making contexts. Other tool-related factors did not significantly moderate the predictive validity.

CONCLUSIONS

SCJ tools offer a promising balance between structure and professional judgment. However, all tools have inherent limitations and require careful contextual application. The findings highlight the need for dynamic tools integrating risk and needs assessments and call for practitioner training to improve tool implementation. This study provides evidence-based guidance to inform the development, adaptation, and use of child risk assessment tools in global child protection systems.

Recurrence of Substantiated Maltreatment Reports between Low-Income Parents With Disabilities and Their Propensity-Score Matched Sample Without Disabilities

Using the Longitudinal Studies of Child Abuse and Neglect dataset, 127 low-income parents with disabilities and a propensity score matched sample of 254 parents without disabilities were compared for the rates of repeated substantiated child maltreatment allegations and potentially distinct risks for substantiated child maltreatment recurrence. The number of substantiated child maltreatment allegations was not significantly higher for low-income parents with disabilities (M = 1.17, SD = 1.83) than their matched sample (M = .93, SD = 1.44) (t = -1.29, p = .197). Findings from the negative binomial regression indicated that parental disability was also not a significant predictor for repeated substantiated child maltreatment allegations among low-income parents after controlling other risk factors (Exp(B) = 1.16, p < .353). In both groups, black parents were more likely to have repeated substantiated maltreatments than white parents. For parents with disabilities, being an older parent, receiving SNAP benefits, having a daughter, and having a child in continued out-of-home care significantly increased the risk for repeated substantiations while having a GED or higher education degree and living with a larger number of family decreased the risk. For parents without disabilities, family instability was the only additional risk factor for repeated maltreatment substantiations.

Predictors of Maternal Recidivism in the Child Protection System

Child maltreatment recidivism is typically measured and studied at the individual level. Conditions that give rise to child abuse and neglect, however, typically affect multiple children in a given family. In the current study, we estimated maltreatment recidivism at the maternal level and examined its risk as a function of maternal sociodemographic characteristics that may change over time. Using linked administrative records, we identified a subset of first-time mothers in California whose first child was reported to the child protection system (CPS) between birth and age 5 and who then gave birth to another child (n = 14,715). Following the firstborn child's CPS reporting, nearly half of these mothers (43.3%) were re-reported concerning the non-firstborn children during the first 5 years of the child's life. Risk factors consistently documented across births were associated with a heightened risk of maternal CPS recidivism. Our study advances an understanding of the full extent of maltreatment recidivism by broadening the focus from individual children.

The CPS Workers' Child Removal Decision in Cases of Domestic and Witnessed Violence

This study aimed to explore maternal and paternal risks and protective factors that may influence the Child Protection Service (CPS) workers' child removal decision in case of domestic and witnessed violence. In all, 218 case reports of Italian CPSs were retrospectively analyzed through the Protocol of Risk and Protective Factors. The sample was then split up into two groups on the basis of the CPS professionals' placement decision after the investigation (child removal decision versus parents support and monitoring intervention). Two statistical approaches were used to identify the patterns of risk and protective factors associated with maternal and paternal assessments: logistic regression models and decision tree analysis. Results showed that mothers who are victims of Intimate Partner Violence experienced the child removal in about half of the cases, while fathers showed a higher removal rate. Differences emerged between mothers' and fathers' risk profiles, suggesting that workers attributed a different weight to some factors depending on whether they concerned the mother or the father. Only the proximal risk factor poor empathy skills was significant for both mothers and fathers. For the mothers' group, one of the most important factors was the presence of direct forms of child maltreatment in addition to witnessed violence, while for fathers' group the drug abuse emerged a crucial relevant proximal risk factor.

Predictive Risk Modeling for Recurrence of Child Maltreatment Using Cases from the National Child Maltreatment Data System in Korea: Exploratory Data Analysis Using Data Mining Algorithm

The purpose of this study was to explore and identify patterns of risk predictors of maltreatment recurrence using predictive risk modeling (PRM). This study used the administrative dataset from the National Child Maltreatment Information System recorded by Korean CPS (Child Protective Service) workers. The information, including recurrent maltreatment, was collected in 2012; then, those reported cases were followed for 2 years through 2014. The data included information about child, family, caregiver, maltreatment, and service characteristics and consisted of male (50.22%) and female (49.78%) children with an average age of 9 years (n = 4319). We examined the association of risk factors with recurrence using conditional inference trees (CTREE): a tree-based data mining algorithm for classification that allows the exploration of the interconnection between hypothesized risk factors. Study findings showed that a history of prior CPS involvement was the first decision point in the decision tree structure of recurrence. The effect of other risk factors depended on the presence of prior CPS involvement. In the absence of prior CPS involvement, cases with (a) a single-parent status and (b) a caregiver's alcohol abuse living in other types of households (two-parent households, kinship care, and children without parents) were associated with recurrence. In the presence of prior CPS involvement, cases with out-of-home care or others (long- or short-term foster care and emergency placement) in the final decision of child placement (a) where in-home care in the initial decision of child placement within the presence of physical abuse and (b) where social isolation without physical abuse was related to recurrence. Cases with (a) a male caregiver and (b) a female caregiver with social isolation and without social isolation yet employed were at high risk for recurrence under the circumstance of in-home care in the final decision of child placement. This exploratory study found multiple connections among the factors in the prediction of recurrence. The CTREE helps unravel the complexity embedded in maltreatment recurrence by capturing its patterns. This information can deepen our knowledge of associations between risk factors in the prediction of recurrence and be used as a reference to inform child maltreatment policy and prevention.

