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

Abstract

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.

Keywords

family violence risk/protective factors aggression/violence

Child protective services (CPS) agencies shoulder the heavy burden of receiving and addressing allegations of child abuse and neglect—more than 2 million in the United States annually (Child Maltreatment 2016, 2018). This burden carries responsibilities to assess and take steps to reduce current and future risk to children’s safety and ascertain needs for connecting children and families to services and resources that can promote recovery and well-being. Unfortunately, a substantial portion of children experience reoccurring contact with CPS. About half of children identified as alleged victims in CPS reports are already named victims in a prior accepted report (Kim, Wildeman, Jonson-Reid, & Drake, 2017). For reports that go on to be substantiated, about one third involve children who were victims of a prior substantiated report (Child Maltreatment 2016, 2018). Moreover, within as little as a year of an accepted report, one third of families will acquire a new report of abuse or neglect (Dakil, Sakai, Lin, & Flores, 2011). This high recidivism reflects the ongoing struggle of CPS agencies to estimate risk and link families with effective interventions for preventing recurrent abuse and neglect (Cuccaro-Alamin, Foust, Vaithianathan, & Putnam-Hornstein, 2017).

Efforts to better understand and conceptualize children’s experiences of maltreatment and adversity may yield valuable information for determining risk and identifying child and family needs toward reducing revictimization and recidivism. Research has made strides in recent years in moving beyond conceptualizing childhood maltreatment, trauma, adversity, and loss as consisting of simple unitary constructs to a more dimensional approach that seeks to understand risk mechanisms associated with co-occurrence, frequency, and severity of these experiences (Jackson, Gabrielli, Fleming, Tunno, & Makanui, 2014). Indeed, there have been several studies demonstrating differential outcomes on the basis of co-occurrence of adversity types (Grasso, Dierkhising, Branson, Ford, & Lee, 2016; Grasso, Petitclerc, et al., 2016; Grasso et al., 2013; Margolin, Vickerman, Oliver, & Gordis, 2010), exposure frequency (Bolger & Patterson, 2001; Jackson et al., 2014; Manly, Cicchetti, & Barnett, 1994; Miller, Chen, & Parker, 2011), and severity (Briggs-Gowan et al., 2019; Bryant & Range, 1997; Grasso et al., 2009; Jackson et al., 2014). Also, several studies have shown that polyvictimization and more frequent and severe maltreatment, adversity, and trauma substantively increase risk of revictimization and reoccurrence of these and other types of adverse experiences (Dierkhising, Ford, Branson, Grasso, & Lee, 2019; Finkelhor, Ormrod, & Turner, 2007). No known study, however, has examined whether co-occurrence, frequency, or severity of adversity is predictive of CPS recidivism.

Despite this advancement in how adversity is quantified and conceptualized in research, practice has not fully caught up. The quality and comprehensiveness of methods for assessing children’s experiences of maltreatment and other adversities vary substantially, both within and across CPS agencies; often it is suboptimal (Jenkins, Tilbury, Mazerolle, & Hayes, 2017). Although investigation of allegations involves determining whether there exists sufficient evidence that abuse or neglect occurred and establishing a level of risk to child safety, use of standardized assessment instruments for quantifying severity and making these determinations is inconsistent. In addition, substantiated allegations of abuse or neglect are well documented in CPS records and serve as dichotomous indicators of a child’s CPS maltreatment history; however, they are void of dimensional or contextual information that may serve utility in estimating risk and understanding the scope and severity of a child’s experiences.

Notably, knowledge of substantiation of abuse or neglect does not appear to be particularly useful in differentiating families that will go on to have subsequent CPS contact. Although about 20% of CPS reports are substantiated, a sizable portion of children with unsubstantiated allegations are re-reported, with rates of subsequent substantiation ranging from 10% to 80% depending on the amount of time that has passed since the initial report (Bae, Solomon, & Gelles, 2007; English, Marshall, Coghlan, Brummel, & Orme, 2002; Thompson & Wiley, 2009; Way, Chung, Jonson-Reid, & Drake, 2001). Two studies reported that substantiation status did not significantly predict substantiation of subsequent reports (Bae et al., 2007; English et al., 2002).

Other studies have identified risk factors for recidivism that are not well captured in administrative codes indicating substantiation status of traditional maltreatment categories. These include characteristics of domestic violence (DV), contextual stressors, and indicators of caregiver impairment, such as ongoing drug and alcohol abuse (Barth, Gibbons, & Guo, 2006; DePanfilis & Zuravin, 1999; English, Marshall, Brummel, & Orme, 1999; Sledjeski, Dierker, Brigham, & Breslin, 2008). Importantly, although administrative data on substantiated reports are easily queried, case records contain a vast amount of narrative information collected from multiple sources at intake and during investigations over time that could be mined to capture this contextual information, including severity of maltreatment and other forms of adversity.

Indeed, research studies have shown success in extracting case record information and employing coding procedures to quantify maltreatment severity using the Maltreatment Classification System (Huffhines et al., 2016; Runyan et al., 2005) and similar protocols (Grasso et al., 2009; Kaufman, Jones, Stieglitz, Vitulano, & Mannarino, 1994) and linking severity scores to mental health and other outcomes. No known study, however, has examined whether adverse and traumatic experiences not traditionally captured and quantified in CPS maltreatment categories (e.g., family separation, community violence, family suicidality) can also be effectively derived from information in case records and applied to risk assessment. Given documented risk of reexposure and revictimization associated with these other forms of adversity and trauma (Dierkhising et al., 2019; Finkelhor et al., 2007; Grasso, Dierkhising, et al., 2016), extracting case record information pertinent to these additional experiences may further inform risk assessment and CPS decision making and planning.

Not surprisingly, CPS agencies lack the resources to systematically extract, appraise, and apply case record information for quantifying aspects of maltreatment and adversity. Rather, caseworkers rely more heavily on basic administrative data (e.g., tally of allegation types, substantiation status) to inform decisions (Jenkins et al., 2017). This is problematic given the reported disconnect between administrative classifications of abuse and neglect and maltreatment determined by way of a thorough case review, with one study identifying 50% of a CPS sample with evidence of emotional abuse after reviewing case record information compared to only 9% with substantiated emotional abuse (Trickett, Mennen, Kim, & Sang, 2009). Of course, emotional abuse and neglect are more difficult to substantiate than more explicit forms of maltreatment, such as physical and sexual abuse, for which physical forensic evidence often determines level of risk (Jenkins et al., 2017; Slep & Heyman, 2006). The lack of utilization of narrative case record information may reflect the impracticality of caseworkers’ ability to extract data from voluminous electronic and paper documents that tend to be managed in dated systems that are not easily queried (Huffhines et al., 2016). Devising practical methods for enabling caseworks to synthesize and fully utilize case record information for purposes of quantifying severity, estimating risk, and case conceptualization and planning may prove worthwhile.

