Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 28.
Published in final edited form as: Child Youth Serv Rev. 2019 Jan 22;99:81–86. doi: 10.1016/j.childyouth.2019.01.027

Caseworker assessment of child risk and functioning and their relation to service use in the child welfare system

Christian M Connell a,b,*, Christopher T Bory c, Cindy Y Huang d, Maegan Genovese b, Colleen Caron e, Jacob Kraemer Tebes b
PMCID: PMC8318199  NIHMSID: NIHMS1667990  PMID: 34326564

Abstract

Children within the child welfare system are more likely to experience emotional and behavioral problems than children not involved with the system. Many states have adopted standardized risk and assessment measures to inform decision-making on appropriate levels of care related to placement or service intensity for children within the system. This study examined the relationship of caseworker ratings of risk across multiple domains to youth functioning and service use for a sample of children open to the child welfare system. The study identified a stratified random sample of youth who were between the ages of five and 21 and open to the child welfare system (n = 184). Stratification was based on current placement (i.e., in-home, foster home, congregate care, and juvenile justice placements). Administrative data was used to access caseworker ratings of risk across child, parent, and family domains using a standardized risk assessment tool. Children’s caseworkers (n = 103) completed a standardized measure of child functioning and reported on youth utilization of services across multiple sectors including specialty mental health, school-based, juvenile justice, and medical settings. Regression analyses using variance-corrected estimation for clustered data (by caseworker) revealed higher levels of child risk were associated with poorer child functioning, which, in turn, were associated with higher rates of multi-sector service use. Recommendations and future directions are discussed.

Keywords: Risk assessment, Child welfare services, Mental health services, Children in child welfare

1. Introduction

Numerous studies have shown that children and youth involved with the child welfare system (CWS) have significantly elevated rates of behavioral health problems, with as many as 80% having emotional or behavioral disorders, developmental delays, or other evidence indicating the need for behavioral health services (Burns et al., 2004; Farmer et al., 2001; Farmer et al., 2010; Horwitz et al., 2012; Horwitz et al., 2012; Kolko, Selelyo, & Brown, 1999; Taussig, 2002). National estimates suggest, for example, that nearly 48% of youth between the ages of two and 14 involved in a child maltreatment investigation have clinically elevated mental health needs (Burns et al., 2004). Newer data suggest that rates of mental health needs may have declined in recent years (Stein et al., 2016), but youth known to CWS continue to demonstrate higher levels of clinical need than the general population.

Given the level of need, state child welfare agencies function as ‘de facto’ public behavioral health systems responsible for identification and treatment of behavioral health problems for youth in their care (Lyons & Rogers, 2004, p. 971). Youth involved with child welfare systems, for instance, have disproportionately higher contact with mental health services than youth involved in other social service systems (Farmer et al., 2001; Halfon, Berkowitz, & Klee, 1992; Staudt, 2003; Takayama, Bergman, & Connell, 1994). To more efficiently provide access to child welfare services, state systems increasingly have adopted managed care practices to allocate behavioral health resources to children and families receiving services (Courtney, 2000; Stratton, 2005). The shift to a managed care practice environment places a premium on identifying those children and families receiving child welfare services who are most in need of targeted behavioral health services.

Consistent with an ecological-transactional model of child maltreatment (Cicchetti, Toth, & Maughan, 2000), many youth involved with the CWS experience elevated rates of exposure to risk across multiple domains, including parental substance use or mental illness, domestic violence, and poverty (Lau et al., 2005; Litrownik et al., 2005; Lynch & Cicchetti, 1998; Manly, Cicchetti, & Barnett, 1994). This exposure further increases the likelihood of negative emotional and behavioral health outcomes for children (Attala, Bauza, Pratt, & Vieira, 1995; Burns et al., 2004; Cicchetti & Lynch, 1993; Ireland & Widom, 1994; Masten & Wright, 1998; Raviv, Taussig, Culhane, & Garrido, 2010) and is associated with negative safety and permanency outcomes for children involved with the child welfare system (Connell, Bergeron, Katz, Saunders, & Tebes, 2007; Connell, Katz, Saunders, & Tebes, 2006; Courtney & Wong, 1996). Risk in multiple contexts, as well as the evelvated level of behavioral health need and case complexity suggests need for a multi-sector response with services and supports drawn from multiple child-serving systems (Farmer et al., 2010).

Despite evidence of significant need for behavioral health services, the rate of access among the child welfare population remains low. Only one-quarter to one-third of children receive specialty mental health services after entering the child welfare system (Burns et al., 2004; Horwitz, Hurlburt, Goldhaber-Fiebert, et al., 2012; Horwitz, Hurlburt, Heneghan, et al., 2012), and only one-third of children receive services from multiple child-serving sectors (e.g., medical, behavioral, school, or juvenile justice systems) within 18 months of child welfare system contact (Farmer et al., 2010). Research does suggest that rates of service use within CWS are targeted toward those at greatest level of need based on caseworker ratings of risk, clinical functioning, and placement in more restrictive settings (e.g., congregate care), particularly within the context of managed care settings (Fong, French, Rubin, & Wood, 2015; Horwitz, Hurlburt, Goldhaber-Fiebert, et al., 2012; Horwitz, Hurlburt, Heneghan, et al., 2012; Landsverk, Garland, & Leslie, 2002; Leslie, Hurlburt, Landsverk, Barth, & Slymen, 2004; Raghavan et al., 2006; Staudt, 2003; Yampolskaya, Sharrock, Clark, & Hanson, 2017).