Differentiating Perpetrators of Intimate Partner Violence From Other Violent Offenders Using a Statistical Learning Model: The Role of Cognition and Life History Variables

Intimate partner violence (IPV) is a widespread crime that victimizes over 4-million women per year in the United States and results in significant monetary cost and unmeasured physical and psychological consequences for victims. Specialized IPV offender treatment programs demonstrate limited effectiveness, which may be due to an insufficient understanding of the factors that differentiate between IPV perpetrators and non-IPV violent offenders. In this study, we utilized classification and regression tree (CART) analysis to identify combinations of factors that best discriminate IPV perpetrators from non-IPV violent offenders. We also compared cognitive abilities between IPV perpetrators and non-IPV violent offenders using standardized neurocognitive tests. CART analysis presented two pathways for identifying offenders as IPV perpetrators: (a) extensive nonviolent criminal history and (b) moderate-to-severe expression of interpersonal traits of psychopathy without attentional deficits. In addition, a third pathway identified non-IPV violent offenders: (c) low levels of interpersonal psychopathic traits and no history of neurodevelopmental diagnosis. IPV perpetrators demonstrated intact cognition relative to test norms, and study groups did not significantly differ on cognitive performance. These findings suggest that individuals with multiple arrests for nonviolent crime or individuals with interpersonal traits of psychopathy without attentional difficulties may be at enhanced risk for IPV perpetration.

Re-referrals to Child Protective Services for children born to young parents

BACKGROUND

Children of teenage parents are at increased risk for child maltreatment but there is limited information about the risks teenage parenthood poses for maltreatment recurrence after prior Child Protective Service (CPS) involvement.

OBJECTIVE

Determine whether children born to teenage parents are at increased risk of maltreatment re-referral to CPS compared to children born to older parents.

PARTICIPANTS AND SETTING

Children under 6 years with substantiated reports of parental maltreatment between 2002 and 2018 were identified from National Child Abuse and Neglect Data System (NCANDS) data.

METHODS

Youngest known parental age at child's birth, in years, was categorized as <20, 20-24, 25-29, or ≥30 (referent group). The number of re-referrals within 5 years was the outcome of interest. Negative binomial regression analyses assessed whether being born to a teen parent was associated with a greater risk of re-referral.

RESULTS

In a study population of 2,680,961 children, having a teen parent, one aged 20-24 years, or 25-29 years at birth was associated with 19% (95% CI 1.18-1.19), 16% (95% CI 1.16-1.17), and 11% (95% CI 1.11-1.12) greater risk of re-referral compared to having a parent aged 30 years or older adjusting for child age, gender, race-ethnicity, state, and initial maltreatment type.

CONCLUSION

The small, but significant increased risk for re-referral to CPS for child maltreatment in young children with at least one teenage or young parent at the time of birth compared to children with older parents suggests that specifically targeting young parents with appropriate services when initially referred to CPS may help reduce the risk of maltreatment recurrence.

A latent class analysis of modifiable risk factors associated with child maltreatment re-reporting and recurrence

BACKGROUND

Maltreatment re-reporting and recurrence represent missed opportunities for prevention and early intervention in child welfare settings.

OBJECTIVES

This study identified latent classes of risk among families who experienced a child maltreatment re-report or maltreatment recurrence within 12-months of initial case closure.

PARTICIPANTS AND SETTING

Administrative child welfare data from a large urban county were subject to secondary analysis. Samples included children who experienced a maltreatment re-report (n = 4390), and children who experienced a second maltreatment substantiation (n = 694).

METHODS

Five modifiable risk factors (i.e., mental health, substance abuse, domestic violence, disability, parenting challenges) were extracted from the initial investigation and subject to latent class analysis. Case characteristics (i.e., age, gender, race, ethnicity, maltreatment type) were then compared across the latent classes in a post-hoc analysis.

RESULTS

Re-report classes were characterized by (1) "Few Identified Challenges" (56%, n = 2458), (2) "Mental Health and Domestic Violence Challenges" (26%, n = 1133), and (3) "Substance Abuse, Domestic Violence, Mental Health, and Parenting Challenges" (18%, n = 790). Re-report classes differed according to child age, race, ethnicity, neglect and physical abuse allegations. Recurrence classes were characterized by (1) "Domestic Violence Challenges" (48%, n = 333), (2) "Mental Health Challenges" (15%, n = 104), and (3) "Domestic Violence, Mental Health, and Parenting Challenges" (37%, n = 257). Recurrence classes differed according to child race and age.

CONCLUSIONS

Findings underscore the complex and co-occurring nature of maltreatment risk, and provide insights to strengthen assessment and intervention practices to reduce repeated contacts with child welfare systems.