The current study is the first known study to determine the extent to which a standardized coding protocol could be applied to narrative case record information for purposes of quantifying severity of maltreatment- and non-maltreatment-classified adversity, loss, and traumatic experiences and predicting CPS outcomes. We employed the Yale-Vermont Adversity in Childhood Scale (Y-VACS), a novel coding protocol designed to assess severity of five CPS maltreatment categories (DV, psychological abuse, physical abuse, sexual abuse, neglect) and 13 other adverse experiences (separation from family, family suicidality, community violence, bullying, sexual assault by nonfamily member, accidents, natural disasters, health-related traumas, medical neglect, death of a loved one, exposure to caregiver substance abuse, exposure to caregiver arrest, and fire). We supplemented the Y-VACS with two additional coding domains that assessed severity of DV specific to psychological and physical intimate partner violence (IPV).

Development of the Y-VACS was inspired by research documenting the significant overlap and co-occurrence of both intrafamilial types of adversity and other forms of adversity, loss, and trauma. Sponsored by the National Institute of Mental Health (NIMH), the Y-VACS was designed to improve upon existing methods for quantifying severity that are less comprehensive or focus exclusively on traditional categories of maltreatment (Bethell et al., 2017; Holbrook et al., 2014). Several recent studies have demonstrated that severity scores generated by the Y-VACS predict depression and trauma-related symptoms in children, as well as stress-related epigenetic and biological markers (Kaufman, Gelernter, Hudziak, Tyrka, & Coplan, 2015; Kaufman et al., 2018; Oh et al., 2018).

We extracted case record information from a randomly selected sample of 100 families with 150 children referred to CPS during a 12-month period. Four expert coders applied Y-VACS ratings to narrative information extracted from case records. Specific study aims were to examine (a) the scope and severity of maltreatment and adversity types and their associations, (b) linkages between adversity severity and CPS allegation outcomes, and (c) the utility of adversity severity in predicting new allegations of abuse or neglect within 12 months of referral.

Method

Procedures

De-identified data were extracted via a comprehensive chart review of electronic and paper case records maintained by the State of Connecticut Department of Children and Families (DCF). The study was determined to be nonhuman subjects research and therefore exempt by the Connecticut Children’s Medical Center and University of Connecticut Health Center Human Subjects Review Boards. A total of 100 unique cases with an accepted report of abuse or neglect between August 1, 2013, and July 31, 2014 were randomly selected from two regional offices: Hartford (n = 50) and Willimantic (n = 50). The purpose for stratifying by region was to obtain a sample that was racially and ethnically representative of the overall population of children served in the state. Reports used for selection are referred to as “index reports” and associated allegations as “index allegations.” Selections were discarded and new selections made if (a) a case was selected twice due to more than one report within the same time period, (b) an allegation was against school personnel, or (c) if a case was sealed, which can occur when a case is deemed “high profile” or is associated with a state employee. New selections were made less than 10 times over the course of data collection.

Data extraction lasted approximately 6 months and spanned the entire history of a case preceding the index report, as well as 12 months following the index report. Data extraction procedures were uniform across data collection sites. Four trained data extractors reviewed the physical version of the case record before extraction to ensure that all pieces of relevant data were captured. Data were collected from a variety of sources within case records (i.e., checklists, data fields, narratives, case summaries, assessments, criminal histories, and correspondence documents). Type of data included child, parent, and family sociodemographic characteristics and descriptive information regarding maltreatment allegations and dispositions. Documented allegation types included physical, emotional, and sexual abuse, and physical, emotional, educational, and medical neglect. Index allegations were either substantiated, unsubstantiated, or triaged to differential response, known as Family Assessment Response (FAR) in Connecticut.

Case characteristics

Data were acquired on 100 cases involving a total of 150 children named as a victim in the index report. Number of index report victims per family ranged from one to seven (M = 1.51, SD = 0.92). These children ranged in age from newborn to 17 years (M = 8.03, SD = 5.04), with 40% under the age of 6 years. About half were female (48%). The majority of children were identified as White/Caucasian (44%), followed by Black/African American (23.3%), Hispanic/Latino (16%), multiracial (4%), and Asian (2.7%), with 10% nondisclosed. Demographic characteristics including sex, age, and race/ethnicity were consistent with the 2016 census of children referred to DCF, suggesting that random selection of cases resulted in a demographically representative sample. Consistent with population statistics, the two regions significantly differed in the proportion of non-Hispanic White/Caucasian children, with 81% in Willimantic and only 14% in Hartford. Age and sex did not significantly differ across regions.

Maltreatment and adversity severity coding

The Y-VACS (Holt-Gosselin et al., 2016) was applied to narrative information extracted from case records for quantifying severity of five CPS maltreatment categories (DV, psychological abuse, physical abuse, sexual abuse, neglect) and 13 other adverse experiences (separation from family, family suicidality, community violence, bullying, sexual assault by nonfamily member, accidents, natural disasters, health-related traumas, medical neglect, death of a loved one, exposure to caregiver substance abuse, exposure to caregiver arrest, and fire). The Y-VACS was supplemented with two additional categories designed to focus on DV specific to IPV and differentiate severity by psychological and physical violence. Thus, violence between adults in the home, regardless of partner status, was captured under the DV category, whereas violence between intimate partners in the home was specifically captured under the psychological and physical IPV categories. These categories were not mutually exclusive. On each adversity category, severity ratings ranged from 0 (“not present”) to 3 (“most severe”). Each adversity item has a set of criteria corresponding to the 0 to 3 ratings. Ratings were applied to each child identified as a victim in the index report (N = 150) and to all information pertinent to the index child from the first time the child was named a victim in a CPS report to the index report.

The Y-VACS includes a suite of instruments that consists of child self-report, caregiver report, adult self-report on childhood experiences, and clinician rating scales. The current study exclusively used the clinician rating scales on extracted case record information. To note, when information from multiple sources and methods is available, the clinician rating scales can be completed using all information available, including the self- and caregiver-reported scales included in the full suite of instruments.