The present study expands our understanding of the relation of caseworker assessments of risk severity and child functioning to predict individual- and multi-sector service use among children open to CWS. We examine two primary questions: (1) what is the relationship of caseworker ratings of child, parent, and family risk to child functioning; and (2) what is the relation of caseworker ratings of risk and child functioning to the child’s individual- and multi-sector service use? We hypothesized that increased risk exposure across parent and child domains would be associated with poorer child functioning. Although caseworker ratings of risk should inform service planning, we hypothesized that their ratings of the child’s recent behavioral functioning would have a stronger relationship with both individual and multi-sector service use (i.e., that service use would be more closely aligned with current behavioral need than ratings of child-, parent-, and family-level risk).

2. Methods

2.1. Design

We obtained administrative data records and surveyed child welfare caseworkers based on a statewide, stratified random sample of children and youth between the ages of 5 and 21 with open cases in the child welfare system who had a caseworker complete a state-required child and family risk assessment during a 3-month study period (May–August 2011). Open cases were those in which the child or family were assigned a child welfare or juvenile justice caseworker and included youth in either in-home or out-of-home placements due to child maltreatment/child welfare needs, behavioral health needs, or juvenile justice system involvement. Risk assessment data (described below) and other administrative data (e.g., age, caseworker assignment, type of placement) were extracted from the Rhode Island Children’s Information System (RICHIST; the child welfare management information system). We used de-identified administrative data, and caseworkers only referred to the child and case by a unique child ID to maintain confidentiality of the children and families. University IRB approval was obtained for this research.

As part of a larger study of the relation of caseworker risk assessment to child welfare placement types (Huang, Bory, Caron, Tebes, & Connell, 2014), we identified a statewide sample of 1735 children with a valid risk assessment during an eight-month period in 2011. Youth below age 5 were excluded to ensure age-level requirements for the caseworker assessment measures of child functioning. To minimize variability associated with recall among caseworkers, we limited the sample to those youth who had been assessed by their caseworker during the three months prior to implementation of caseworker surveys for the study. In addition, we excluded youth who were opened to an adoption or guardianship subsidy placement, because youth in these placements typically did not have a caseworker assigned and were not likely to be referred to services within the state’s system of care. For family-level cases with more than one child involved in the CWS who met study criteria, we randomly selected one child for inclusion. These criteria resulted in a final statewide cohort of 936; the follow-up sampled consisted of a random sample reflecting approximately 25% of this cohort (n = 218 cases) stratified by child placement. To minimize staff burden we limited the sampling process to a maximum of three cases per caseworker. We replaced cases in which this limit was exceeded with a randomly selected case for caseworkers who had fewer than three cases selected. Caseworkers for selected children were mailed an assessment packet with the study instruments and instructions for completion and secure return of materials. A total of 190 packets were returned, and 184 were sufficiently complete to be included in the final study sample (84.4% completion rate). A total of 103 caseworkers provided completed assessment packets: 45% provided information on a single child, 32% on two children, and 23% on three children from their caseload.

2.2. Sample

Demographic and placement descriptive information, as well as mean scores for caseworker ratings of child-related risk and functioning measures are presented in Table 1. The sample included slightly more males (53.8%). Half of the sample were identified as Caucasian(49.5%), followed by Hispanic (26.6%), African American (13.0%), and other racial/ethnic groups (8.2%) – racial/ethnic information was missing for approximately 3% of youth. The mean age was 14.3 years (SD = 3.6 years). Over half (52.2%) of the sample had a previous substantiated maltreatment incident, and 81.0% had a history of out-of-home placement. Current living arrangements for the child welfare sample were as follows: 29.3% percent in-home, 21.7% foster home (including both non-relative and kinship foster care placements), 26.6% congregate care (including group home, shelter, and institutional placements), and 22.3% juvenile justice (including in-home and detention placements).

Table 1.

Child demographic characteristics and ratings of child-level risk and functioning (N = 184).

Overall
Mean SD
Child age 14.3 years 3.6 years
N %
Gender
 Male 99 53.8
 Female 85 46.2
Race/ethnicity
 African American 24 13.0
 Hispanic 49 26.6
 Caucasian 91 49.5
 Other race/ethnicity 15 8.2
 Unknown 5 2.7
Prior history maltreatment 88 52.2
Prior history of placement 149 81.0
Current placement
 In-home (CWS) 54 29.3
 Foster home 40 21.7
 Congregate care 49 26.6
 Juvenile justice 41 22.3
Mean SD
Risk domain
 Parental risk 1.9 3.0
 Youth risk 2.8 2.6
 Family risk 0.4 1.1
Child functioning (CAFAS-CAI total score) 80.0 60.6
Open in a new tab

2.3. Measures

2.3.1. Demographic and case history covariates

Administrative data included basic demographic indicators of age, gender, and racial/ethnic identity, as well as history of substantiated maltreatment, history of out-of-home placement prior to the current episode, and current placement setting (coded to reflect in-home, foster home, congregate care, and juvenile justice placement). Because few youth were in a juvenile justice detention placement, youth in either in-home or detention placements were combined to reflect this latter placement setting.