Quantifying Severity of Maltreatment, Adversity, and Trauma From Child Protective Services Case Record Files

Children referred to child protective services (CPS) for allegations of abuse or neglect often have diverse experiences of maltreatment, adversity, and trauma. Severity of these experiences is associated with greater mental health impairment and increased risk of revictimization and other adversities. Although aspects of these experiences are often captured during CPS investigations and stored in case records as narrative documents, much of this information is underutilized in estimating risk and service planning. The current study extracted case record information from a randomly selected sample of 100 families, with 150 children referred to CPS during a 12-month period. The Yale-Vermont Adversity in Childhood Scale (Y-VACS) was applied to extracted information for quantifying severity of various forms of childhood maltreatment, adversity, and trauma. Study aims were to examine (a) the scope and severity of maltreatment, adversity, and trauma types and their associations; (b) linkages between severity and CPS allegation types and outcomes; and (c) the utility of severity in predicting new allegations of abuse or neglect within 12 months of referral. Results indicated feasibility in quantifying severity of maltreatment and other adversities from case record information and revealed associations between adversity severity and CPS allegation types and outcomes. Severity of psychological intimate partner violence and neglect were predictive of new allegations of abuse or neglect within 12 months of referral. Findings support moving beyond an incident-based CPS strategy to one that better incorporates case record information to assess risk.

Engaging Families in Voluntary Prevention Services to Reduce Future Child Abuse and Neglect: a Randomized Controlled Trial

Research on secondary prevention is mixed as to whether referrals to evidence-based practices and other community interventions reduce maltreatment recidivism. This evaluation sought to rigorously assess whether a community service-linkage model to evidence-based programs decreased subsequent reports of abuse and neglect among families reported to the child protection system (CPS). Between 2016 and 2018, administrative CPS records were used to identify families with at least one child under 5 years of age recently reported for maltreatment in Orange County, California, but where no follow-up services were mandated. Families were randomized to one of two community outreach and engagement conditions (i.e., targeted provider outreach vs. treatment-as-usual). Re-reporting outcomes were tracked in administrative records for a period of 12 months. A total of 4873 families met eligibility criteria, with 2231 (45.0%) randomized to the intervention arm and 2642 (54.0%) families to the control group. Among families in the intervention group, 811 (36.4%) were successfully contacted by a community provider. Of those families contacted, 145 (18.0%) agreed to participate in voluntary services. Overall, we did not detect any significant difference in re-reporting between our treatment and control groups (P = .433). Likewise, no significant differences in re-reporting were observed when we stratified the treatment group by levels of outreach and service engagement. Findings indicate low levels of successful engagement of families in voluntary services. Notwithstanding the absence of significant findings, this study demonstrates the potential for using community-based RCTs and administrative records to rigorously evaluate secondary prevention programs in the child protection system.

Identifying children at high risk for recurrence child abuse

PURPOSE

To evaluate the characteristics of abused children, families and abusive event and to identify risk factors associated with recurrence of child abuse.

METHODS

Retrospective data from 133 children aged between 2 months to 15 years old who were diagnosed as abuse between year 2002 and 2017. Thirteen items related to characteristics of the child, families, abusive event were selected. These factors were analyzed by multivariate logistic regression model for association with repeated child abuse.

RESULTS

Total of 133 subjects with average age of 5.25 ± 4.65 years old. There were 54 cases (40.60%) reported of repeated abuse. Majority of repeated abuse type in this study was physical abuse (73.68%). Most perpetrators were child's own parents (45.10%). Factors associated with increased risk of repeated abuse were child age 1-5 years old (AOR = 4.95/95%CI = 1.06-23.05), 6-10 years old (AOR = 6.80/95%CI = 1.22-37.91) and perpetrator was child's own parent (AOR = 21.34/95%CI = 3.51-129.72). Three cases of mortality were found with single-visit children and one case in recurrence. Most of death cases were children less than 1-year-old with average age of 7 months. Causes of death were subdural hematoma with skull and ribs fracture.

CONCLUSIONS

Identifying risk factors for repeated child abuse help in recognizing child at risk to provide prompt intervention. This study found two factors associated with higher risk of abuse recurrence: child age 1-10 years old and abusive parents. Children who presented with these risk factors should be recognized and intensively monitored.

Risk factors for nonaccidental burns in children

Background

The relative influences of baseline risk factors for pediatric nonaccidental burns have not been well described. We evaluated baseline characteristics of pediatric nonaccidental burn patients and their primary caretakers.

Methods

A single-center retrospective cohort study was conducted of pediatric (age < 17) burn patients from July 1, 2013, to June 30, 2018. The primary outcome was nonaccidental burn, defined as burn secondary to abuse or neglect as determined by the inpatient child protection team or Child Protective Services. Univariate and multivariate analyses were performed.

Results

Of 489 burn patients, 47 (9.6%) suffered nonaccidental burns. Nonaccidental burn patients more frequently had a history of Child Protective Services involvement (48.9% vs 9.7%, P < .001), as did their primary caretakers (59.6% vs 10.9%, P < .001). Non-Hispanic black children had higher rates of Child Protective Services referral (50.7% vs 26.7%, P < .001) and nonaccidental burn diagnosis (18.9% vs 5.6%, P < .001) than children of other races/ethnicities. On multivariate analysis, caretaker involvement with CPS (odds ratio 7.53, 95% confidence interval 3.38-16.77) and non-Hispanic black race/ethnicity (odds ratio 3.28, 95% confidence interval 1.29-8.36) were associated with nonaccidental burn.