Psychometric properties of the Y-VACS were previously established on a sample of 171 children in which concurrent validity was supported by associations between Y-VACS severity scores and several alternative measures of self- and caregiver-reported adversity and trauma exposure, as well as maltreatment data from CPS records (Holbrook et al., 2014). Several other studies have established predictive validity by demonstrating associations between Y-VACS severity scores and depression and trauma-related symptoms in children, as well as stress-related epigenetic and biological markers over time (Kaufman et al., 2015; Kaufman et al., 2018; Oh et al., 2018).

In the current study, four Y-VACS coders included a master’s level researcher, two doctoral-level medical anthropologists, and a doctoral-level clinical psychologist. Coders reviewed and provided ratings for all cases and held regular meetings to establish consensus ratings for each case. For purposes of analysis, data included both consensus ratings and averaged severity ratings. Analysis of both consensus and averaged ratings enabled us to examine the utility of different methods of synthesizing these data and make suggestions on how these methods might be applied in practice.

Data analysis

Interrater reliability on adversity severity ratings was evaluated using several metrics computed with SPSS software (Version 25). Interclass correlation coefficients (ICCs) were calculated using a two-way random effects model on mean ratings across the four coders, to evaluate consistency across raters, and on single raters, to evaluate absolute agreement. Criteria for evaluating ICC values were as follows: ≤.40 (fair agreement), .41 to .60 (moderate agreement), .61 to .80 (good agreement), .81 to 1.00 (excellent agreement) (Landis, King, Choi, Chinchilli, & Koch, 2011).

Descriptive statistics included the use of t tests, chi-square goodness-of-fit tests, and Pearson’s and Spearman’s correlation coefficients. Secondary analyses included partial correlations with sibling status (i.e., 1 = sibling in the data set, 0 = no sibling in the data set) as a covariate. Logistic regression was used to identify significant predictors of new allegations of abuse or neglect.

Results

Adversity severity

Adversity severity ratings were generated for 13 Y-VACS categories and the two supplemental IPV categories. Five Y-VACS categories were not scorable due to insufficient information in charts, either a consequence of low base rate or lack of documentation of these types of adversities; thus, bullying, accidental trauma, natural disasters, death of a loved one, and fire were not examined further. Table 1 presents mean and single-rater ICCs on adversity severity ratings across adversity types. Interrater reliability ranged from moderate to good agreement for single-rater ICCs and moderate to excellent agreement for mean ICCs. Table 1 also presents mean severity ratings by adversity types based on both averaged ratings and consensus ratings. The final column in Table 1 presents correlations between averaged and consensus ratings, which ranged from .84 to .95.

Table 1.

Interrater Reliability on Adversity Severity Ratings (N = 150).

Adversity Category	Single ICC	95% CI		Mean ICC	95% CI		Averaged Ratings		Consensus Ratings		Correlations Between Averaged and Consensus Ratings
Adversity Category	Single ICC	Lower	Upper	Mean ICC	Lower	Upper	M	SD	M	SD	Correlations Between Averaged and Consensus Ratings
DV	.71	.63	.78	.91	.87	.93	1.57	1.00	1.70	1.09	.95
Psych. IPV	.65	.56	.73	.88	.84	.91	1.52	1.07	1.65	1.17	.92
Phys. IPV	.55	.46	.64	.83	.77	.88	1.07	0.82	1.03	0.97	.87
Psych. Abuse	.46	.36	.56	.77	.69	.84	0.83	0.82	1.05	1.08	.84
Phys. Abuse	.74	.68	.81	.92	.89	.94	0.57	0.74	0.60	0.81	.93
Sex Abuse	.71	.64	.78	.91	.87	.93	0.17	0.44	0.18	0.50	.86
Neglect	.63	.54	.71	.87	.82	.91	1.49	0.93	1.59	1.12	.91
Loss	.54	.44	.63	.82	.76	.87	1.10	0.75	1.11	0.86	.87
CG Sub.	.75	.68	.81	.92	.90	.95	1.01	1.0	1.12	1.17	.94
Suicidality	.54	.44	.63	.82	.76	.87	0.28	0.61	0.29	0.74	.88
Criminality	.57	.48	.66	.84	.78	.89	0.68	0.60	0.73	0.70	.87
CV	.80	.74	.85	.94	.92	.96	0.15	0.48	0.15	0.54	.95
Sex Assault	.48	.38	.58	.79	.71	.85	0.04	0.22	0.03	0.27	.85
Health	.67	.59	.75	.89	.85	.92	0.12	0.39	0.17	0.52	.91

Note. CG Sub. = caregiver substance abuse; CI = confidence interval; CV = community violence; DV = domestic violence; Health = health-related trauma; ICC = interclass correlation coefficients; Phys. IPV = physical intimate partner violence; Psych. IPV = psychological intimate partner violence.

Tables 2 and 3 present a correlation matrix of adversity severity types using averaged and consensus ratings, respectively (see Supplemental Tables S1 and S2 for partial correlations controlling for sibling status). In general, adversity severity types reflecting abuse and neglect and caregiver-related adversities were highly intercorrelated, with the exception of sexual abuse severity. Additional positive associations included loss severity with psychological IPV, psychological abuse, and neglect severity, as well as health-related trauma severity with neglect severity and caregiver substance abuse.

Table 2.

Pearson Correlation (r) Matrix of Averaged Adversity Severity Ratings (N = 150).

Adversity Type	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. DV	—	.78**	.90**	.26*	.15*	−.12	.20**	.13	.12	.21**	.54**	.06	.00	−.13
2. Psych. IPV		—	.72**	.27**	.12	−.06	.27**	.14	.13	.32**	.38**	.11	.12	−.14
3. Phys. IPV			—	.26*	.12	−.11	.23**	.23**	.10	.15	.56**	.07	−.05	−.11
4. Psych. Abuse				—	.41**	.18*	.26**	.43**	.21**	.12	.20**	−.03	.13	.10
5. Phys. Abuse					—	.11	.03	.01	.04	.14	.09	.16*	.05	−.02
6. Sex Abuse						—	.04	.16*	.05	−.01	−.09	.01	.13	.12
7. Neglect							—	.32*	.53**	.22**	.31**	.09	.00	.22**
8. Loss								—	.13	−.03	.37**	.04	.01	.10
9. CG Sub.									—	.15	.25**	.13	−.01	.15*
10. Suicidality										—	.13	.01	−.02	−.02
11. Criminality											—	.11	−.14	−.10
12. CV												—	.00	.01
13. Sex Assault													—	.06
14. Health														—

Note. CG Sub. = caregiver substance abuse; CV = community violence; DV = domestic violence; Health = health-related trauma; Psych. IPV = psychological intimate partner violence; Phys. IPV = physical intimate partner violence.

p < .05. **p < .01.