2.3.2. Child and family risk

Caseworker appraisals of risk were obtained from administrative records using a standardized risk assessment tool completed at placement intake that has been shown to predict placement restrictiveness (Huang et al., 2014). The instrument assesses risk across 17 domains comprising three subscales: parenting risk (e.g., parental bonding, financial stability, mental health, substance use, parental history of child abuse, and history of violence or criminal behavior), youth risk (e.g., mental health or developmental risks, substance use, educational risk, permanency risk, and medical/dental risks), and family risk (e.g., family violence, access to kinship or related supports, and family financial stability).

Risk level for each domain was rated on a four-point scale of 0 (no risk), 1 (low risk), 2 (moderate risk), to 3 (high risk); each level includes specific behavioral criteria (e.g., “Use of substances results in emotionally abusive and/or violent behavior” is classified as “High Risk” on the child substance use domain). Based on a comprehensive case review, caseworkers rate a child on each domain. Subscale scores are calculated by summing items within each domain. The 8-item parent risk domain has a range of 0–24, the 6-item youth risk domain has a range of 0–18, and the 3-item family risk domain has a range of 0–9.

2.3.3. Child functioning

The CAFAS Checklist for Adult Informant (CAFAS-CAI; Hodges, 1995) was used to gather information to assess a youth’s behavior within the past three months across eight domains (school/work, home, community, behavior toward others, moods/emotions, self-harmful behavior, substance use, and thinking). The CAFAS-CAI consists of a series of statements about the youth’s behavior that are rated as true or false. Caseworkers indicated whether the youth engaged in specific behaviors during the previous three-month period, as well as the frequency or severity of these behaviors. Responses were converted to a CAFAS score through a coding process developed by study team members who had completed formal CAFAS training and certification. Items from the CAFAS-CAI were mapped to the original CAFAS instrument and a series of decision rules were applied to mirror the criteria for assigning CAFAS scores to each functional domain. These decision rules were reviewed in team meetings to ensure consensus and to set criteria when the language of the CAFAS-CAI was not identical to that of the original CAFAS instrument. Consistent with the original instrument, CAFAS-CAI scores for each domain were rated as minimal or no impairment (0), mild (10), moderate (20), or severe (30), and a CAFAS-CAI total score was calculated by summing the subscale scores on each of the eight domains, with a range of 0 to 240. Higher scores are indicative of greater levels of impairment; scores of 100 or more are indicate marked impairment, and scores of 140 or more indicate severe impairment (Hodges, 2000).

2.3.4. Service use

The Child and Adolescent Service Assessment (CASA; Ascher, Farmer, Burns, & Angold, 1996; Farmer, Angold, Burns, & Costello, 1994) was used to assess child utilization of services across a range of settings and domains. An adapted version of the CASA was used to collect service use across 40 items comprising four primary service domains: (1) specialty mental health services (e.g., mental health center, inpatient alcohol or drug treatment unit); (2) school (e.g., school guidance counselor); (3) justice system (e.g., Probation Officer or Juvenile Correction Counselor); and (4) general medical service (e.g., family doctor for any identified problems). The caseworker was asked whether the youth had accessed each service type within the past three months. For purposes of the present analyses, any affirmative response within a given sector was coded to reflect that the youth had accessed a particular service type (e.g., specialty mental health, school-based services, etc.). In addition, we calculated a multi-sector service use indicator, defined as the use of services from more than one sector to represent the need for more networked and integrated services. Farmer et al. (2010) calculated a similar indicator in previous research on service use within a system of care.

2.4. Data analysis

Multiple regression was used to examine the relationship between caseworker ratings of risk and child functioning, while controlling for demographic characteristics, history of prior substantiated maltreatment or out-of-home placement, and current placement setting. Logistic regression analyses were used to examine the relationship between risk, child functioning and service utilization (i.e., multi-sector, specialty mental health, school, justice system, and general medical), while controlling for demographic characteristics, prior history of maltreatment and out-of-home placement, and current placement setting. All analyses used cluster-robust standard errors to account for potential within-caseworker-level correlated error. Analyses were conducted using Stata 14.2 (StataCorp LLC, 2018).