Conclusion

Caretaker history of Child Protective Services involvement and non-Hispanic black race/ethnicity were associated with increased odds of pediatric nonaccidental burn. Prospective research is necessary to determine whether these represent true risk factors for nonaccidental burn or are the result of other confounders, such as socioeconomic status.

Latent class analysis risk profiles: An effective method to predict a first re-report of maltreatment?

Recurrence of child maltreatment is a significant concern causing substantial individual, family and societal cost. Variable-based approaches to identifying targets for intervention may not reflect the reality that families may experience multiple co-occurring risks. An alternative approach was tested using baseline data from the National Survey of Child and Adolescent Well-being (NSCAW) I and II to develop Latent Class Analysis models of family risk classes using variables derived from prior studies of re-reporting. The samples were collected approximately 10 years apart offering a chance to test how the approach might be impacted by demographic or policy shifts. The association between baseline classes and later re-reports was tested using both samples. A two-class model of high versus low presence of baseline risk resulted that was strongly associated with later likelihood of re-report and results were relatively stable across the two studies. Person-centered approaches may hold promise in the early identification of families that require a more comprehensive array of supports to prevent re-reports of maltreatment.

Repeat reports among cases reported for child neglect: A scoping review

BACKGROUND

In the United States (US), child welfare policy prioritizes prevention of future harm (e.g., repeat reports) after a report of maltreatment. The majority of reports include some form of child neglect, but no prior review of the recurrence literature has focused on neglect.

OBJECTIVE

This review sought to help guide future research, policy and practice by summarizing recurrence findings related to child neglect with attention to the broader ecological context in which maltreatment occurs.

PARTICIPANTS

The final review included 34 US studies of maltreatment recurrence. Twenty-eight studies compared child neglect with at least one other form of maltreatment and six studies examined recurrence among neglect cases.

METHODS

Eleven online databases were searched to locate relevant empirical studies. This review attended specifically to contextualizing findings according to other modifiable factors as well as methodological variation. A scoping review approach was used to summarize findings.

RESULTS

Of the 28 studies comparing neglect to other types of maltreatment, 14 found increased risk for neglect, 12 found no association, and two reported a lower risk. When significant, the effect size ranged from 10% to over three times higher risk for neglect. Poverty or material need was the most commonly included control (15 studies), with two thirds finding that lower resource families had higher risk.

CONCLUSION

Methodological variability across studies confounds current ability to guide practice or policy. More research is needed that can replicate and extend findings with comparable samples and model specifications that take into account the regional and policy context.

Family profiles of maltreated children in Singapore: A latent class analysis

Children who enter the child protection system often have complex family problems and have experienced early adverse experiences. Using latent class analysis, this study aimed to identify family classes of child protection cases in Singapore, to ascertain the prevalence of these family classes, and to test the association of family class membership to subsequent recurrence of harm. A sample of 440 cases who entered the Child Protective Service in Singapore was analyzed based on eight familial factors on the household and caregiver levels. A four-class solution was found to demonstrate the best fit: (a) the large household group was intergenerational and majority lived with extended family members, (b) the harsh parenting group showed high levels of parenting problems and the caregiver justifying his abuse/neglect, (c) the high criminality group had high levels of caregiver substance abuse and caregiver arrest and incarceration history, and (d) the low disadvantage group rated low on all the familial factors. A Cox Regression revealed that in comparison to the low disadvantage group, the harsh parenting group was twice as likely to have recurrence of harm. There were also differences across family classes with regard to age at entry into child protection, gender composition and abuse types. The findings and practice implications are discussed.

The use of model based recursive partitioning as an analytic tool in child welfare

Child welfare agencies are tasked with investigating allegations of child maltreatment and intervening when necessary. Researchers are turning to the field of predictive analytics to optimize data analysis and data-driven decision making. To demonstrate the utility of statistical algorithms that preceded the current predictive analytics, we used Model Based (MOB) recursive partitioning, a variant of regression analysis known as decision trees, on a dataset of cases and controls with a binary outcome of serious maltreatment (defined as hospitalization or death). We ran two models, one which split a robust set of variables significantly correlated with the outcome on the partitioning of a proxy variable for environmental poverty, and one which ran the same variable set partitioned on a variable representing confirmed prior maltreatment. Both models found that what most differentiated children was spending greater than 2% of the timeframe of interest in foster care, and that for some children, lack of Medicaid eligibility almost doubled or tripled the odds of serious maltreatment. We find that decision trees such as MOB can augment risk assessment tools and other data analyses, informing data-driven program and policy decision making. We caution that decision trees, as with any other predictive tool, must be evaluated for inherent biases that may be contained in the proxy variables and the results interpreted carefully. Predictive analytics, as a class, should be used to augment, but not replace, critical thinking in child welfare decision making.