Table 3.

Pearson Correlation (r) Matrix of Adversity Severity Ratings by Consensus (N = 150).

Adversity Type	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. DV	—	.67**	.77**	.18*	.18*	−.06	.19*	.11	.01	.20*	.43**	.08	.11	−.09
2. Psych. IPV		—	.61**	.29**	.05	.04	.25**	.21**	.15	.24**	.25**	.05	.08	−.09
3. Phys. IPV			—	.19*	.17*	.04	.09	.10	−.01	.12	.42**	.11	.10	−.00
4. Psych. Abuse				—	.40**	.18*	.20*	.36**	.15	.17*	.12	.04	.12	.07
5. Phys. Abuse					—	.18*	−.01	.07	−.02	.26**	.09	.16	.15	.14
6. Sex Abuse						—	−.03	−.00	−.01	−.06	−.11	−.06	.01	.01
7. Neglect							—	.18*	.36**	.09	.26**	.12	−.02	.26**
8. Loss								—	.18*	.06	.30**	.11	.08	.13
9. CG Sub.									—	.04	.13	.09	.03	.13
10. Suicidality										—	.05	.03	.05	−.03
11. Criminality											—	.11	−.13	−.02
12. CV												—	−.05	.00
13. Sex Assault													—	.03
14. Health														—

p < .05. **p < .01.

There were several associations between adversity severity ratings and demographic characteristics. Child age at the index report was positively associated with severity of DV (averaged ratings: r = .18, p = .028), psychological abuse (averaged ratings: r = .42, p < .001; consensus ratings: r = .35, p < .001), physical abuse (averaged ratings: r = .43, p < .001; consensus ratings: r = .44, p < .001), loss (averaged ratings: r = .24, p = .003; consensus ratings: r = .22, p = .008), caregiver substance abuse (averaged ratings: r = .16, p = .048), and sexual assault by a nonfamily member (averaged ratings: r = .30, p < .001; consensus ratings: r = .27, p = .001).

Being female was associated with increased severity of DV (consensus ratings: r_s = .17, p = .035), psychological IPV (consensus ratings: r_s = .20, p = .014), physical IPV (averaged ratings: r_s = .19, p = .019), psychological abuse (consensus ratings: r_s = .19, p = .018), neglect (averaged ratings: r_s = .17, p = .035), caregiver substance abuse (averaged ratings: r_s = .20, p = .014; consensus ratings: r_s = .18, p = .030), and sexual assault by a nonfamily member (averaged ratings: r_s = .17, p = .041).

Race/ethnicity minority status (1 = minority, 0 = nonminority) was associated with physical abuse (averaged ratings: r_s = .23, p = .005; consensus ratings: r_s = .20, p = .014) and family criminality (averaged ratings: r_s = .18, p = .032; consensus ratings: r_s = .20, p = .015).

Having a sibling in the data set was associated with increased severity of DV (averaged ratings: r_s = .29, p < .001; consensus ratings: r_s = .26, p = .002), psychological IPV (averaged ratings: r_s = .33, p < .001; consensus ratings: r_s = .33, p < .001), physical IPV (averaged ratings: r_s = .24, p = .003; consensus ratings: r_s = .19, p = .019), and sexual assault by a nonfamily member (averaged ratings: r_s = .18, p = .032) and less severe ratings of loss (averaged ratings: r_s = .33, p < .001) and health-related trauma (averaged ratings: r_s = –.26, p = .001; consensus ratings: r_s = –.28, p = .001).

Index allegations

Index allegations (not mutually exclusive) included physical (11.3%), emotional (3.3%), and sexual abuse (2%) and physical (77.3%), emotional (29.3%), educational (2.7%), and medical neglect (3.3%). DV exposure was documented as a problem by CPS as part of neglect allegations for 30.7% of the sample. Among allegation types, those triaged to differential response, rather than investigation, included 23.5% (4/17) physical and 40% (2/5) emotional abuse allegations, and 35.3% (41/116) physical, 27.3% (12/44) emotional, 75% (3/4) educational, and 20% (1/5) medical neglect allegations. Among investigated allegations, those substantiated included 23.1% (3/13) physical abuse, and 32% (24/75) physical, 46.9% (15/32) emotional, and 25% (1/4) medical neglect allegations, with 22.7% (34/150) of children having at least one substantiated allegation. No allegations of emotional or sexual abuse, or educational neglect were substantiated. Boys were more likely to have an index allegation of physical abuse (r_s = .18), and girls were more likely to have an index allegation of physical neglect (r_s = .19). There were no significant differences by age, racial/ethnic minority status, or whether a child had a sibling in the data set.

Past allegations

More than half (63.3%; 95/150) of children had at least one past allegation of abuse or neglect. Past allegation types (not mutually exclusive) included physical (22.7%), emotional (7.3%), and sexual (2.7%) abuse, and physical (52.7%), emotional (27.3%), educational (4.7%), and medical neglect (4.7%). DV exposure was documented as a problem by CPS as part of neglect allegations for 20.7% of the sample. Among allegation types, those triaged to differential response included 8.8% physical (3/34) and 36.4% (4/11) emotional abuse, and 22.8% (18/79) physical, 17.1% (7/41) emotional, 28.6% (2/7) educational, and 42.9% (3/7) medical neglect. Among investigated allegation types, those substantiated included 12.9% (4/31) physical and 57.1% (4/7) emotional abuse, and 29.5% (18/61) physical, 20.6% (7/34) emotional, and 40% (2/5) educational neglect, with 34.7% (33/95) of children with a past allegation having at least one allegation substantiated. No allegations of sexual abuse or medical neglect were substantiated. Older children were more likely to have past allegations of physical abuse (r_s = .26), educational neglect (r_s = .25), and medical neglect (r_s = .20). Older children were also more likely to have at least one past allegation that was substantiated (r_s = .30). There were no significant differences by sex, racial/ethnic minority status, or whether a child had a sibling in the data set.