3. Results

3.1. Risk and child functioning

The first set of analyses examined the relation of caseworker ratings of child, caregiver, and family risk to ratings of child functioning, while controlling for child demographic characteristics and indicators of child welfare involvement (see Table 2). The overall regression model fit the data well (χ2(13) = 91.05, p < .001) and accounted for significant variation in caseworker ratings of functional impairment (overall R2 = 0.37). Caseworker ratings of youth risk were associated with significantly greater impairment in functioning (β = 0.24); parent and family risk indicators were not significantly associated with youth functioning. Older youth experienced greater impairments in functioning (β = 0.22), as did those with a history of out-of-home placement (β = 0.18). Current placement was also associated with significant effects on ratings of functioning, youth in foster home placements were rated more favorably than those in in-home placements (β = −0.19), while those in congregate care placements had significantly greater ratings of impairment (β = 0.27).

Table 2.

Relation of casework risk assessment to youth functioning.

B Robust SE 95% CI
Lower Upper
Constant 80.92** 4.30* 72.50 89.34
Age 3.73** 1.07 1.63 5.84
Gender (male)a −5.50 7.18 −19.58 8.58
Raceb
 African American 13.89 11.48 −8.60 36.38
 Hispanic −9.31 9.37 −27.67 9.06
 Other −9.51 16.79 −42.42 23.41
Child welfare history
 Prior maltreatment −2.10 10.81 −23.28 19.09
 Prior placement 27.05 10.74 6.00 48.10
Current placementc
 Foster home −28.35** 10.02 −47.99 −8.70
 Congregate care 37.28** 11.50 14.74 59.83
 Juvenile justice −5.08 12.54 −29.65 19.50
Risk
 Parent risk score 1.63 1.48 −1.27 4.53
 Youth risk score 5.49** 1.63 2.29 8.70
 Family risk score 1.81 3.44 −4.93 8.56
Open in a new tab

Note. CI = confidence interval for B.

a

Reference is female.

b

Reference is White.

c

Reference is In-Home CWS.

**

p < .01.

*

p < .05.

3.2. Risk, child functioning, and service utilization

The final set of analyses examined the relation of caseworker ratings of risk and youth functioning to multi-sector service utilization, as well as likelihood of specialty mental health, school-based, juvenile justice, and medical service use to address behavioral health issues (see Table 3). Caseworkers indicated that the majority of youth were receiving specialty mental health services (83.1%), 49.4% used school-based services, 33.9% had justice-related services, and 33.7% used general medical services; two-thirds (66.3%) received services from two or more sectors.

Table 3.

Relation of caseworker risk assessment and youth functioning to service use.

Predictors Multi-sector Specialty mental health School Justice General medical
OR [95% CI] OR [95% CI] OR [95% CI] OR [95% CI] OR [95% CI]
Constant 0.39t [0.08, 1.92] 1.36 [0.20, 9.13] 0.13* [0.02, 0.94] 0.00** [0.00, 0.01] 1.07 [0.16, 7.00]
Age 0.93 [0.83, 1.04] 0.95 [0.83, 1.08] 1.03 [0.92, 1.16] 1.76** [1.17, 2.66] 0.86* [0.76, 0.98]
Gender (male)a 3.23** [1.39, 7.50] 1.84 [0.62, 5.46] 1.92 [0.92, 4.00] 22.66* [1.95, 263.99] 0.67 [0.26, 1.70]
Raceb
 African American 0.95 [0.26, 3.49] 0.33 [0.05, 1.97] 1.22 [0.30, 4.91] 10.53* [1.06, 104.91] 0.23* [0.07, 0.78]
 Hispanic 1.70 [0.59, 4.91] 0.74 [0.23, 2.41] 1.16 [0.49, 2.74] 7.65** [1.75, 33.51] 0.47 [0.17, 1.34]
 Other 0.77 [0.17, 3.41] 0.53 [0.07, 3.69] 2.02 [0.54, 7.57] 21.60** [2.59, 179.93] 0.08** [0.01, 0.48]
Child welfare history
 Prior maltreatment 1.31 [0.44, 3.90] 1.30 [0.29, 5.81] 1.40 [0.57, 3.44] 0.34 [0.05, 2.17] 1.29 [0.50, 3.29]
 Prior placement 0.51 [0.13, 1.96] 1.25 [0.34, 4.67] 0.45 [0.14, 1.49] 0.12* [0.02, 0.80] 0.96 [0.22, 4.20]
Current placementc
 Foster care 1.71 [0.52, 5.62] 1.65 [0.42, 6.44] 1.16 [0.39, 3.40] 1.27 [0.17, 9.37] 0.93 [0.27, 3.17]
 Congregate care 2.41 [0.54, 10.64] 1.38 [0.18, 10.61] 2.28 [0.68, 7.68] 1.12 [0.33, 3.79] 0.85 [0.25, 2.93]
 Juvenile justice 2.40 [0.67, 8.65] 0.52 [0.12, 2.26] 0.49 [0.14, 1.74] 573.89** [33.78, 9749.85] 0.41 [0.13, 1.31]
Risk rating
 Parent risk 0.76** [0.63, 0.92] 0.72** [0.59, 0.89] 0.88 [0.75, 1.03] 0.98 [0.65, 1.47] 0.88 [0.76, 1.03]
 Youth risk 0.94 [0.76, 1.17] 1.08 [0.83, 1.41] 0.95 [0.80, 1.12] 1.04 [0.84, 1.28] 1.18* [1.01, 1.37]
 Family risk 1.18 [0.71, 1.97] 2.14** [1.21, 3.80] 1.12 [0.71, 1.76] 1.38 [0.55, 3.45] 0.96 [0.62, 1.50]
Child functioningd 1.48** [1.29, 1.69] 1.48** [1.18, 1.87] 1.27** [1.16, 1.39] 1.16 [0.99, 1.35] 1.25** [1.14, 1.36]
Open in a new tab

Note. CI = confidence interval for odds ratio (OR).

a

Reference is female.

b

Reference is White.

c

Reference is in-home placement.

d

Re-scaled for 10-point increase.