Predicting child maltreatment: A meta-analysis of the predictive validity of risk assessment instruments

Risk assessment is crucial in preventing child maltreatment since it can identify high-risk cases in need of child protection intervention. Despite widespread use of risk assessment instruments in child welfare, it is unknown how well these instruments predict maltreatment and what instrument characteristics are associated with higher levels of predictive validity. Therefore, a multilevel meta-analysis was conducted to examine the predictive accuracy of (characteristics of) risk assessment instruments. A literature search yielded 30 independent studies (N=87,329) examining the predictive validity of 27 different risk assessment instruments. From these studies, 67 effect sizes could be extracted. Overall, a medium significant effect was found (AUC=0.681), indicating a moderate predictive accuracy. Moderator analyses revealed that onset of maltreatment can be better predicted than recurrence of maltreatment, which is a promising finding for early detection and prevention of child maltreatment. In addition, actuarial instruments were found to outperform clinical instruments. To bring risk and needs assessment in child welfare to a higher level, actuarial instruments should be further developed and strengthened by distinguishing risk assessment from needs assessment and by integrating risk assessment with case management.

Predictors of substantiated re-reports in a sample of children with initial unsubstantiated reports

Many children with unsubstantiated reports of child abuse and neglect repeatedly return to the child protection system, indicating that unsubstantiated reports may represent actual child maltreatment or risk for future maltreatment. Identifying patterns of re-reporting and predictors that may be associated with later substantiated re-reporting could help to identify children who are very likely to be maltreated. This knowledge may guide the development of policies and interventions to prevent further maltreatment and the risk for re-reports. The aims of this study were to: (1) measure the period between the time of the initial reports that were not substantiated and the time of first substantiated re-reports; and (2) identify factors associated with the risk of later substantiated re-reporting. The study analyzed secondary data from the Longitudinal Studies on Child Abuse and Neglect (LONGSCAN) through survival analysis. Of the 378 children with initially unsubstantiated reports, 81% were re-reported, of which almost two-thirds were substantiated. Children who were younger, non-white, and had caregivers with more depressive symptoms were at increased risk of a substantiated re-report. Among those that were later substantiated, 20% were substantiated within one year. Findings suggest that targeted preventative services should be developed and provided for families who are reported for the first time, even if not substantiated.

Prevalence and Determinants of Preterm Birth in Tehran, Iran: A Comparison between Logistic Regression and Decision Tree Methods

Objectives

Preterm birth (PTB) is a leading cause of neonatal death and the second biggest cause of death in children under five years of age. The objective of this study was to determine the prevalence of PTB and its associated factors using logistic regression and decision tree classification methods.

Methods

This cross-sectional study was conducted on 4,415 pregnant women in Tehran, Iran, from July 6–21, 2015. Data were collected by a researcher-developed questionnaire through interviews with mothers and review of their medical records. To evaluate the accuracy of the logistic regression and decision tree methods, several indices such as sensitivity, specificity, and the area under the curve were used.

Results

The PTB rate was 5.5% in this study. The logistic regression outperformed the decision tree for the classification of PTB based on risk factors. Logistic regression showed that multiple pregnancies, mothers with preeclampsia, and those who conceived with assisted reproductive technology had an increased risk for PTB (p < 0.05).

Conclusion

Identifying and training mothers at risk as well as improving prenatal care may reduce the PTB rate. We also recommend that statisticians utilize the logistic regression model for the classification of risk groups for PTB.

What Happened Next: Interviews With Mothers After a Finding of Child Maltreatment in the Household

Child Protective Services (CPS) identifies over 700,000 victims of child maltreatment in the United States annually. Research shows that risk factors for these children may persist despite CPS intervention. Mothers have unique and often untapped perspectives on the experiences and consequences of CPS intervention that may inform future practice. We explored these perspectives through interviews with 24 mothers after a first-time CPS finding of maltreatment not resulting in out-of-home placement. Male partners were primary perpetrators in 21 cases, with mothers or sitters identified as perpetrators in remaining cases. Data were analyzed using grounded theory. Mothers described risk factors or Roots of maltreatment prior to CPS involvement and reported variable experiences with Recognition of and Response to maltreatment. Divergent Outcomes emerged: I Feel Stronger and We're No Better These findings provide an understanding of household experiences around child maltreatment that may support practice and policy changes to improve outcomes for vulnerable children.

Development of a prediction model for child maltreatment recurrence in Japan: A historical cohort study using data from a Child Guidance Center

To develop a prediction model for the first recurrence of child maltreatment within the first year after the initial report, we carried out a historical cohort study using administrative data from 716 incident cases of child maltreatment (physical abuse, psychological abuse, or neglect) not receiving support services, reported between April 1, 1996 through March 31, 2011 to Shiga Central Child Guidance Center, Japan. In total, 23 items related to characteristics of the child, the maltreatment, the offender, household, and other related factors were selected as predictive variables and analyzed by multivariate logistic regression model for association with first recurrence of maltreatment. According to the stepwise selection procedure six factors were identified that include 9-13year age of child (AOR=3.43/95%CI=1.52-7.72), <40year age of the offender (AOR=1.65/95%CI=1.09-2.51), offender's history of maltreatment during childhood (AOR=2.56/95%CI=1.31-4.99), household financial instability or poverty (AOR=1.64/95%CI=1.10-2.45), absence of someone in the community who could watch over the child (AOR=1.68/95%CI=1.16-2.44), and the organization as the referral source (AOR=2.21/95%CI=1.24-3.93). Using these six predictors, we generated a linear prediction model with a sensitivity and specificity of 45.2% and 82.4%, respectively. The model may be useful to assess the risk of further maltreatment and help the child and family welfare administrations to develop preventive strategies for recurrence.