New allegations

Forty-two children (28%) were named victim in at least one new report of abuse or neglect made within 1 year following the index report. Among children with at least one new allegation, allegation types (not mutually exclusive) included physical (19%; 8/42) and emotional (7.1%; 3/42) abuse, and physical (85.7%; 36/42), emotional (47.6%; 20/42), and medical neglect (2.4%; 1/42). Among those with a new allegation, 35.7% (15/42) had an allegation for which DV exposure was documented as a problem. Among allegation types, those triaged to differential response included 37.5% (3/8) physical and 66.7% (2/3) emotional abuse, and 25% (9/36) physical and 20% (4/20) emotional neglect. Among investigated allegations, those substantiated included 22.2% (6/27) physical and 25% (4/16) emotional neglect, with 16.7% (7/42) of children with a new allegation having at least one allegation substantiated. No allegations of physical or emotional abuse, or medical neglect were substantiated. Younger children were more likely to have a new allegation of emotional neglect (r_s = .23). There were no significant differences by sex, racial/ethnic minority status, or whether a child had a sibling in the data set. Finally, having a new allegation was not significantly predicted by having a past allegation or having a past or index allegation that was substantiated, nor was it associated with having a specific type of past or index allegation.

Allegation types, dispositions, and adversity severity

Tables 4 and 5 present nonparametric correlations between averaged and consensus adversity severity scores, respectively, allegations, and allegation outcomes (see Supplemental Tables S3 and S4 for partial correlations controlling for sibling status). Past and index allegations and outcomes are collapsed to reflect presence/absence. Greater severity on DV and psychological and physical IPV was associated with having a past/index allegation of physical and emotional neglect. Greater physical abuse severity was associated with having a past/index allegation of physical and emotional abuse and medical neglect. Sexual abuse severity was only associated with having a past/index allegation of sexual abuse. Greater neglect severity was associated with having a past/index allegation of physical, educational, and medical neglect. Caregiver substance abuse severity was associated with having a past/index allegation of physical and educational neglect. Familial suicidality severity was associated with having a past/index allegation of medical neglect. Familial criminality severity was associated with having a past/index allegation of physical neglect. Witness to community violence severity was associated with having a past/index allegation of physical and emotional abuse and medical neglect. Sex assault severity was associated with educational neglect. Health trauma severity was associated with medical neglect.

Table 4.

Spearman’s Correlation (r_s) Matrix of Averaged Adversity Severity Scores and Allegation Outcomes (N = 150).

Adversity Type	Past/Index Allegations									New Allegations Any
Adversity Type	PA	EA	PN	EN	SA	ED	MN	Any Sub.	Any Diff. Resp.	New Allegations Any
DV	.04	.05	.21**	.34**	−.11	.08	.01	.29**	−.10	.10
Psych. IPV	−.08	.16	.18*	.25**	−.06	−.03	−.06	.19*	.05	.17*
Phys. IPV	.03	.00	.17*	.34**	−.06	.07	−.04	.23**	−.07	.16
Psych. Abuse	.29**	.29**	.02	.04	−.02	.02	.18*	.09	.01	.04
Phys. Abuse	.57**	.27**	−.02	.10	.08	.06	.20*	.22**	−.03	−.02
Sex Abuse	.04	−.13	.10	−.04	.39**	.04	−.04	−.06	.03	.05
Neglect	.03	.09	.34**	.00	−.06	.21*	.21*	.25**	−.10	.20*
Loss	.10	−.11	.12	−.13	.00	.13	.17*	.14	−.12	−.03
CG Sub.	−.08	.05	.23**	.05	−.04	.18*	.08	.16*	−.09	.08
Suicidality	−.05	.09	.12	.13	−.10	.02	.21*	.06	−.16	−.02
Criminality	.03	−.06	.17*	.12	−.04	−.03	.12	.39**	−.05	.05
CV	.19*	.18*	.11	.00	.08	.11	.18*	.03	.00	.04
Sex Assault	.06	.03	.06	−.09	−.05	.19*	−.06	.07	.05	.00
Health	.09	−.01	.11	−.02	.00	.07	.31**	.06	−.09	.07

Note. CG Sub. = caregiver substance abuse; CV = community violence; Diff. Resp. = differential response; DV = domestic violence; EA = emotional abuse; ED = educational neglect; EN = emotional neglect; Health = health-related trauma; MN = medical neglect; PA = physical abuse; Phys. IPV = physical intimate partner violence; PN = physical neglect; Psych. IPV = psychological intimate partner violence; SA = sexual abuse; Sub. = substantiation.

p < .05. **p < .01.

Table 5.

Spearman’s Correlation (r_s) Matrix of Consensus Adversity Severity Scores and Allegation Outcomes (N = 150).

Adversity Type	Past/Index Allegations									New Allegations Any
Adversity Type	PA	EA	PN	EN	SA	ED	MN	Any Sub.	Any Diff. Resp.	New Allegations Any
DV	.02	.06	.18*	.29**	−.12	.10	.02	.31**	−.09	.05
Psych. IPV	−.13	.14	.19*	.26**	−.07	−.05	−.05	.23**	.02	.18*
Phys. IPV	.11	.03	.15	.31**	−.01	.13	−.06	.17*	.05	.13
Psych. Abuse	.27**	.32**	.01	.05	.04	.02	.15	.15	−.08	.10
Phys. Abuse	.53**	.27**	−.03	.14	.08	.06	.19*	.25*	−.06	−.05
Sex Abuse	.07	−.13	.10	−.02	.40**	.04	−.03	−.05	.01	.04
Neglect	.01	.12	.32**	−.01	−.02	.16	.20*	.32**	−.16*	.24**
Loss	.04	−.09	.07	−.12	−.07	.12	.13	.15	−.06	−.13
CG Sub.	−.10	.03	.22**	.05	−.02	.19*	.06	.10	−.04	.04
Suicidality	−.07	.15	.10	.15	−.08	.06	.19*	.12	−.11	−.14
Criminality	.04	−.05	.19*	.12	−.03	.02	.14	.36**	.02	−.05
CV	.19*	.13	.09	−.00	.03	.10	.18*	.01	−.06	.10
Sex Assault	−.02	.09	.04	.00	−.04	.27**	−.05	.11	.01	−.01
Health	−.08	.00	.11	−.03	.01	.03	.26**	.07	−.08	.02

p < .05. **p < .01.

DV, psychological and physical IPV, physical abuse, neglect, and family criminality severity were positively associated with having a prior/index allegation that was substantiated. Neglect severity was negatively associated with having a prior/index allegation resulting in differential response.