**

p < .01.

*

p < .05.

t

p < 0.10.

Model 1 (multi-sector services) provided a good fit to the data (χ2(14) = 46.64, p < .001). Results indicated increased child functional impairment was associated with greater likelihood of multi-sector service use (OR = 1.48); each 10-point increase in CAFAS-CAI total score was associated with a nearly 50% increase in the likelihood of multi-sector service use. Males also had increased odds of mult-sector service use (OR = 3.23), while higher ratings of parental risk were associated with reduced odds of multi-sector service use (OR = 0.76). Child welfare history and current placement were not significantly related to multi-sector service use after accounting for demographic indicators and caseworker ratings of risk and functioning.

Model 2 (specialty mental health services) provided a good fit to the data (χ2(14) = 39.01, p < .001). Again, increased child functional impairment was associated with greater likelihood of service use in this domain (OR = 1.48) and higher ratings of parental risk were associated with reduced odds of use (OR = 0.72). In addition, higher ratings of family risk were associated with greater likelihood of specialty mental health service use (OR = 2.14); other factors were not significantly related.

Model 3 (school services) provided a good fit to the data (χ2(14) = 45.96, p < .001). After controlling for demographic and child welfare-related covariates, only caseworker ratings of impaired functioning predicted greater likelihood of school-related service use (OR = 1.27).

Model 4 (justice-related services) provided a good fit to the data (χ2(14) = 36.47, p < .001). Ratings of child functional impairment were not only associated with likelihood of service use in this domain, though a number of demographic and child welfare indicators were also associated with increased likelihood including older youth (OR = 1.76), males (OR = 22.66), and youth of minority racial or ethnic backgrounds. Youth with a history of prior placement were less likely to utilize juvenile justice services (OR = 0.12). Finally, youth were were involved in a juvenile justice placement (either in-home or in detention) were significantly more likely to be accessing justice-related services than those in in-home placement (OR = 573.89), as would be expected since assignment to a probation officer constituted one of the service-related indicators and a majority of youth in in-home juvenile placements has such an assignment.

Finally, model 5 (general medical services) provided a good fit to the data (χ2(14) = 75.57, p < .001). As with each of the other outcomes, increased child functional impairment was associated with greater likelihood of service use in this domain (OR = 1.25). Higher ratings of youth risk were also associated with increased odds of use (OR = 1.18). Older youth were less likely to be receiving general medical services (OR = 0.86, as were African American (OR = 0.23) or youth of other racial/ethnic backgrounds (OR = 0.08). Other child welfare involvement indicators were not significantly related to service use in this domain.

4. Discussion

The present study furthers our understanding of the relation of risk exposure in the child and family domains to child functioning, and the joint influence each has on service use within an active child welfare population including children in in-home, foster home, congregate care, and juvenile justice placements. Results provided limited supported of our hypotheses of the association between caseworker ratings of child and family risk and child functioning, and more consistent support of our hypotheses about the relation of impaired child functioning and utilization of behavioral health services within and across various sectors. With respect to the first study question, higher levels of caseworker-appraised youth risk were associated with a moderate increase in ratings of functional impairment. However, caseworker ratings of parent and family risk were not related significantly to ratings of child functioning. In addition to effects of risk appraisal, youth with a history of out-of-home placement had more negative caseworker ratings of their functioning, while current placement setting presented a mixed set of findings – youth in foster home placements had lower ratings of functional impairment, and youth in congregate placements had higher ratings of impairment relative to those in in-home placement.

With respect to the second set of analyses, caseworker ratings of child functioning were strongly and consistently related to service use within three of the four sectors assessed, as well as multi-sector service use across domains. The one exception to this finding was for juvenile justice services, which were primarily associated with demographic (e.g., older, male, minority youth) and placement (e.g., in- and out-of-home justice placement) characteristics. Also consistent with hypotheses, child functioning, rather than caseworker appraisal of risk, served as the primary predictor of service involvement. In the full model, caseworker ratings of youth, parent, and family risk were minimally related to service use outcomes after other factors were accounted for, with only a couple of exceptions. Youth risk was positively related to general medical service use, while parental risk scores actually were associated with lower rates of specialty mental health and multi-sector service use.