Using a pattern-centered approach to assess sexual risk-taking in study abroad students

OBJECTIVES

The purpose of this study was to examine the impact of several potential factors related to sexually risky behaviors in study abroad students. The authors utilized a pattern-centered analysis to identify specific groups that can be targeted for intervention.

PARTICIPANTS

The sample consisted of 173 students who studied abroad in a variety of international locations for an average of 4 months.

METHODS

Participants completed questionnaires informed by the Triandis Theory of Interpersonal Behavior that have been predictive of risky sex in traditional traveling environments.

RESULTS

The analyses revealed 3 different pathways for risky sexual behavior: Environmental involvement, historical condom use, and intentions to engage in risky sex.

CONCLUSION

These findings can be used for identification of specific high-risk groups of students who can be targeted for predeparture prevention programs.

Risk factors for child maltreatment recurrence: An updated systematic review

OBJECTIVES

Children who have been maltreated are at increased risk of further maltreatment. Identification of those at highest risk of further maltreatment is a priority for professionals working in child protection services. The current study is intended to consolidate and expand on previous work on recurrence of child maltreatment. It has sought to identify risk factors for maltreatment recurrence in the recent literature in the expectation that this may help in the practical identification of children at risk.

METHODS

We conducted a systematic review of cohort studies published between 2003 and 2009, identifying factors associated with maltreatment recurrence in children. Studies included demonstrated differing levels of substantiation of maltreatment.

RESULTS

Fifteen studies met inclusion criteria but showed significant heterogeneity, varying in setting, recruitment of subjects, types of maltreatment considered and length of follow-up. Previous findings were replicated and expanded in the current study in relation to a range of factors, including rates of maltreatment recurrence, maltreatment types, frequency of previous episodes of maltreatment, child and family considerations, home environment and service provision. Factors were identified irrespective of level of maltreatment substantiation.

CONCLUSION

This study provides further systematic evidence of the existence of a number of factors associated with child maltreatment recurrence. It points to the possibility of practical application of its findings within the wider context of decision making in child protection services, with the ultimate aim of reducing recurrence of maltreatment in individual cases.

Predictive analytics and child protection: constraints and opportunities

This paper considers how predictive analytics might inform, assist, and improve decision making in child protection. Predictive analytics represents recent increases in data quantity and data diversity, along with advances in computing technology. While the use of data and statistical modeling is not new to child protection decision making, its use in child protection is experiencing growth, and efforts to leverage predictive analytics for better decision-making in child protection are increasing. Past experiences, constraints and opportunities are reviewed. For predictive analytics to make the most impact on child protection practice and outcomes, it must embrace established criteria of validity, equity, reliability, and usefulness.

Critical success factors in awareness of and choice towards low vision rehabilitation

PURPOSE

The goal of the current study was to examine the critical factors indicative of an individual's choice to access low vision rehabilitation services.

METHODS

Seven hundred and forty-nine visually impaired individuals, from the Montreal Barriers Study, completed a structured interview and questionnaires (on visual function, coping, depression, satisfaction with life). Seventy-five factors from the interview and questionnaires were entered into a data-driven Classification and Regression Tree Analysis in order to determine the best predictors of awareness group: positive personal choice (I knew and I went), negative personal choice (I knew and did not go), and lack of information (Nobody told me, and I did not know).

RESULTS

Having a response of moderate to no difficulty on item 6 (reading signs) of the Visual Function Index 14 (VF-14) indicated that the person had made a positive personal choice to seek rehabilitation, whereas reporting a great deal of difficulty on this item was associated with a lack of information on low vision rehabilitation. In addition to this factor, symptom duration of under nine years, moderate difficulty or less on item 5 (seeing steps or curbs) of the VF-14, and an indication of little difficulty or less on item 3 (reading large print) of the VF-14 further identified those who were more likely to have made a positive personal choice. Individuals in the lack of information group also reported greater difficulty on items 3 and 5 of the VF-14 and were more likely to be male.

CONCLUSIONS

The duration-of-symptoms factor suggests that, even in the positive choice group, it may be best to offer rehabilitation services early. Being male and responding moderate difficulty or greater to the VF-14 questions about far, medium-distance and near situations involving vision was associated with individuals that lack information. Consequently, these individuals may need additional education about the benefits of low vision services in order to make a positive personal choice.

Risk factors for recurrent injuries in victims of suspected non-accidental trauma: a retrospective cohort study

Background

Many children who are victims of non-accidental trauma (NAT) may be repeatedly evaluated for injuries related to maltreatment. The purpose of this study was to identify risk factors for repeated injuries in children with suspected NAT.

Methods

We conducted a retrospective cohort study using claims data from a pediatric Medicaid accountable care organization. Children with birth claims and at least one non-birth related claim indicating a diagnosis of NAT or skeletal survey in 2007–2011 were included. Recurrent events were defined as independent episodes of care involving an urgent/emergent care setting that included a diagnosis code specific for child abuse, a CPT code for a skeletal survey, or a diagnosis code for an injury suspicious for abuse. Cox proportional hazards models were used to examine risk factors for recurrent events.

Results

Of the 1,361 children with suspected NAT, a recurrent NAT event occurred in 26% within 1 year and 40% within 2 years of their initial event. Independent risk factors for a recurrent NAT event included a rural residence, age < 30 months old, having only 1 or 2 initially detected injuries, and having a dislocation, open wound, or superficial injury at the previous event (p ≤ 0.01 for all).

Conclusions

Over 25% of children who experienced a suspected NAT event had a recurrent episode within one year. These children were younger and more likely to present with “minor” injuries at their previous event.

A developmental approach to the risk of a first recurrence in child protective services

The objective of this study is to estimate the risk of a first recurrence over a five-year period following initial child protective services (CPS) intervention and identify the characteristics associated with the risk of recurrence for three different age groups. Recurrence is defined as the first substantiated report within the observation period after initial services have ended. The study involved a cohort of 25,897 Quebec children who received postinvestigation services for the first time and whose cases were closed between 2005 and 2009. Survival analysis was used to estimate the five-year risk of recurrence and Cox regression to model the risk of recurrence for three age groups. The covariates introduced into the regression analyses were characteristics of the child and initial services. The risk of recurrence in the five years following termination of initial CPS services was 36% for the entire cohort and varied depending on the child's age at the time of case closure. Children aged 6-11 when their cases were closed had the highest risk of recurrence. Although Aboriginal descent and prior CPS investigations have a consistent effect on the risk of recurrence in all three age groups, the effects of other covariates, such as out-of-home placement and court involvement, vary or are even reversed, depending on the child's age. These findings highlight the need to adopt a differential approach that takes into account the child's age, both in the provision of protective services and in research involving the population receiving such services.

The effect of parental intellectual disability status on child protection service worker decision making

BACKGROUND

There is evidence to suggest that parents with an intellectual disability (ID) constitute a higher proportion of child-protective services (CPS) cases than would be expected based on the prevalence of ID in the general population. Researchers have suggested that the stereotypic assumptions and expectations that CPS workers have about parents with an ID might influence decisions and responses made to such parents. This study examined whether parental ID (having an ID vs. not) had an effect on CPS workers' emotional reactions, attributions and decisions about risk to the child, whether to remove the child and workers' general willingness to help the parent.

METHOD

Two hundred and twelve CPS workers read vignettes describing parents who were labelled as either having or not having an ID. Workers responded to the vignettes by making ratings of their emotional reactions, attributions and decisions regarding risk, removal and helping.

RESULTS

CPS workers made significantly higher ratings of pity, willingness to help and risk for parents with an ID than for parents without an ID. Lower ratings of anger and disgust were found for parents with an ID than for parents without an ID. Parents' intellectual status did not have a direct effect on workers' attributions or removal decisions.

CONCLUSIONS

The results show evidence for the influence of stereotypes regarding parental ID due to its differential effect on CPS workers' emotional reactions and decisions about child risk and their willingness to help.

Contextual and individual-level predictors of abused children's reentry into out-of-home care: a multilevel mixture survival analysis

OBJECTIVES

This study examined the effects of individual and contextual factors on reentry into out-of-home care among children who were discharged from child protective services in fiscal year 2004-2005. The objectives were to: (1) examine individual and contextual factors associated with reentry, (2) explore whether there are meaningful groups of youth who differ in terms of risk for reentry, and (3) determine whether relatively homogeneous clusters of child welfare agencies, based on contextual characteristics, differ significantly in terms of the reentry rates of the children whom they serve.

METHOD

The study design involved a multilevel longitudinal analysis of administrative data based on an exit cohort. Two Cox proportional hazards multilevel mixture models were tested. The first model included multiple individual level predictors and no agency level predictors. The second model included both levels of predictors.

RESULTS

The results of multilevel Cox regression mixture modeling indicated that at the individual level, younger age, being placed in out-of-home care because of neglect and having physical, health problems corresponded to a decreased likelihood for reentry. At the agency level, lower average expenditures per child and contracting out case management services were associated with faster reentry into out-of-home care.

CONCLUSIONS

This study demonstrates that children who reenter out-of-home care appear to be a homogeneous population and that reentry is associated with both contextual factors and individual characteristics.

PRACTICE IMPLICATIONS

The most important implication that can be drawn from the study findings is that reentry may be most effectively prevented by focusing on such factors at the organizational level as contracting out case management services and funding allocation. Child welfare agencies that are responsible for an array of services and decide to contract out case management should consider the use of performance-based contracts and emphasize and strengthen quality assurance approaches for contracted services. In addition, to compensate for lower funding allocated for children served in out-of-home care, child welfare workers should become more familiar with community resources and help connect families to these supports.

Factors associated with caregiver stability in permanent placements: a classification tree approach

OBJECTIVE

Identify individual and environmental variables associated with caregiver stability and instability for children in diverse permanent placement types (i.e., reunification, adoption, and long-term foster care/guardianship with relatives or non-relatives), following 5 or more months in out-of-home care prior to age 4 due to substantiated maltreatment.

METHODS

Participants were 285 children from the Southwestern site of Longitudinal Studies of Child Abuse and Neglect (LONGSCAN). Caregiver instability was defined as a change in primary caregiver between ages 6 and 8 years. Classification and regression tree (CART) analysis was used to identify the strongest predictors of instability from multiple variables assessed at age 6 with caregiver and child reports within the domains of neighborhood/community characteristics, caregiving environment, caregiver characteristics, and child characteristics.

RESULTS

One out of 7, or 14% of the 285 children experienced caregiver instability in their permanent placement between ages 6 and 8. The strongest predictor of stability was whether the child had been placed in adoptive care. However, for children who were not adopted, a number of contextual factors (e.g., father involvement, expressiveness within the family) and child characteristics (e.g., intellectual functioning, externalizing problem behaviors) predicted stability and instability of permanent placements.

CONCLUSIONS

Current findings suggest that a number of factors should be considered, in addition to placement type, if we are to understand what predicts caregiver stability and find stable permanent placements for children who have entered foster care. These factors include involvement of a father figure, family functioning, and child functioning.

PRACTICE IMPLICATIONS

Adoption was supported as a desired permanent placement in terms of stability, but results suggest that other placement types can also lead to stability. In fact, with attention to providing biological parents, relative, and non-relative caregivers with support and resources (e.g., emotional, financial, and optimizing father involvement or providing a stable adult figure) the likelihood that a child will have a stable caregiver may be increased.

Household, family, and child risk factors after an investigation for suspected child maltreatment: a missed opportunity for prevention

OBJECTIVE

To determine whether a Child Protection Services investigation for suspected child maltreatment is associated with subsequent improvements in household, caregiver, and child risk factors.

DESIGN

Retrospective cohort study.

SETTING

The Longitudinal Studies of Child Abuse and Neglect, a multicenter cohort study of the antecedents and consequences of child maltreatment.

PARTICIPANTS

A total of 595 children with the same maternal caregiver responding to Longitudinal Studies of Child Abuse and Neglect surveys at ages 4 and 8 years.

MAIN EXPOSURE

Investigation for suspected child maltreatment between ages 4 and 8 years.

MAIN OUTCOME MEASURES

Adjusted differences in 7 modifiable risk factors (social support, family functioning, poverty, maternal education, maternal depressive symptoms, anxious or depressive child behaviors, and aggressive or destructive child behaviors) at age 8 years.

RESULTS

Of 595 subjects, 164 (27.6%) experienced an investigation for suspected child maltreatment between ages 4 and 8 years. At age 8 years, investigated subjects were not perceptibly different from noninvestigated subjects in social support, family functioning, poverty, maternal education, or child behavior problems after adjusting for baseline risk factors. Mothers of investigated subjects did have more depressive symptoms than mothers of noninvestigated peers at the child's age of 8 years. Substantiation of child maltreatment by Child Protective Services did not alter these findings.

CONCLUSIONS

Our finding that an investigation for suspected child maltreatment is not associated with relative improvements in common, modifiable risk factors suggests that we may be missing an opportunity for secondary prevention of maltreatment and maltreatment consequences.

A time series modeling approach in risk appraisal of violent and sexual recidivism

For over half a century, various clinical and actuarial methods have been employed to assess the likelihood of violent recidivism. Yet there is a need for new methods that can improve the accuracy of recidivism predictions. This study proposes a new time series modeling approach that generates high levels of predictive accuracy over short and long periods of time. The proposed approach outperformed two widely used actuarial instruments (i.e., the Violence Risk Appraisal Guide and the Sex Offender Risk Appraisal Guide). Furthermore, analysis of temporal risk variations based on specific time series models can add valuable information into risk assessment and management of violent offenders.

Predicting child maltreatment among Puerto Rican children from migrant and non-migrant families

OBJECTIVES

The objectives of the present study were to (1) describe the prevalence of child maltreatment among migrant and non-migrant Puerto Rican families and (2) identify socio-demographic and cultural (i.e., acculturation pattern, familismo) predictors of maltreatment within these two samples.

METHOD

Representative community samples of Puerto Rican children (ages 5-13 at baseline) and their adult caretakers were interviewed at two sites: the South Bronx in New York City (n=631 families) and the Standard Metropolitan Areas of San Juan and Caguas in Puerto Rico (n=859 families). Participants were re-interviewed 1 and 2 years following the baseline assessment.

RESULTS

While prevalence rates of maltreatment (physical abuse, 10%; sexual abuse 1%; neglect, 10%; and multi-type, 6%) did not differ between the two sites at baseline assessment, site differences emerged over time. Rates of physical abuse at follow-up were significantly higher in the Bronx compared to Puerto Rico. Further, for families living in the Bronx, living in poverty predicted chronic maltreatment, whereas living above the poverty line predicted new cases of maltreatment at follow-up. For families living in Puerto Rico, those who experienced physical abuse or multi-type maltreatment at baseline were more likely to report chronic maltreatment at follow-up regardless of poverty level. Cultural factors were not related to baseline or follow-up maltreatment at either site.

CONCLUSION

Findings suggest that while rates of child maltreatment may be similar in migrant and non-migrant Puerto Rican families and when compared to prevalence rates in the US, predictors of maltreatment may differ.

PRACTICE IMPLICATIONS

Since predictors of maltreatment may vary across population subgroups, studying homogenous samples will lead to more effective and targeted interventions.