Having a new allegation of abuse or neglect within 12 months of the index report was not associated with age, sex, or racial/ethnic minority status, but was associated with psychological IPV and neglect severity for both averaged and consensus ratings (i.e., at the bivariate level). When entered together in a logistic regression model using consensus ratings, neglect severity, area under curve (AUC) = 0.647, β = 0.43, SE = 0.18, p = .015, odds ratio (OR) = 1.6, 95% confidence interval (CI) = [1.09, 2.19], but not psychological IPV severity, AUC = 0.611, β = 0.27, SE = 0.17, p = .107, OR = 1.3, 95% CI = [.94, 1.82], remained significantly predictive of having a new allegation. Controlling for age at index allegation, sex, racial/ethnic minority status, past substantiated reports, and having a sibling in the data set did not change the significance of these results, with neglect severity significantly predictive of having a new allegation, β = 0.53, SE = 0.19, p = .006, OR = 1.7, 95% CI = [1.16, 2.46].

The same analysis with averaged ratings yielded consistent results, with neglect severity, AUC = 0.628, β = 0.43, SE = 0.20, p = .032, OR = 1.5, 95% CI = [1.04, 2.28], but not psychological IPV severity, AUC = 0.605, β = 0.32, SE = 0.18, p = .077, OR = 1.4, 95% CI = [0.97, 1.95], remaining significantly predictive of having a new allegation. Controlling for age at index allegation, sex, racial/ethnic minority status, past substantiated reports, and having a sibling in the data set did not change the significance of these results, with neglect severity significantly predictive of having a new allegation, β = 0.56, SE = 0.22, p = .011, OR = 1.2, 95% CI = [1.14, 2.72].

Discussion

The current study demonstrated that extracted case record information can be used to reliably quantify the severity of maltreatment and other adverse experiences among CPS-referred youth. These included adverse experiences that go beyond those captured by traditional CPS maltreatment categories, including traumatic loss, familial suicidality, familial criminality, exposure to community violence, health-related trauma, and caregiver substance abuse. As severity on several of these other adversities was associated with maltreatment severity and CPS outcomes, mining case records for this information may provide some utility in formulating a comprehensive conceptualization of a child’s adverse experiences, estimating risk, and planning for services. However, several non-maltreatment-related adversities were not quantifiable using case records due to either low base rates or lack of documentation of these other experiences. These included bullying, accidental trauma, natural disasters, death of a loved one, and fire.

Severity of DV exposure, and psychological and physical IPV, specifically, were reliably quantified using case record information, despite that they do not constitute a separate CPS maltreatment category, but rather are documented under physical or emotional neglect in Connecticut, as well as in several other states. It is worth noting that neglect is considered a deprivation-based adversity, whereas DV exposure is better characterized as a threat-based adversity associated with distinct biobehavioral consequences (McLaughlin, Sheridan, & Lambert, 2014). Indeed, in the current data set, measures of DV severity were associated with psychological and physical abuse severity. Also, the CPS-documented rate of DV exposure in past/index allegations was smaller than what was found following a comprehensive chart review (44% vs. 78%), suggesting that current CPS documentation fails to wholly capture the extent of DV exposure among referred children. This is particularly notable given that, at the bivariate level, psychological IPV severity was predictive of new allegations of abuse and neglect in the year following the index report.

Neglect severity was also predictive of having a new allegation within 1 year of the index allegation and assumed most of the variance in the model when entered along with psychological IPV severity. This is not surprising given that the majority of new allegations were for physical neglect. The utility of quantifying neglect severity was further demonstrated in that it remained predictive of new allegations when controlling for age, sex, racial/ethnic minority status, having a sibling in the data set, any past substantiated reports, and any past allegations of emotional or physical neglect. Thus, knowing about the severity of neglect appears to add incremental value to administrative data in estimating short-term recidivism risk.

Findings suggest value in CPS agencies moving beyond an incident-based approach by investing in ways to better utilize case record information to understand, quantify, and conceptualize severity of children’s cumulative exposure to maltreatment and other adversities. One step toward this goal may involve CPS agencies developing and implementing procedures for systematically synthesizing and quantifying adversity-related information. There have been a number of research studies in which case record information was used to code maltreatment severity (Grasso et al., 2009; Huffhines et al., 2016; Kaufman et al., 1994; Runyan et al., 2005; Taussig, Culhane, Garrido, Knudtson, & Petrenko, 2013), although the Y-VACS is the only coding protocol that includes non-maltreatment-related adversities. In light of research highlighting the significant overlap of maltreatment and non-maltreatment-related adversities and trauma (Finkelhor, Ormrod, Turner, & Hamby, 2005; Grasso, Dierkhising, et al., 2016; Grasso, Petitclerc, et al., 2016), with linkages to emergent mental health symptoms and impairment, it may benefit CPS agencies to collect and extract information about other forms of adversity when conceptualizing cases, estimating risk, and establishing plans for safety and prevention. Of course, implementing these protocols is labor-intensive and requires training to achieve reliability. The challenge will be to develop more practical methods that can be standardized across the CPS workforce.

To note, findings in the current study were consistent whether adversity severity ratings were quantified by averaging across the four expert coders or determined via group consensus. The latter method demands the most time and staff coordination. Furthermore, interrater reliability was good to excellent, suggesting that once reliable, coders may be capable of producing valid severity ratings independently, without the need for multiple coders. CPS agencies interested in implementing a protocol for producing adversity severity ratings might decide to train and establish reliability among staff or caseworkers who could serve to review charts for the purposes of quantifying adversity severity ratings. A second step toward this goal may involve agencies transitioning to more sophisticated electronic record systems capable of managing information in such a way as to provide easily queried “dashboard” summaries of adversity information that can be more readily applied to decision making in time-sensitive situations.

A limitation of the current study includes our modest sample size, which prevented us from applying more sophisticated methods for modeling the nested structure of the data. Another limitation was our inability to utilize the full suite of instruments available in the Y-VACS to obtain and examine information from multiple informants and via multiple methods. Adversity severity ratings are only as valid as the information used to produce them. This study relied exclusively on information available in case records—much of it gleaned from supplemental narrative information that varied in quantity and quality across cases. Future efforts to quantify adversity severity ratings for research or more practical reasons should consider incorporating a larger breadth of data sources. Future studies should also follow cases over a longer period of time to examine the utility of adversity severity ratings in predicting recidivism that extends beyond 1 year. Finally, studies examining feasibility of adversity severity coding for practical purposes should examine whether caseworkers can reliably code severity using case record information. Despite these limitations, the current study represents a preliminary step toward supporting a more comprehensive and data-driven CPS approach for understanding and quantifying the complexity of children’s exposure to maltreatment and other forms of adversity and trauma.