Our findings suggest that the types of youth risk indicators frequently assessed by caseworkers as part of CWS intake and service planning process are associated with Children’s behavioral functioning, but that level of functioning is a stronger predictor of service utilization than risk appraisal, even though case practices often link risk assessment directly to service planning. The lack of association between indicators of parent/family risk and child functioning was somewhat surprising, and counter to our expectations. We had hypothesized that these contextual risks would negatively influence child functioning, but it may be that caseworkers are more attuned to caregiver and family-level risks that influence child safety or permanency decisions, rather than risks associated with behavioral functioning – while youth-related risks may reflect more directly on their behavioral functioning.

The lack of association, or in some cases the negative association, between risk indicators and service use is also of interest. The express purpose of the caseworker risk assessment used in this study was to guide treatment planning, so we anticipated that the level of risk across different domains would be related to differences in patterns of service use. These findings suggest that multi-sector services are more closely aligned with a child’s current behavioral functioning than ratings of risk, which may have stronger role in determinations of placement and restrictiveness (e.g., Huang et al., 2014). Our findings also are consistent with a those reported by Farmer et al. (2010), who found that the presence of parental risks such as mental health or substance use problems interefered with the ability of a family to access child-focused services. Alternatively, it may be that for households where parental risk factors are observed, the parent or caregiver becomes the focus of service referrals, rather than the child. Our study focused specifically on child service utilization, and information related to caregiver service referrals or utilization was not available. This relation merits further examination to understand the mechanisms by which parental risk factors influence child-focused service access or become an orienting framework for service referrals instead of child-focused supports.

5. Study limitations

These results should be viewed within the context of some potential limitations. Though the sample was stratified and representative of each placement type, the sample size was relatively small and limited to a single statewide sample. Some caution is warranted in generalizing these results to different state samples. Another limitation is the concurrent assessment of child functioning and service utilization in the current study design. Although risk assessment was completed independently of ratings of child functioning and service utilization, the latter two domains were completed at the same assessment period. To minimize contamination, caseworkers were asked to complete the ratings of functioning based on their knowledge of the youth and then asked to report service-related information based on those services and supports the family had received over the past 3-month period. It is possible that knowledge of services may have influenced caseworker ratings of functioning. Further research utilizing intake ratings of both risk and functioning to subsequent service referrals and utilization would strengthen our understanding of the joint influences of risk and functioning in this area. Finally, the application of the risk assessment tool, which is a clinical decision-making tool used by caseworker, may serve as another limitation. It is possible that certain risk factors are missed based on the training and background of the caseworker and the setting in which they work and interaction with youth (e.g., social worker in child welfare vs. probation officer). These limitations call for a need for future studies to examine differences in ratings of youth risks and functioning based on caseworker setting, and how these differences may impact service referrals within CWS.

6. Conclusions

CWS are increasingly focused on incorporating system of care principles and practices to ensure that children and families with complex, multi-service sector needs receive essential services (Fluke & Oppenheim, 2010; Stroul, Blau, & Friedman, 2010). Central aspects of such an approach include the use of effective screening and assessment tools to identify children in need of services and supports, as well as integrated or multi-sector service networks capable of providing flexible and community-based services in the least restrictive setting possible (Tebes et al., 2005). Our findings suggest that child welfare agencies may benefit from supplementing risk assessment tools with measures of child functioning to inform service planning within this context. Functioning measures, at least in the present context, may provide additional useful information about a child’s service-related needs across multiple contexts – though it is possible that parent and family-level risk assessment provide valuable information to inform family-focused child welfare services.

Acknowledgments

This project was funded through the Rhode Island Department of Children, Youth, and Families (RIDCYF). Support for Dr. Huang was provided by a National Institute on Drug Abuse (NIDA) funded Postdoctoral Training Program (T32 DA019426). The authors would like to acknowledge the following people for their support and feedback on this study: The authors wish to acknowledge the contributions of assistance and support made by Lauren Moss-Racusin, The Consultation Center, and Leon Saunders and David Allenson, RIDCYF, to the completion of this article. In addition, the authors want to thank members of the Division of Prevention and Community Research, Yale University School of Medicine, for helpful comments and suggestions on an earlier draft of the article.

Footnotes

Conflict of interest

None.

References

  1. Ascher BH, Farmer EM, Burns BJ, & Angold A (1996). The child and adolescent services assessment (CASA): Description and psychometrics. Journal of Emotional and Behavioral Disorders, 4(1), 12–20. 10.1177/106342669600400102. [DOI] [Google Scholar]
  2. Attala JM, Bauza K, Pratt H, & Vieira D (1995). Integrative review of effects on children witnessing domestic violence issues. Comprehensive Pediatric Nursing, 18(3), 163–172. [DOI] [PubMed] [Google Scholar]
  3. Burns BJ, Phillips SD, Wagner HR, Barth RP, Kolko DJ, Campbell Y, & Landsverk J (2004). Mental health need and access to mental health services by youth involved with child welfare: A national survey. Journal of the American Academy of Child and Adolescent Psychiatry, 43(8), 960–970. 10.1097/01.chi.0000127590.95585.65. [DOI] [PubMed] [Google Scholar]
  4. Cicchetti D, & Lynch M (1993). Toward an ecological/transactional model of community violence and child maltreatment: Consequences for Children’s development. Psychiatry, 56, 96–118. [DOI] [PubMed] [Google Scholar]
  5. Cicchetti D, Toth SL, & Maughan A (2000). An ecological–transactional model of child maltreatment. In Sameroff AJ, & Lewis M (Eds.). Handbook of developmental psychopathology (pp. 689–722). (2nd edition). New York: Kluwer Academic/Plenum Press. [Google Scholar]
  6. Connell CM, Bergeron N, Katz KH, Saunders L, & Tebes JK (2007). Re-referral to child protective services: The influence of child, family, and case characteristics on risk status. Child Abuse & Neglect, 31, 573–588. 10.1016/j.chiabu.2006.12.004. [DOI] [PubMed] [Google Scholar]
  7. Connell CM, Katz KH, Saunders L, & Tebes JK (2006). Leaving foster care - the influence of child and case characteristics on foster care exit rates. Children and Youth Services Review, 28, 780–798. 10.1016/j.childyouth.2005.08.007. [DOI] [Google Scholar]
  8. Courtney ME (2000). Managed care and child welfare services: What are the issues? Children and Youth Services Review, 22, 87–91. 10.1016/S0190-7409(00)00067-0. [DOI] [Google Scholar]
  9. Courtney ME, & Wong YI (1996). Comparing the timing of exits from substitute care. Children and Youth Services Review, 18(4–5), 307–334. 10.1016/0190-7409(96)00008-4. [DOI] [Google Scholar]
  10. Farmer EM, Angold A, Burns BJ, & Costello EJ (1994). Reliability of self-reported service use: Test–retest consistency of Children’s responses to the child and adolescent services assessment. Journal of Child and Family Studies, 3, 307–325. 10.1007/BF02234688. [DOI] [Google Scholar]
  11. Farmer EM, Burns BJ, Chapman M, Phillips SD, Angold A, & Costello EJ (2001). Use of mental health services by youth in contact with social services. Social Service Review, 75(4), 605–624. 10.1086/323165. [DOI] [Google Scholar]
  12. Farmer EM, Mustillo S, Wagner HR, Burns BJ, Kolko DJ, Barth RP, & Leslie LK (2010). Service use and multi-sector use for mental health problems by youth in contact with child welfare. Children and Youth Services Review, 32, 815–821. 10.1016/j.childyouth.2010.01.019. [DOI] [Google Scholar]
  13. Fluke JD, & Oppenheim E (2010). Getting a grip on systems of care and child welfare using opposable thumbs. Evaluation and Program Planning, 33, 41–44. 10.1016/j.evalprogplan.2009.05.012. [DOI] [PubMed] [Google Scholar]
  14. Fong HF, French B, Rubin D, & Wood JN (2015). Mental health services for children and caregivers remaining at home after suspected maltreatment. Children and Youth Services Review, 58, 50–59. 10.1016/j.childyouth.2015.08.010. [DOI] [Google Scholar]
  15. Halfon N, Berkowitz G, & Klee L (1992). Mental health service utilization by children in foster care in California. Pediatrics, 89(6), 1238–1244. [PubMed] [Google Scholar]
  16. Hodges K (1995). CAFAS checklist for adult informant. Ann Arbor, MI: Ypsilanti: EasternMichigan University. [Google Scholar]
  17. Hodges K (2000). Child and adolescent functional assessment scale (2nd revision). AnnArbor, MI: Ypsilanti: Eastern Michigan University. [Google Scholar]
  18. Horwitz SM, Hurlburt MS, Goldhaber-Fiebert JD, Heneghan AM, Zhang J, Rolls-Reutz J, … Stein RE (2012). Mental health services use by children investigated by child welfare agencies. Pediatrics, 130, 861–869. 10.1542/peds.2012-1330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Horwitz SM, Hurlburt MS, Heneghan A, Zhang J, Rolls-Reutz J, Fisher E,Landsverk J, … Stein RE (2012). Mental health problems in young children investigated by U.S. child welfare agencies. Journal of the American Academy of Child & Adolescent Psychiatry, 51(6), 572–581. 10.1016/j.jaac.2012.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Huang CY, Bory CT, Caron C, Tebes JK, & Connell CM (2014). Relationship of risk assessment to placement characteristics in a statewide child welfare population. Children and youth services review, 46, 85–90. 10.1016/j.childyouth.2014.07.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Ireland T, & Widom CS (1994). Childhood victimization and risk for alcohol and drug arrests. International Journal of Addiction, 29, 235–274. 10.3109/10826089409047380. [DOI] [PubMed] [Google Scholar]
  22. Kolko DJ, Selelyo J, & Brown EJ (1999). The treatment histories and service involvement of physically and sexually abusive families: Description, correspondence, and clinical correlates. Child Abuse & Neglect, 23(5), 495–576. 10.1016/S0145-2134(99)00022-8. [DOI] [PubMed] [Google Scholar]
  23. Landsverk J, Garland AF, & Leslie LK (2002). Mental health services for children reported to child protective services. In Myers JEB, & Briere J (Eds.). The APSAC handbook on child maltreatment(2nd edition). Thousand Oaks, CA: Sage. [Google Scholar]
  24. Lau AS, Leeb RT, English D, Graham JC, Briggs EC, Brody KE, & Marshall JM (2005). What’s in a name? A comparison of methods for classifying predominant type of maltreatment. Child Abuse & Neglect, 29, 533–551. 10.1016/j.chiabu.2003.05.005. [DOI] [PubMed] [Google Scholar]
  25. Leslie LK, Hurlburt MS, Landsverk J, Barth R, & Slymen DJ (2004). Outpatient mental health services for children in foster care: A national perspective. Child Abuse & Neglect, 28(6), 697–712. 10.1016/j.chiabu.2004.01.004. [DOI] [PubMed] [Google Scholar]
  26. Litrownik AJ, Lau A, English DJ, Briggs E, Newton RR, Romney S, & Dubowitz H (2005). Measuring the severity of child maltreatment. Child Abuse & Neglect, 29, 553–573. 10.1016/j.chiabu.2003.08.010. [DOI] [PubMed] [Google Scholar]
  27. Lynch M, & Cicchetti D (1998). An ecological-transactional analysis of children and contexts: The longitudinal interplay among child maltreatment, community violence, and Children’s symptomatology. Development and Psychopathology, 10, 235–257. 10.1017/S095457949800159X. [DOI] [PubMed] [Google Scholar]
  28. Lyons JS, & Rogers L (2004). The US Child Welfare System: A de facto public behavioral health care system. Journal of the American Academy of Child and Adolescent Psychiatry, 43(8), 971–973. 10.1097/01.chi.0000127575.23736.90. [DOI] [PubMed] [Google Scholar]
  29. Manly JT, Cicchetti D, & Barnett D (1994). The impact of subtype, frequency, chronicity, and severity of child maltreatment on social competence and behavior problems. Development and Psychopathology, 6(121), 143. 10.1017/S0954579400005915. [DOI] [Google Scholar]
  30. Masten AS, & Wright MO (1998). Cumulative risk and protection models of child maltreatment. In Rossman BBR, & Rosenberg MS (Eds.). Multiple victimization of children: Conceptual, developmental, research and treatment issues (pp. 7–30). NY: Haworth Press. 10.1300/J146v02n01_02. [DOI] [Google Scholar]
  31. Raghavan R, Leibowitz AA, Andersen RM, Zima BT, Schuster MA, &Landsverk J (2006). Effects of Medicaid managed care policies on mental health service use among a national probability sample of children in the child welfare system. Children and Youth Services Review, 28, 1482–1496. 10.1016/j.childyouth.2006.04.001. [DOI] [Google Scholar]
  32. Raviv T, Taussig HN, Culhane SE, & Garrido EF (2010). Cumulative risk exposure and mental health symptoms among maltreated youth placed in out-of-home care. Child Abuse & Neglect, 34(10), 742–751. 10.1016/j.chiabu.2010.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. StataCorp LLC (2018). Stata version 14.2. College Station, TX: StataCorp LLC. [Google Scholar]
  34. Staudt MM (2003). Mental health services utilization by maltreated children: Research findings and recommendations. Child Maltreatment, 8, 195–203. [DOI] [PubMed] [Google Scholar]
  35. Stein RE, Hurlburt MS, Heneghan AM, Zhang J, Kerker B, Landsverk J, & Horwitz SM (2016). For better or worse? Change in service use by children investigated by child welfare over a decade. Academic Pediatrics, 16(3), 240–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stratton P (2005). Levels of care and treatment foster care. Foster Family-based TreatmentAssociation; 1–10. [Google Scholar]
  37. Stroul B, Blau G, & Friedman R (2010). Updating the system of care concept and philosophy. Washington, DC: Georgetown University Center for Child and Human Development, National Technical Assistance Center for Children’s Mental Health. [Google Scholar]
  38. Takayama JI, Bergman AB, & Connell FA (1994). Children in foster care in the State of Washington: Health care utilization and expenditure. Journal of the American Medical Association, 271(23), 1850–1855. 10.1001/jama.271.23.1850. [DOI] [PubMed] [Google Scholar]
  39. Taussig H (2002). Risk behaviors in maltreated youth placed in foster care: A longitudinal study of protective and vulnerability factors. Child Abuse & Neglect, 26, 1179–1199. 10.1016/S0145-2134(02)00391-5. [DOI] [PubMed] [Google Scholar]
  40. Tebes JK, Bowler SM, Shah S, Connell CM, Ross E, Simmons R, … Kaufman JS (2005). Service access and service system development in a Children’s behavioral health system of care. Evaluation & Program Planning, 28, 151–160. 10.1016/j.evalprogplan.2004.10.008. [DOI] [Google Scholar]
  41. Yampolskaya S, Sharrock PJ, Clark C, & Hanson A (2017). Utilization of mental health services and mental health status among children placed in out-of-home care: A parallel process latent growth modeling approach. Child Psychiatry & Human Development, 48(5), 728–740. 10.1007/s10578-016-0699-3. [DOI] [PubMed] [Google Scholar]

RESOURCES