Supplemental Material

SupplementalTables-Final – Supplemental material for Quantifying Severity of Maltreatment, Adversity, and Trauma From Child Protective Services Case Record Files

Supplemental material, SupplementalTables-Final for Quantifying Severity of Maltreatment, Adversity, and Trauma From Child Protective Services Case Record Files by Damion J. Grasso, Susan DiVietro, Rebecca Beebe, Meghan Clough and Garry Lapidus in Journal of Interpersonal Violence

Footnotes

Acknowledgements

We wish to acknowledge Ms. Mary Painter and Ms. Linda Madigan from the Connecticut Department of Children and Families (DCF), as well as the DCF Commissioner at the time of the research study, Ms. Joette Katz, and her administration. Data extraction and management was supported by a contract between the Connecticut DCF and the Connecticut Children’s Injury Prevention Center. Data analysis and manuscript preparation were provided in-kind.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Damion J. Grasso

Author Biographies

Damion J. Grasso, PhD, is an assistant professor of psychiatry and pediatrics at the University of Connecticut School of Medicine. A principal component of his work focuses on the development of trauma-related psychopathology in children exposed to interpersonal violence and adversity, as well as prevention and intervention strategies.

Susan DiVietro, PhD, is a medical anthropologist and research scientist at the Connecticut Children’s Medical Center. Her research focuses on screening and assessment of intimate partner violence in families.

Rebecca Beebe, PhD, is a medical anthropologist and research scientist at the Connecticut Children’s Medical Center. Her research focuses on interventions for family violence with an emphasis on families involved in the child welfare system.

Meghan Clough, MA, is a research associate at Connecticut Children’s Medical Center in the Injury Prevention Center. She received her master’s degree from Central Connecticut State University. She managed the data from this study.

Garry Lapidus, PA-C, MPH, is the director of the Connecticut Children’s Injury Prevention Center and a national leader in injury prevention research, education, and training. A significant component of his work focuses on families affected by domestic violence.

References

Bae

H. -O.

Solomon

P. L.

Gelles

R. J.

(2007). Abuse type and substantiation status varying by recurrence. Children and Youth Services Review, 29, 856-869. doi:10.1016/j.childyouth.2007.01.002

Barth

R. P.

Gibbons

Guo

(2006). Substance abuse treatment and the recurrence of maltreatment among caregivers with children living at home: A propensity score analysis. Journal of Substance Abuse Treatment, 30, 93-104. doi:10.1016/j.jsat.2005.10.008

Bethell

C. D.

Carle

Hudziak

Gombojav

Powers

Wade

Braveman

(2017). Methods to assess adverse childhood experiences of children and families: Toward approaches to promote child well-being in policy and practice. Academic Pediatrics, 17(7), S51-S69. doi:10.1016/j.acap.2017.04.161

Bolger

K. E.

Patterson

C. J.

(2001). Developmental pathways from child maltreatment to peer rejection. Child Development, 72, 549-568. doi:10.1111/1467-8624.00296

Briggs-Gowan

M. J.

Estabrook

Henry

Grasso

D. J.

Burns

McCarthy

K. J.

. . . Wakschlag

L. S.

(2019). Parsing dimensions of family violence exposure in early childhood: Shared and specific contributions to emergent psychopathology and impairment. Child Abuse & Neglect, 87, 100-111.

Bryant

S. L.

Range

L. M.

(1997). Type and severity of child abuse and college students’ lifetime suicidality. Child Abuse & Neglect, 21, 1169-1176. doi:10.1016/S0145-2134(97)00092-6

Child Maltreatment 2016. (2018). Retrieved from https://www.acf.hhs.gov/cb/research-data-technology/statistics-research/child-maltreatment

Cuccaro-Alamin

Foust

Vaithianathan

Putnam-Hornstein

(2017). Risk assessment and decision making in child protective services: Predictive risk modeling in context. Children and Youth Services Review, 79, 291-298. doi:10.1016/j.childyouth.2017.06.027

Dakil

S. R.

Sakai

Lin

Flores

(2011). Recidivism in the child protection system: Identifying children at greatest risk of reabuse among those remaining in the home. Archives of Pediatrics & Adolescent Medicine, 165, 1006-1012. doi:10.1001/archpediatrics.2011.129

10.

DePanfilis

Zuravin

S. J.

(1999). Predicting child maltreatment recurrences during treatment. Child Abuse & Neglect, 23, 729-743. doi:10.1016/S0145-2134(99)00046-0

11.

Dierkhising

C. B.

Ford

J. D.

Branson

Grasso

D. J.

Lee

(2019). Developmental timing of polyvictimization: Continuity, change, and association with adverse outcomes in adolescence. Child Abuse & Neglect, 87, 40-50. doi:10.1016/j.chiabu.2018.07.022

12.

English

D. J.

Marshall

D. B.

Brummel

Orme

(1999). Characteristics of repeated referrals to child protective services in Washington state. Child Maltreatment, 4, 297-307. doi:10.1177/1077559599004004003

13.

English

D. J.

Marshall

D. B.

Coghlan

Brummel

Orme

(2002). Causes and consequences of the substantiation decision in Washington state child protective services. Children and Youth Services Review, 24, 817-851. doi:10.1016/S0190-7409(02)00241-4

14.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Re-victimization patterns in a national longitudinal sample of children and youth. Child Abuse & Neglect, 31, 479-502. doi:10.1016/j.chiabu.2006.03.012

15.

Finkelhor

Ormrod

R. K.

Turner

H. A.

Hamby

S. L.

(2005). The victimization of children and youth: A comprehensive, national survey. Child Maltreatment, 10, 5-25. doi:10.1177/1077559504271287

16.

Grasso

D. J.

Boonsiri

Lipschitz

Guyer

Houshyar

Douglas-Palumberi

. . . Kaufman

(2009). Posttraumatic stress disorder: The missed diagnosis. Child Welfare, 88(4), 157-176. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20405781

17.

Grasso

D. J.

Dierkhising

C. B.

Branson

C. E.

Ford

J. D.

Lee

(2016). Developmental patterns of adverse childhood experiences and current symptoms and impairment in youth referred for trauma-specific services. Journal of Abnormal Child Psychology, 44, 871-886. doi:10.1007/s10802-015-0086-8

18.

Grasso

D. J.

Petitclerc

Henry

D. B.

McCarthy

K. J.

Wakschlag

L. S.

Briggs-Gowan

M. J.

(2016). Examining patterns of exposure to family violence in preschool children: A latent class approach. Journal of Traumatic Stress, 29, 491-499. doi:10.1002/jts.22147

19.

Grasso

D. J.

Saunders

B. E.

Williams

L. M.

Hanson

Smith

D. W.

Fitzgerald

M. M.

(2013). Patterns of multiple victimization among maltreated children in navy families. Journal of Traumatic Stress, 26, 597-604. doi:10.1002/jts.21853

20.

Holbrook

O’Loughlin

Althoff

Douglas-Palumberi

Kaufman

Hudziak

(2014, October). The Yale-Vermont adversity in childhood scale: A quantitative approach to adversity assessment. Paper presented at the American Academy of Child and Adolescent Psychiatry’s 61st Annual Meeting, San Diego, CA.

21.

Holt-Gosselin

Schreck

Ametti

O’Loughlin

Holbrook

Hudziak

J. J.

Althoff

(2016). 6.127 relationship between adverse childhood experiences and metabolic and psychiatric outcomes using the Yale-Vermont Adversity in Childhood scale. Journal of the American Academy of Child & Adolescent Psychiatry, 55(10), S244-S245. doi:10.1016/j.jaac.2016.09.443

22.

Huffhines

Tunno

A. M.

Cho

Hambrick

E. P.

Campos

Lichty

Jackson

(2016). Case file coding of child maltreatment: Methods, challenges, and innovations in a longitudinal project of youth in foster care. Children and Youth Services Review, 67, 254-262. doi:10.1016/j.childyouth.2016.06.019

23.

Jackson

Gabrielli

Fleming

Tunno

A. M.

Makanui

P. K.

(2014). Untangling the relative contribution of maltreatment severity and frequency to type of behavioral outcome in foster youth. Child Abuse & Neglect, 38, 1147-1159. doi:10.1016/j.chiabu.2014.01.008

24.

Jenkins

B. Q.

Tilbury

Mazerolle

Hayes

(2017). The complexity of child protection recurrence: The case for a systems approach. Child Abuse & Neglect, 63, 162-171. doi:10.1016/j.chiabu.2016.11.020

25.

Kaufman

Gelernter

Hudziak

J. J.

Tyrka

A. R.

Coplan

J. D.

(2015). The Research Domain Criteria (RDoC) project and studies of risk and resilience in maltreated children. Journal of the American Academy of Child & Adolescent Psychiatry, 54, 617-625. doi:10.1016/j.jaac.2015.06.001

26.

Kaufman

Jones

Stieglitz

Vitulano

Mannarino

A. P.

(1994). The use of multiple informants to assess children’s maltreatment experiences. Journal of Family Violence, 9, 227-248. doi:10.1007/BF01531949

27.

Kaufman

Wymbs

N. F.

Montalvo-Ortiz

J. L.

Orr

Albaugh

M. D.

Althoff

. . . Hudziak

(2018). Methylation in OTX2 and related genes, maltreatment, and depression in children. Neuropsychopharmacology, 43, 2204-2211. doi:10.1038/s41386-018-0157-y

28.

Kim

Wildeman

Jonson-Reid

Drake

(2017). Lifetime prevalence of investigating child maltreatment among U.S. children. American Journal of Public Health, 107, 274-280.

29.

Landis

J. R.

King

T. S.

Choi

J. W.

Chinchilli

V. M.

Koch

G. G.

(2011). Measures of agreement and concordance with clinical research applications. Statistics in Biopharmaceutical Research, 3, 185-209.

30.

Manly

J. T.

Cicchetti

Barnett

(1994). The impact of subtype, frequency, chronicity, and severity of child maltreatment on social competence and behavior problems. Development and Psychopathology, 6, 121-143. doi:10.1017/S0954579400005915

31.

Margolin

Vickerman

K. A.

Oliver

P. H.

Gordis

E. B.

(2010). Violence exposure in multiple interpersonal domains: Cumulative and differential effects. Journal of Adolescent Health, 47, 198-205. doi:10.1016/j.jadohealth.2010.01.020

32.

McLaughlin

K. A.

Sheridan

M. A.

Lambert

H. K.

(2014). Childhood adversity and neural development: Deprivation and threat as distinct dimensions of early experience. Neuroscience & Biobehavioral Reviews, 47, 578-591. doi:10.1016/j.neubiorev.2014.10.012

33.

Miller

G. E.

Chen

Parker

K. J.

(2011). Psychological stress in childhood and susceptibility to the chronic diseases of aging: Moving toward a model of behavioral and biological mechanisms. Psychological Bulletin, 137, 959-997. doi:10.1037/a0024768

34.

D. L.

Jerman

Purewal Boparai

S. K.

Koita

Briner

Bucci

Harris

N. B.

(2018). Review of tools for measuring exposure to adversity in children and adolescents. Journal of Pediatric Health Care, 32, 564-583. doi:10.1016/j.pedhc.2018.04.021

35.

Runyan

D. K.

Cox

C. E.

Dubowitz

Newton

R. R.

Upadhyaya

Kotch

J. B.

. . . Knight

E. D.

(2005). Describing maltreatment: Do child protective service reports and research definitions agree? [Special issue]. Child Abuse & Neglect, 29, 461-477. doi:10.1016/j.chiabu.2004.06.015

36.

Sledjeski

E. M.

Dierker

L. C.

Brigham

Breslin

(2008). The use of risk assessment to predict recurrent maltreatment: A Classification and Regression Tree Analysis (CART). Prevention Science, 9, 28-37. doi:10.1007/s11121-007-0079-0

37.

Slep

A. M. S.

Heyman

R. E.

(2006). Creating and field-testing child maltreatment definitions: Improving the reliability of substantiation determinations. Child Maltreatment, 11, 217-236. doi:10.1177/1077559506288878

38.

Taussig

H. N.

Culhane

S. E.

Garrido

Knudtson

M. D.

Petrenko

C. L. M.

(2013). Does severity of physical neglect moderate the impact of an efficacious preventive intervention for maltreated children in foster care? Child Maltreatment, 18, 56-64. doi:10.1177/1077559512461397

39.

Thompson

Wiley

T. R.

(2009). Predictors of re-referral to child protective services: A longitudinal follow-up of an urban cohort maltreated as infants. Child Maltreatment, 14, 89-99. doi:10.1177/1077559508325317

40.

Trickett

P. K.

Mennen

F. E.

Kim

Sang

(2009). Emotional abuse in a sample of multiply maltreated, urban young adolescents: Issues of definition and identification. Child Abuse & Neglect, 33, 27-35. doi:10.1016/j.chiabu.2008.12.003

41.

Way

Chung

Jonson-Reid

Drake

(2001). Maltreatment perpetrators: A 54-month analysis of recidivism. Child Abuse & Neglect, 25, 1093-1108. doi:10.1016/S0145-2134(01)00258-7

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB