Table of Contents | Previous | Next |
4. Risk Assessment at Time of Investigation
The psychosocial risk factors present for a child and family when the case is brought to the attention of CWS are typically used to shape decisions about which children will go on to get formal child welfare services and which will not. The systematic collection and evaluation of child and family characteristics related to child safety is the intended function of risk assessment approaches now used in almost all child welfare jurisdictions (Mitchell et al., in press). NSCAW collected risk-assessment data similar to what are often used in child welfare practice, as well as information about how the child welfare workers weighted that information to make decisions. This information can contribute to understanding the mechanisms by which risk-assessment information is turned into service pathways. This chapter examines, by child setting, the presence of 25 risk factors, in 8 categories, as determined by the child welfare worker at the time of the investigation. Which of these factors are deemed as most critical in guiding the outcome of the case is analyzed, as well as the child welfare worker's assessment that the child would experience another report of abuse or neglect in the near future based on various levels of service provision. Over time, this study will provide further information from the child welfare worker's appraisal of the parent, home, and child characteristics. In addition, we will see how those appraisals are related to case decision-making and, ultimately, how these factors and decisions are related to child and family outcomes.
The longitudinal design of NSCAW will eventually allow for the study of the relationship of risk-assessment items to future decisions. At intake, the most important contribution of the data is to help identify the characteristics of the population coming to the attention of child welfare agencies, in order to determine what services might be designed to best address their current situations.
4.1 Risk Assessment
Risk assessment is a set of procedures used to structure information gathering and decision-making regarding child welfare cases. In our survey of child welfare managers, we determined that the vast majority of PSUs use a structured decision-making or risk assessment tool (Mitchell et al., in press). These tools vary considerably. Risk-assessment items that addressed parent, child, and secondary caregiver risks (or protective factors) were taken from these and other tools that have been used for research to help identify salient predictors of child welfare decisions. In addition, the risk assessment section included predictions from child welfare workers about the longer-term outcomes for cases, depending on the type of services provided. Although risk-assessment items ideally are used in order to generate decisions, these instruments were completed for NSCAW well after the case decisions were made and thus may not reflect how these risk factors were used in the decision-making process at the time of the investigation. The risk assessment data are an especially valuable component of the analyses, as they give a sense of the circumstances and climate in the child's home at the time of the investigation and add to our baseline picture of all NSCAW children and families.
4.1.1 History of Child Welfare Services
Prior to completion of the risk-assessment items related to family and child dynamics, child welfare workers were asked the following four questions that may have been significant in determining the child welfare worker's decision in the case: Were there any prior reports of maltreatment to the agency? Was there a prior investigation of abuse or neglect? Was there a prior incident of substantiated abuse or neglect? and Was there any prior child welfare services history, not including investigations? Although it was not specified in the questions administered, it is possible that the family orientation of many child welfare workers could have influenced them to answer questions in a way that would include prior history of the family as a whole, rather than just the study child, so it should be noted that these analyses are more family- than child-specific. This history data is important in understanding families' and children's patterns of involvement with CWS.
According to the investigative workers, approximately half (51%) of all children/families involved with CWS have had prior reports of maltreatment to the agency. The proportion for children in out-of-home care is close to two-thirds (66%), which is significantly higher (p < .001) than the proportion for children remaining at home (49%). Within the in-home subpopulation, children/families receiving child welfare services (59%) are significantly more likely to have had prior reports of maltreatment than children/families not receiving services (46%; p < .001). These findings are consistent with what might be expected, with children and families who have experienced prior reports of maltreatment receiving higher levels of service, including placement of the child out of the home. Also to be expected due to the child's age, children aged 2 and younger and their families are significantly less likely than children in all other age groups and their families to have prior reports of maltreatment to the agency, both overall and for children remaining at home (p < .001). There were no significant differences based on race/ethnicity with regard to prior reports of maltreatment (Table 4-1).
Recognizing that not all reports result in investigations, we next asked child welfare workers who indicated prior reports of maltreatment if there were prior investigations of abuse or neglect. Overall, 94% of all children/families with prior reports of maltreatment also had prior investigations of abuse or neglect. (Based on the assumption that a report would necessarily precede an investigation, this item was asked only in cases where a prior report was indicated. If this 94% were applied to the entire NSCAW population, it would indicate that 48% of the children had a prior investigation of abuse or neglect.) This proportion did not differ significantly based on child setting or age; however, children of other races/ethnicities and their families were found to be significantly more likely than White children and their families, both overall and for children remaining at home, to have prior investigations of abuse or neglect (Table 4-2). The very high proportion of prior investigations may, in part, be due to agency administrative practices that may cause reports that are screened out (and not investigated) or investigated (but not substantiated) to be difficult to access. That is, some or all reports are not kept in the records unless they result in investigations.
When child welfare workers indicated that there was a prior investigation of abuse or neglect, they were then asked about the presence of prior incidents of substantiated abuse or neglect. Over half (55%) of the children/families with prior investigations had prior incidents of substantiated abuse or neglect. As with the item on prior reports of maltreatment, children in out-of-home care and their families are significantly more likely (p < .001) than those remaining at home to have prior incidents of substantiated abuse or neglect (71% vs. 52%). There is also a trend (p = .05) within the subpopulation of children remaining at home for children/families receiving services to be more likely than children/families not receiving child welfare services to have prior incidents of substantiated abuse or neglect. There are no significant differences on this item based on age or race/ethnicity (Table 4-3).
| Characteristic | TOTAL | Setting | |||||||
|---|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | ||||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | |||
| Percent^/ (SE) | |||||||||
| Age | 0-2 | 37.6 a b c (2.7) |
28.7 (3.8) |
44.0 (3.7) |
32.8 d e f (3.0) |
72.1 (3.6) |
48.7 (8.1) |
--- | 61.1 (4.2) |
| 3-5 | 54.5 (3.3) |
51.7 (4.3) |
60.5 (5.3) |
54.2 (3.4) |
49.7 (10.8) |
60.4 (15.7) |
--- | 59.7 (9.9) |
|
| 6-10 | 50.9 (2.3) |
45.4 (3.4) |
62.4 (4.5) |
49.3 (2.5) |
64.7 (5.5) |
67.1 (9.7) |
79.9 (14.3) |
66.9 (5.6) |
|
| 11+ | 58.4 (3.0) |
53.5 (5.2) |
62.3 (4.1) |
56.1 (3.8) |
77.3 (6.6) |
81.2 (6.1) |
71.5 (12.5) |
73.2 (6.0) |
|
| Race/ Ethnicity | African American | 53.0 (2.9) |
46.1 (4.0) |
61.0 (4.7) |
50.4 (3.3) |
71.4 (9.3) |
67.2 (9.0) |
72.0 (9.8) |
69.9 (4.8) |
| White | 51.8 (1.9) |
47.2 (2.8) |
60.1 (3.7) |
50.5 (2.1) |
67.3 (5.2) |
60.4 (9.4) |
68.5 (15.7) |
62.9 (4.7) |
|
| Hispanic | 46.3 (3.5) |
43.2 (4.2) |
49.5 (4.1) |
44.8 (3.5) |
64.7 (15.4) |
55.2 (7.8) |
87.6 (9.3) |
62.6 (8.6) |
|
| Other | 49.1 (4.5) |
39.6 (6.9) |
61.4 (8.8) |
45.7 (5.3) |
74.8 (11.1) |
80.1 (10.6) |
83.7 (14.5) |
78.1 (6.3) |
|
| TOTAL | 50.9 (1.5) |
45.6 h (2.1) |
58.6 (2.3) |
49.1 g (1.6) |
69.0 (3.5) |
63.2 (6.8) |
72.5 (9.7) |
66.2 (3.3) |
|
|
a Children age 3-5 and their families are significantly more likely than children aged 0-2 and their families to have prior reports of maltreatment (X2 = 22.1, p < .001). (back) b Children age 6-10 and their families are significantly more likely than children aged 0-2 and their families to have prior reports of maltreatment (X2 = 12.7, p < .001). (back) c Children age 11+ and their families are significantly more likely than children aged 0-2 and their families to have prior reports of maltreatment (X2 = 32.5, p < .001). (back) d Children age 3-5 remaining at home and their families are significantly more likely than children aged 0-2 remaining at home and their families to have prior reports of maltreatment (X2 = 28.1, p < .001). (back) e Children age 6-10 remaining at home and their families are significantly more likely than children aged 0-2 remaining at home and their families to have prior reports of maltreatment (X2 = 16.0, p < .001). (back) f Children age 11+ remaining at home and their families are significantly more likely than children aged 0-2 remaining at home and their families to have prior reports of maltreatment (X2 = 25.2, p < .001). (back) g Children in out-of-home care and their families are significantly more likely than children remaining at home and their families to have prior reports of maltreatment (X2 = 20.6, p < .001). (back) |
| Characteristic | TOTAL | Setting | |||||||
|---|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | ||||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home^^ | |||
| Percent^/ (SE) | |||||||||
| Age | 0-2 | 92.8 (2.0) |
91.3 (3.7) |
94.4 (1.9) |
92.4 (2.5) |
97.8 (1.2) |
86.7 (6.9) |
--- | 93.8 (3.1) |
| 3-5 | 93.6 (3.5) |
91.1 (5.4) |
97.9 (1.0) |
93.2 (3.8) |
97.4 (2.1) |
99.3 (0.7) |
--- | 98.9 (0.7) |
|
| 6-10 | 93.5 (2.0) |
92.5 (3.3) |
97.3 (1.6) |
93.9 (2.4) |
79.9 (13.0) |
99.3 (0.5) |
--- | 90.9 (6.6) |
|
| 11+ | 94.2 (2.0) |
92.9 (3.2) |
95.0 (1.8) |
93.7 (2.2) |
94.0 (5.9) |
99.6 (0.4) |
94.2 (4.2) |
96.7 (2.2) |
|
| Race/ Ethnicity | African American | 93.3 (2.6) |
93.0 (4.9) |
95.5 (1.9) |
93.9 (3.2) |
86.4 (9.2) |
92.1 (5.0) |
100 | 90.7 (5.2) |
| White | 93.8 a (1.6) |
92.1 (2.3) |
96.3 (1.2) |
93.4 b (1.7) |
93.5 (4.5) |
98.9 (1.0) |
95.3 (4.9) |
96.3 (2.1) |
|
| Hispanic | 91.7 (4.0) |
88.6 (6.6) |
97.5 (1.5) |
91.0 (4.4) |
97.6 (2.0) |
96.3 (2.7) |
--- | 96.6 (1.8) |
|
| Other | 98.7 (0.8) |
100 | 98.1 (1.4) |
99.2 (0.6) |
92.1 (7.9) |
100 | --- | 96.3 (3.6) |
|
| TOTAL | 93.6 (1.5) |
92.1 (2.6) |
96.4 (0.8) |
93.5 (1.8) |
91.3 (4.4) |
96.3 (2.0) |
95.9 (2.9) |
94.4 (2.1) |
|
|
a Children of other races/ethnicities and their families are significantly more likely than White children and their families to have prior investigations of abuse or neglect (X2 = 11.0, p ≤ .001). (back) |
Finally, all child welfare workers were asked if, apart from investigations, there was a prior history with CWS. This would include any ongoing in-home services or out-of-home care. Close to one-third (30%) of families being investigated for abuse and neglect also have a known, prior CWS history. Again, this proportion is significantly higher (p < .001) for children in out-of-home care and their families—almost twice that of children remaining at home and their families (53% vs. 27%). A large and significant difference (p < .001) is also present within the subpopulation of children remaining at home, with families receiving child welfare services again more likely to have prior CWS history than families not receiving services (41% vs. 22%). With regard to age, the youngest children are significantly less likely than the oldest children to have prior CWS history, both overall and for children remaining at home; but this would be expected due to the decreased opportunity for such history. There are no significant differences in the presence of prior CWS history based on race/ethnicity (Table 4-4).
| Characteristic | TOTAL | Setting | |||||||
|---|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | ||||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | |||
| Percent^ / (SE) | |||||||||
| Age | 0-2 | 54.9 (4.4) |
51.5 (8.7) |
54.0 (7.1) |
52.5 (5.4) |
59.3 (11.8) |
62.9 (6.7) |
--- | 60.9 (7.9) |
| 3-5 | 46.5 (6.0) |
44.1 (7.0) |
47.7 (9.3) |
45.3 (6.3) |
78.2 (9.1) |
65.3 (11.4) |
--- | 63.7 (9.0) |
|
| 6-10 | 54.8 (4.6) |
46.7 (6.0) |
65.3 (5.8) |
52.5 (4.9) |
87.5 (3.7) |
70.6 (6.3) |
--- | 72.3 (6.8) |
|
| 11+ | 61.6 (4.5) |
56.0 (6.1) |
60.8 (7.6) |
57.7 (4.8) |
86.8 (4.6) |
82.5 (8.2) |
89.6 (5.6) |
79.4 (7.7) |
|
| Race/ Ethnicity | African American | 58.3 (4.7) |
53.6 (6.8) |
58.9 (6.9) |
55.6 (5.1) |
83.8 (4.9) |
65.7 (7.9) |
80.4 (10.7) |
70.3 (6.4) |
| White | 52.7 (3.5) |
46.0 (5.3) |
56.9 (5.9) |
49.5 (3.8) |
82.3 (4.5) |
72.4 (5.9) |
60.8 (25.7) |
75.3 (5.0) |
|
| Hispanic | 58.5 (8.2) |
58.2 (10.0) |
60.0 (12.2) |
58.7 (9.5) |
43.2 (21.4) |
72.6 (13.5) |
--- | 56.8 (12.6) |
|
| Other | 46.6 (9.7) |
26.9 (10.4) |
60.9 (13.8) |
41.3 (11.2) |
59.7 (7.1) |
87.0 (11.7) |
--- | 72.4 (7.8) |
|
| TOTAL | 54.8 (2.6) |
48.9 (3.6) |
58.3 (3.9) |
52.1 a (2.9) |
74.9 (6.2) |
71.5 (3.6) |
66.9 (16.1) |
70.6 (3.9) |
|
|
|
A logistic regression controlling for age, gender, race/ethnicity, and child setting confirmed bivariate results that the oldest children are significantly more likely than the youngest children to have prior CWS history. Further, children remaining at home with services are significantly more likely than children remaining at home with no services to have prior CWS history. In addition, the model results indicated that children in each of the out-of-home placement types are significantly more likely (p < .001 for all three placement types) than children remaining at home with no services to have had prior CWS history (Table 4-5). A similar regression run (not shown) using “in-home services” as the reference group for the child setting variable further indicated that children in foster homes are significantly more likely than
children remaining at home with services and their families to have had prior CWS history (OR = 2.55, CI = 1.72, 3.79; p < .001).
| Characteristic | TOTAL | Setting | |||||||
|---|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | ||||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | |||
| Percent^/ (SE) | |||||||||
| Age | 0-2 | 26.9 c (2.1) |
15.2 (2.2) |
37.8 (3.8) |
21.5 d (2.0) |
69.9 (5.3) |
35.3 (5.1) |
--- | 53.9 (5.0) |
| 3-5 | 30.9 (4.1) |
25.2 (4.7) |
41.8 (7.9) |
30.0 (4.5) |
56.2 (8.8) |
43.6 (14.7) |
--- | 44.6 (9.7) |
|
| 6-10 | 27.6 (2.9) |
20.5 (3.5) |
39.4 (4.4) |
24.9 (2.9) |
55.8 (6.0) |
55.4 (8.7) |
21.5 (13.5) |
53.4 (5.8) |
|
| 11+ | 34.7 (3.4) |
26.6 (4.5) |
43.4 (6.7) |
31.6 (3.6) |
58.8 (9.5) |
52.9 (9.0) |
86.7 (5.1) |
54.6 (5.6) |
|
| Race/ Ethnicity | African American | 34.4 (3.8) |
27.5 (5.4) |
41.8 (5.8) |
31.8 (4.3) |
56.7 (3.5) |
49.5 (7.3) |
72.7 (12.7) |
51.0 (4.1) |
| White | 30.2 (2.4) |
22.2 (2.5) |
42.6 (4.2) |
27.5 (2.3) |
60.3 (5.7) |
46.4 (8.3) |
63.4 (17.5) |
52.6 (4.1) |
|
| Hispanic | 22.8 (4.7) |
15.8 (5.1) |
30.4 (4.9) |
19.3 (4.5) |
82.0 (6.6) |
35.7 (7.7) |
97.9 (1.8) |
57.5 (6.7) |
|
| Other | 28.6 (5.2) |
16.6 (5.0) |
47.5 (10.4) |
25.7 (5.5) |
52.3 (11.8) |
58.0 (18.7) |
--- | 57.0 (9.5) |
|
| TOTAL | 29.9 (2.0) |
22.0 b (2.5) |
40.7 (3.2) |
27.0 a (2.1) |
61.7 (3.3) |
46.6 (5.7) |
70.1 (11.5) |
53.0 (2.4) |
|
|
a Children in out-of-home care and their families are significantly more likely than children remaining at home and their families to have prior child welfare services history (X2 = 45.3, p < .001). (back) b Children remaining at home with services and their families are significantly more likely than children remaining at home with no services and their families to have prior child welfare services history (X2 = 23.8, p < .001). (back) c Children aged 11+ and their families are significantly more likely than children aged 0-2 and their families to have prior child welfare services history (X2 = 6.2, p ≤ .01). (back) |
Overall, results of these analyses indicate that fairly large proportions of children/families being investigated for maltreatment have had involvement with CWS prior to the current investigation. About half (51%) have had prior reports of maltreatment, one-quarter (25%) have had a prior incident of substantiated abuse or neglect, and close to one-third (30%) have had prior CWS history other than that related to an investigation. Children in out-of-home care and their families are more likely than children remaining at home and their families to have prior reports, substantiations, and other CWS history. Children at home receiving services and their families also have greater prior CWS involvement than children at home not receiving services and their families. The consistency of these findings appears to support the notion of a central
role of prior CWS history in current decision-making—children and families with higher levels of past CWS involvement have a higher level of current service receipt. The child's race/ ethnicity does not appear to be related to prior CWS involvement, and the child's age may only be a factor insofar as increasing age provides more opportunity for CWS involvement. Changes in CWS record keeping that might occur from the implementation of multiple response systems—in which services receipt is not recorded because it is provided following diversion to community agencies—could affect the ability of child welfare agencies to use previous CWS involvement for decision-making.
| OR | 95% CI | ||
|---|---|---|---|
| Age | 11+ | (reference group) | |
| 0-2 | .65* | .47, .89 | |
| 3-5 | .91 | .61, 1.36 | |
| 6-10 | .71 | .45, 1.10 | |
| Gender | Female | (reference group) | |
| Male | .90 | .68, 1.18 | |
| Race/ Ethnicity | White | (reference group) | |
| African American | 1.14 | .74, 1.76 | |
| Hispanic | .67 | .37, 1.21 | |
| Other | .89 | .55, 1.44 | |
| Child Setting/ Service | In-home, No Services | (reference group) | |
| In-home, services | 2.39**_i | 1.70, 3.35 | |
| Foster home | 6.10**_ii | 4.30, 8.66 | |
| Kinship care | 2.85**_iii | 1.71, 4.75 | |
| Group home care | 7.40**_iv | 2.84, 19.31 | |
|
Cox and Snell pseudo-R2 is .07 |
4.1.2 Caregiver Risk Factors at Time of Investigation
Child welfare workers were asked about the presence of various risk factors in caregiver's lives at the time of the investigation and whether there was a recent history of certain risk factors; the answers were scored as yes or no. We have categorized the 25 risk items into the following 8 types of risks: substance abuse, cognitive and physical health of the caregivers, parenting, family, violence against the caregiver, risk factors of the secondary caregiver, child characteristics, and cooperation by the caregiver. This section presents the prevalence of these risks, which is important in understanding various aspects of the child's living environment, particularly specific characteristics of the caregiver and family, at the time of the investigation. The reader is reminded that this analysis is based on child welfare workers' perceptions of risks, and may disagree with other data sources that address similar risks or behaviors.
Overall, about 8% of caregivers were identified by the child welfare worker completing the risk assessment as actively abusing alcohol, and 9% of caregivers were classified as actively abusing other drugs at the time of the investigation (Table 4-6). Approximately 12% had a recent history of arrest. Although these may not have been related to substance abuse, we have included them in this group of substance-abuse-related risks because these drug-related arrests predominate among arrests of females (Chilton & Jarvis, 2001) and are also important contributors to child welfare reports (Albert & Barth, 1996). This involvement with substance abuse varies considerably by setting—caregivers of children who remained at home following the investigation and who received in-home services are significantly more likely to have been identified by the child welfare worker as abusing alcohol or other drugs or as having a recent history of arrest than caregivers of children who remained at home and did not receive services (p < .001). Furthermore, caregivers of children who live out of the home are significantly more likely to abuse alcohol or other drugs or have a recent history of arrest than caregivers of children who live at home (p < .001). Caregivers of children in kinship foster care are significantly more likely to abuse drugs or have a recent history of arrest than caregivers of children in foster care or group care.
About 15% of caregivers were identified by child welfare workers as having had a serious mental health problem at the time of the investigation; almost 7% had an intellectual or cognitive impairment, and about 5% had a physical impairment (Table 4-7). The proportion of caregivers with serious mental health problems or an intellectual or cognitive impairment as identified by the child welfare worker is highest among caregivers of children who have been placed outside the home. That is, significantly more caregivers of children placed outside the home have a serious mental health problem (p < .001) or intellectual or cognitive impairment than caregivers of children living at home. In addition, within the subpopulation of children remaining at home, significantly more caregivers of children receiving services have a serious mental health problem (p < .001) or intellectual or cognitive impairment than caregivers of children not receiving services. Finally, for children in out-of-home care, caregivers of children in kinship foster care are significantly more likely than caregivers of children in group care to have a serious mental health problem (p < .001). The proportion of caregivers with physical impairments does not vary significantly by setting.
Many caregivers are seen by child welfare workers as having parenting risk factors, and caregivers of children who live out of the home are much more likely to have parenting risk factors than caregivers of children who live at home (Table 4-8). Overall, about one-third of caregivers had poor parenting skills, almost one-fifth had unrealistic expectations, and slightly less than one-tenth used excessive and/or inappropriate discipline. (Just over one-third of those who used inappropriate discipline were assessed by their child welfare workers as having no motivation to change their use of excessive discipline.) These proportions are much higher among caregivers of children who are placed outside the home, for whom 79% had poor parenting skills, 44% had unrealistic expectations, 19% used excessive and/or inappropriate discipline, and 72% had no motivation to change with regard to their inappropriate discipline, according to the child welfare worker. In addition, caregivers of children receiving in-home services are significantly more likely than caregivers of children not receiving child welfare services to have risk factors of poor parenting skills, unrealistic expectations, and excessive or inappropriate discipline (p < .001).
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^ / (SE) | ||||||||
| Active alcohol abuse by PCG^^ | 8.2 (0.7) |
3.3 (0.5) |
12.6 a (1.6) |
5.8 (0.6) |
25.4 (3.3) |
36.7 (4.8) |
22.4 (13.0) |
28.7 b (2.5) |
| Active drug abuse by PCG | 9.2 (0.9) |
3.5 (0.8) |
12.1 c (1.3) |
5.8 (0.7) |
36.8 (3.7) |
48.8 d e (4.7) |
8.1 (3.7) |
37.4 f (3.0) |
| PCG recent history of arrest | 12.4 (1.0) |
7.8 (1.1) |
15.6 g (1.5) |
9.9 (1.0) |
30.4 (2.4) |
44.9 h i (5.5) |
12.5 (4.6) |
33.9 j (3.1) |
|
a Caregivers of children living at home and receiving services are significantly more likely to be active alcohol abusers than caregivers of children living at home and not receiving services (X2 = 27.1, p < .001). (back) b Caregivers of children living out of the home are significantly more likely to be active alcohol abusers than caregivers of children at home (X2 = 29.5, p < .001). (back) c Caregivers of children living at home and receiving services are significantly more likely to be active drug abusers than caregivers of children living at home and not receiving services (X2 = 24.6, p < .001). (back) d Caregivers of children living in kinship foster care are significantly more likely to be active drug abusers than caregivers of children living in foster care (X2 = 7.7, p < .01). (back) e Caregivers of children living in kinship foster care are significantly more likely to be active drug abusers than caregivers of children living in group care (X2 = 8.5, p < .01). (back) f Caregivers of children living out of the home are significantly more likely to be active drug abusers than caregivers of children living at home (X2 = 28.0, p < .001). (back) g Caregivers of children living at home and receiving services were significantly more likely to have a recent history of arrest than caregivers of children living at home and not receiving services (X2 = 16.3, p < .001). (back) h Caregivers of children living in kinship foster care are significantly more likely to have a history of recent arrest than caregivers of children living in foster care (X2 = 6.8, p ≤ .01). (back) i Caregivers of children living in kinship foster care are significantly more likely to have a history of recent arrest than caregivers of children living in group care (X2 = 9.1, p < .01). (back) |
Family risk factors were present in many caregivers' lives at the time of investigation (Table 4-9). Over half of the caregivers had no other supportive caregiver, just over half had high stress in the family, close to one-third had low social support, and almost one-quarter had trouble paying for basic necessities. As with the risk factors discussed above, caregivers of children living out of the home are significantly more likely to have experienced each of these family risk factors as identified by the child welfare worker than caregivers of children living at home (p < .001 for high stress in family, low social support, and trouble paying for basic necessities). In addition, caregivers of children receiving services at home are significantly more likely than caregivers of children with closed cases to have experienced high stress in family, low social support, and trouble paying for basic necessities (p < .001).
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| Serious mental health problem | 15.3 (1.4) |
9.4 (1.4) |
22.4 a (2.5) |
12.9 (1.5) |
43.0 (3.9) |
35.8 b (3.7) |
24.5 (6.9) |
35.8 c (2.4) |
| Intellectual or cognitive impairment | 6.9 (0.9) |
5.2 (1.0) |
9.2 d (1.3) |
6.2 (0.9) |
14.8 (3.0) |
11.4 (3.1) |
10.4 (4.1) |
11.7 e (2.0) |
| Any physical impairments | 5.4 (0.6) |
4.5 (0.9) |
6.1 (0.9) |
5.0 (0.7) |
7.7 (2.2) |
10.9 (3.3) |
5.1 (2.3) |
8.2 (1.3) |
|
a Caregivers of children living at home and receiving services are significantly more likely to have a serious mental health problem than caregivers of children living at home and not receiving services (X2 = 29.4, p < .001). (back) b Caregivers of children living in kinship foster care are significantly more likely to have a serious mental health problem than caregivers of children living in group care (X2 = 30.0, p < .001). (back) c Caregivers of children living out of the home are significantly more likely to have a serious mental health problem than caregivers of children living at home (X2 = 30.0, p < .001). (back) d Caregivers of children living at home and receiving services are significantly more likely to have an intellectual or cognitive impairment than caregivers of children living at home and not receiving services (X2 = 6.9, p ≤ .01). (back) |
A high proportion of caregivers have been the subject of violence (Table 4-10). Almost one-third of caregivers had a history of domestic violence, about one-fifth of caregivers had a history of abuse or neglect, and just over one-tenth of caregivers were experiencing domestic violence at the time of the investigation. As with the other risk factors discussed so far, the prevalence of violence against the caregiver as assessed by the child welfare worker is significantly higher among caregivers of children living out of the home than among caregivers of children living at home (p < .001 for history of abuse or neglect against the caregiver and p < .01 for active domestic violence against the caregiver). In addition, the likelihood of history of domestic violence and history of abuse or neglect against the caregiver is significantly greater for caregivers of children receiving in-home services than caregivers of children with closed cases (p < .001 for history of abuse or neglect against the caregiver).
The child welfare worker was also asked about the presence of risk factors in the secondary caregiver's life. At the time of the investigation, 12% of secondary caregivers were said to be actively abusing alcohol, about 9% actively abusing other drugs, and almost 15% using inappropriate or excessive discipline, and about 13% had a history of abuse or neglect (Table 4-11). Caregivers of children who live at home and receive services were significantly more likely to be identified by the child welfare worker as abusing alcohol or other drugs or as having a history of abuse or neglect than caregivers of children who live at home and do not receive services (p < .001 for drug abuse). In addition, caregivers of children who live out of the home were more likely to have the same three risk factors present than caregivers of children who live at home (p < .001 for alcohol abuse and other drug abuse).
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| Poor parenting | 33.2 (1.9) |
20.4 (2.3) |
47.4 a (2.6) |
27.6 (2.0) |
85.3 (2.2) |
79.7 (4.0) |
79.9 (6.7) |
79.1 b (2.3) |
| Unrealistic expectations | 17.4 (1.2) |
9.4 (1.3) |
28.1 c (2.0) |
14.4 (1.2) |
51.7 (3.8) |
35.0 (6.3) |
64.7 (10.9) |
43.6 d (2.6) |
| Excessive/ inappropriate discipline | 8.3 (0.8) |
4.9 (1.2) |
12.9 e (1.6) |
7.0 (0.9) |
23.6 (2.2) |
16.2 (2.5) |
18.7 (6.1) |
18.8 f (1.5) |
| No motivation to change (re: inappropriate discipline) | 35.1 (3.2) |
21.5 (5.7) |
29.3 (4.5) |
25.0 (4.0) |
77.4 (5.2) |
71.8 (8.5) |
81.5 (9.1) |
71.6 g (3.6) |
|
a Caregivers of children living at home and receiving services are significantly more likely to have poor parenting skills than caregivers of children living at home who are not receiving services (X2 = 55.3, p < .001). (back) b Caregivers of children living out of the home are significantly more likely to have poor parenting skills than caregivers of children living at home (X2 = 35.4, p < .001). (back) c Caregivers of children living at home and receiving services are significantly more likely to have unrealistic expectations of their child than caregivers of children living at home and not receiving services (X2 = 46.9, p < .001). (back) d Caregivers of children living out of the home are significantly more likely to have unrealistic expectations of their child than caregivers of children living at home (X2 = 44.7, p < .001). (back) e Caregivers of children living at home and receiving services are significantly more likely to use excessive and/or inappropriate discipline than caregivers of children living at home who are not receiving services (X2 = 13.5, p < .001). (back) f Caregivers of children living out of the home are significantly more likely to use excessive and/or inappropriate discipline than caregivers of children living at home (X2 = 33.3, p < .001). (back) |
Children may have characteristics that put them at increased risk of abuse or neglect. The child welfare workers were asked about the presence of two of those risks: poor ability to self-protect (for children aged 5 and older) and special needs or behavior problems. Overall, 40% of children aged 5 and older had a poor ability to self-protect and 21% were seen as having special needs or behavior problems (Table 4-12). Children who live at home and receive services are significantly more likely to have had these two risks present as assessed by the child welfare worker than children who live at home and do not receive services (p < .001 for special needs or behavior problems). In addition, children who live out of the home are significantly more likely to have had these risks present than children who live at home (p < .001). Finally, children in group care are significantly more likely (p < .001) than both children in foster care and children in kinship foster care to have had special needs or behavior problems at the time of the investigation.
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| No other supportive caregiver | 53.9 (2.2) |
51.9 (2.8) |
53.5 (2.4) |
52.4 (2.4) |
68.1 (4.7) |
66.9 (4.5) |
72.4 (9.5) |
66.6 a (2.8) |
| High stress in family | 51.7 (1.9) |
42.9 (2.4) |
66.1 b (2.3) |
49.2 (2.1) |
74.3 (4.9) |
76.8 (3.7) |
54.4 (12.6) |
73.2 c (2.9) |
| Low social support | 30.5 (1.7) |
23.5 (1.8) |
39.3 d (2.5) |
27.7 (1.7) |
62.0 (4.1) |
46.5 (4.7) |
65.4 (10.6) |
52.7 e (2.4) |
| Have trouble paying basic necessities | 23.9 (1.7) |
16.3 (1.7) |
33.9 f (2.6) |
21.0 (1.7) |
49.7 (4.8) |
53.6 (4.8) |
24.0 (7.3) |
47.1 g (3.6) |
|
a Caregivers with children living out of the home are significantly more likely to lack another supportive caregiver than caregivers with children living at home (X2 = 10.9, p < .01). (back) b Families with children living at home and receiving services are significantly more likely to have high stress than children living at home and not receiving services (X2 = 47.0, p < .001). (back) c Families with children living at home are significantly more likely to have high stress than families with children living out of the home (X2 = 19.4, p < .001). (back) d Caregivers with children living at home and receiving services are significantly more likely to have low social support than caregivers with children living at home and not receiving services (X2 = 35.5, p < .001). (back) e Caregivers with children living out of the home are significantly more likely to have low social support than caregivers with children living at home (X2 = 29.0, p < .001). (back) f Caregivers with children living at home and receiving services are significantly more likely to have trouble paying for basic necessities than caregivers with children living at home and not receiving services (X2 = 37.0, p < .001). (back) |
Cooperation with child welfare workers has often been shown to predict case outcomes (e.g., Karski, 1996). The vast majority of caregivers (90%) exhibited a “reasonable” level of cooperation, though cooperation was significantly less likely to be reported (p < .001) among caregivers whose children live out of the home than among caregivers whose children live at home (Table 4-13). Caregiver involvement in non-CPS services may also predict child outcomes if use of supplementary human services helps reduce risks to children. Or, involvement in other human services may indicate a more pervasive set of challenges for families. Almost 30% of the caregivers had involvement in specialized non-child welfare services. This is meant to include services provided by outside (i.e., non-Department of Social Services [DSS]) agencies and may include services such as mental health, home visiting, public health nursing, or substance abuse treatment. It is not meant to include services provided by DSS such as income maintenance, Medicaid, and day care. Caregivers of children who live at home and receive child welfare services were significantly more likely to be involved with a non-CPS service as identified by the child welfare worker than caregivers of children who live at home and do not receive services (p < .001). There is no difference in caregiver involvement in other human services between caregivers of children who live at home and children who live out of the home.
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| History of domestic violence against PCG^^ | 30.0 (1.7) |
25.2 (2.3) |
36.4 a (2.3) |
28.2 (1.9) |
49.3 (4.1) |
44.8 (3.9) |
38.8 (12.3) |
44.3 b (3.6) |
| History of abuse/neglect against PCG | 21.6 (1.6) |
14.9 (2.0) |
31.9 c (1.9) |
19.5 (1.7) |
49.4 (4.0) |
33.5 (4.8) |
48.0 (13.2) |
39.4 d (3.2) |
| Active domestic violence | 13.4 (0.9) |
10.8 (1.3) |
15.3 (1.6) |
12.0 (1.2) |
26.3 (2.6) |
30.2 (3.7) |
12.7 (5.0) |
25.1 e (2.3) |
|
a Caregivers of children living at home and receiving services are significantly more likely to have a history of domestic violence than caregivers of children living at home and not receiving services (X2 = 10.0, p < .01). (back) b Caregivers of children living out of the home are significantly more likely to have a history of domestic violence than caregivers of children living at home (X2 = 7.2, p < .01). (back) c Caregivers of children living at home and receiving services are significantly more likely to have experienced abuse/neglect as a child than caregivers of children living at home and not receiving services (X2 = 27.3, p < .001). (back) d Caregivers of children living out of the home are significantly more likely to have experienced abuse/neglect as a child than caregivers of children living at home (X2 = 16.3, p < .001). (back) |
To determine how caregiver risks are associated with the child's eventual service setting, we conducted a multinomial logistic regression of child age, type of maltreatment, child race/ethnicity, child gender, a total caregiver risk score, and risk type scores on setting—in-home, open, and out-of-home versus in-home, no services (Table 4-14). The total caregiver risk score categories were developed by summing the risk factors discussed above, with the exception of “caregiver involvement in non-CPS service” because this could potentially be interpreted as a risk or protective factor. An additional variable, which considered the absence of a secondary caregiver a risk factor, was also created and included in the total. After adjusting for missing values, the total score was categorized into three risk groups (low, medium, and high). Risk type scores were calculated by taking the mean of the variables that made up each risk subcategory, as shown in Tables 4-6 through 4-13. The items were all yes/no questions and were recoded such that a “no” response was given a value of 0 and a “yes” response was given a value of 1. Resulting means, therefore, fell somewhere between 0 and 1. (Note that the child risk subcategory was excluded because, due to the large number of predictor variables utilized, its inclusion exceeded the capacity of the model.)
While the analysis did not reveal any association of setting with the total caregiver risk score, significant associations with risk type scores were observed. Specifically, children of caregivers with high parenting risk scores (comprising poor parenting, unrealistic expectations, excessive/inappropriate discipline, and no motivation to change with regard to inappropriate discipline) are over 5 times as likely to receive services at home and over 11 times as likely to be placed in out-of-home care than to remain at home with no services (p < .001 for both comparisons). Children of caregivers with high substance abuse risk scores (comprising active alcohol abuse by the primary caregiver, active drug abuse by the primary caregiver, and recent history of arrest of the primary caregiver) are almost 7 times more likely to be placed in out-of-home care than to remain at home with no services (p < .001). Finally, children of caregivers with high cooperation scores (comprising reasonable level of caregiver cooperation and caregiver involvement in non-CPS services) are more likely to remain at home with no services than to remain at home with services (p ≤ .001).
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| Active alcohol abuse | 12.0 (1.1) |
9.2 (1.3) |
15.1 a (1.6) |
10.8 (1.2) |
27.5 (3.6) |
25.2 (5.2) |
--- | 22.7 b (2.7) |
| Active drug abuse | 8.9 (0.9) |
5.1 (1.1) |
13.4 c (1.8) |
7.3 (0.9) |
21.9 (3.0) |
30.3 (6.3) |
12.3 (6.4) |
23.7 d (3.3) |
| Inappropriate/ excessive discipline | 14.5 (1.2) |
12.6 (1.5) |
16.2 (1.7) |
13.5 (1.3) |
32.1 (6.1) |
15.0 (2.9) |
18.5 (12.6) |
23.9 (3.3) |
| History of abuse/ neglect | 12.9 (1.1) |
10.2 (1.6) |
16.6 e (2.0) |
11.9 (1.2) |
22.3 (3.6) |
26.1 (4.7) |
20.8 (11.7) |
24.3 f (3.4) |
|
a Secondary caregivers of children living at home and receiving services are significantly more likely to have active alcohol abuse than secondary caregivers of children living at home and not receiving services (X2 = 11.2, p < .01). (back) b Secondary caregivers of children living out of the home are significantly more likely to have active alcohol abuse than secondary caregivers of children living at home (X2 = 22.1, p < .001). (back) c Secondary caregivers of children living at home and receiving services are significantly more likely to have active drug abuse than secondary caregivers of children living at home and not receiving services (X2 = 12.3, p < .001). (back) d Secondary caregivers of children living out of the home are significantly more likely to have active drug abuse than secondary caregivers of children living at home (X2 = 26.7, p < .001). (back) e Secondary caregivers of children living at home and receiving services are significantly more likely to have a history of abuse or neglect than secondary caregivers of children living at home and not receiving services (X2 = 6.2, p ≤ .01). (back) |
To summarize, across all of the 25 risk factors examined, the proportion of children having these present at the time of the investigation ranges from 5% (caregiver has physical impairment) to 71% (caregiver not involved in non-CPS services). When grouped into categories, the “family” risk factors appear to be the most common, with child welfare workers indicating that each of these risks was present in the families of about one-quarter to one-half of the children at the time of the investigation. These factors include no other supportive caregiver in the household, high stress in the family, low social support, and family having trouble paying basic necessities. The comparison of risk factors across child settings revealed a consistent pattern, with children in out-of-home care more likely than children remaining at home, and children remaining at home with services more likely than children remaining at home without services to have most of the risk factors. In addition, with regard to several of the risk factors—drug abuse of primary caregiver, recent arrest history of primary caregiver, and serious mental health problem of caregiver—children placed in kinship care following the current investigation were more likely to be at risk at the time of the investigation. A multinomial logistic regression identified the parenting, substance abuse, and cooperation risk categories as significant in predicting child setting. Those children whose families had more parenting or substance abuse risk factors were less likely to be at home with no services, and those children whose caregivers who scored more favorably on the cooperation risk factor were more likely to be at home with no services.
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| Poor ability to self-protect | 40.1 (1.8) |
33.9 (2.2) |
47.4 a (4.0) |
37.4 (1.8) |
66.7 (4.9) |
63.4 (5.6) |
64.7 (11.8) |
59.9 b (3.8) |
| Special needs/ behavior problems | 21.1 (1.6) |
16.3 (2.0) |
27.1 c (2.4) |
19.2 (1.7) |
37.1 (4.1) |
27.1 (4.9) |
92.0 d e (3.4) |
35.9 f (3.1) |
|
a Children living at home and receiving services are significantly more likely to have a poor ability to self-protect than children living at home and not receiving services (X2 = 9.6, p < .01). (back) b Children living out of the home are significantly more likely to have a poor ability to self-protect than children living at home (X2 = 15.1, p < .001). (back) c Children living at home and receiving services are significantly more likely to have a major special need or behavior problem than children living at home and not receiving services (X2 = 15.6, p < .001). (back) d Children living in group care are significantly more likely to have a major special need or behavior problem than children living in foster care (X2 = 11.5, p = .001). (back) e Children living in group care are significantly more likely to have a major special need or behavior problem than children living in kinship foster care (X2 = 12.4, p < .001). (back) |
4.1.3 Critical Factors in Case Decision-Making
After answering the individual risk assessment items, child welfare workers were asked to identify the two factors that were the most critical in their determination of how to proceed with the case. These data are valuable in understanding the paths various children take through CWS and how they end up on these paths. Understanding the priority concerns of child welfare workers may contribute to developing services that best support less intrusive care for children. The factor mentioned most often by child welfare workers was “reasonable level of caregiver cooperation”—identified by about one-third (34%) of the child welfare workers as one of the two most critical factors influencing their decision on how to proceed with the case. This finding is consistent with much of the other research on the significance of this factor in determining case dispositions. This suggests that the caregivers themselves may play a large role in affecting the outcome of their cases with the child welfare agency, although this also raises the concern that clients who have legitimate concerns about the way their cases are being handled may be disadvantaged if they seem uncooperative. About one-quarter (24%) of the child welfare workers identified the child's inability to self-protect against future maltreatment as a critical factor. Each of the following factors was identified by approximately one in eight child welfare workers as being critical in determining how to proceed with the case: another supportive caregiver present in the home, prior investigation of abuse or neglect, high stress on the family, and the child's special needs or behavior problems (Table 4-15). Drug abuse, active alcohol abuse, domestic violence, and serious mental health problems were all ranked much lower—perhaps because they are less common issues than impaired parenting and because they are contributing factors to impaired and abusive parenting rather than singular reasons for child welfare intervention.
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | ||
| Percent^/ (SE) | ||||||||
| Reasonable level of caregiver cooperation | 90.3 (0.8) |
93.0 (1.0) |
91.4 (1.4) |
92.6 (0.7) |
68.5 (3.9) |
68.2 (5.4) |
84.6 (6.0) |
71.5 a (3.2) |
| Caregiver involvement in non-CPS service | 29.3 (2.0) |
25.2 (2.5) |
39.8 b (3.0) |
29.2 (2.2) |
5.1 (4.0) |
23.8 (4.2) |
51.9 (12.1) |
31.2 (2.8) |
|
a Caregivers of children living at home are significantly more likely to demonstrate a reasonable level of cooperation than caregivers of children living out of the home (X2 = 42.6, p < .001). (back) |
The six factors that were mentioned most frequently overall by child welfare workers as influencing their decisions on the child's case were examined to determine if there were significant differences regarding how frequently they were mentioned based on child setting, child age, or child race/ethnicity. Table 4-16 presents the proportion of child welfare workers that identified each of these six factors as one of the most critical, by child setting. With regard to both “reasonable level of caregiver cooperation” and “another supportive caregiver present in the home,” child welfare workers of children remaining at home are significantly more likely than those of children in out-of-home care to mention these as critical factors affecting their decision-making (p ≤ .001). Additionally, within the subpopulation of children remaining at home, child welfare workers of children with no services are significantly more likely than those of children with services to mention these factors as critical (p < .001). Not only does this finding indicate, as mentioned above, that caregivers may play an important role in affecting the outcome of their case, but it further suggests that the caregiver's level of cooperation and whether or not the caregiver has the support of another caregiver in the household weigh heavily in a child welfare worker's decision. This accounts for some of the difficulty that child welfare researchers have had in explaining placement decisions with more limited data sets that only contain demographic case characteristics such as race, age, and maltreatment type. Further, this suggests the importance of child welfare worker training that helps to optimize engagement of caregivers in services use and that draws on the natural helping resources of families.
| Characteristic | In-Home, Services/ In-Home, No Services | Out-of-Home/ In-Home, No Services | |
|---|---|---|---|
| OR | (95% CI) | ||
| Child Age | 11+ | (reference group) | |
| 0 - 2 | .65 (.34, 1.26) | 1.13 (.51, 2.50) | |
| 3 - 5 | 1.06 (.51, 2.22) | .49 (.22, 1.07) | |
| 6 - 10 | .77 (.40, 1.49) | .68 (.36, 1.28) | |
| Type of Maltreatment | Physical | (reference group) | |
| Sexual | 1.46 (.73, 2.90) | 1.91 (.76, 4.81) | |
| Failure to provide | 1.15 (.68, 1.96) | 1.20 (.68, 2.12) | |
| Failure to supervise | 1.08 (.62, 1.87) | 1.16 (.65, 2.06) | |
| Other | .74 (.40, 1.37) | .64 (.23, 1.76) | |
| Child Race/ Ethnicity | White | (reference group) | |
| African American | 1.05 (.69, 1.58) | 1.00 (.66, 1.53) | |
| Hispanic | 1.30 (.73, 2.33) | 1.03 (.52, 2.05) | |
| Other | 1.17 (.56, 2.43) | .78 (.46, 1.30) | |
| Child Gender | Female | (reference group) | |
| Male | 1.23 (.85, 1.78) | 1.39 (.98, 1.97) | |
| Total Caregiver Risk Score | Low risk | (reference group) | |
| Medium risk | 1.16 (.74, 1.81) | 1.72 (.66, 4.46) | |
| High risk | 1.21 (.54, 2.71) | 1.83 (.66, 5.06) | |
| Risk Type Scores | Substance Abuse | 1.18 (.59, 2.38) | 6.77*_i (4.10, 11.17) |
| Cognitive and physical health | 1.18 (.52, 2.66) | 1.16 (.56, 2.40) | |
| Parenting | 5.09*_ii (2.69, 9.62) | 11.13*_iii (5.05, 24.54) | |
| Family | 1.55 (.76, 3.17) | 2.02 (.84, 4.83) | |
| Violence against caregiver | 1.51 (.77, 2.96) | 1.13 (.56, 2.30) | |
| Secondary caregiver | 1.23 (.54, 2.80) | 1.67 (.82, 3.40) | |
| Cooperation | .38*_iv (.21, .67) | 1.47 (.87, 2.49) | |
| Factor | Percent(SE) |
|---|---|
| Reasonable level of caregiver cooperation | 33.9 (1.7) |
| Child's inability to self-protect | 23.6 (1.4) |
| Another supportive caregiver present in the home | 12.7 (1.1) |
| Prior investigation of abuse or neglect | 12.6 (1.1) |
| High stress on the family | 12.2 (0.8) |
| Child's major special needs or behavior problems | 12.1 (1.3) |
| Poor parenting skills-PCG | 9.3 (0.8) |
| Prior reports of maltreatment to the agency | 9.2 (0.9) |
| PCG recognized problem of inappropriate discipline and showed motivation to change | 9.0 (1.0) |
| Caregiver involvement in non-CPS services | 7.1 (1.0) |
| Prior incident of substantiated abuse or neglect | 7.0 (0.8) |
| Drug abuse-PCG | 6.5 (0.9) |
| Active domestic violence | 4.9 (0.6) |
| Prior child welfare service history (not including investigations) | 4.7 (0.6) |
| Serious mental health or emotional problems-PCG | 4.0 (0.5) |
| Active alcohol abuse-PCG ^^ | 3.4 (0.5) |
| Low social support | 3.4 (0.5) |
| Family had trouble paying for basic necessities | 3.2 (0.5) |
| Excessive and/or inappropriate discipline-SCG | 3.0 (0.4) |
| Excessive and/or inappropriate discipline-PCG | 2.8 (0.5) |
| Unrealistic expectations of the child-PCG | 2.7 (0.4) |
| History of abuse and neglect of PCG | 2.0 (0.3) |
| Drug abuse-SCG | 1.9 (0.4) |
| Active alcohol abuse-SCG ^^^ | 1.8 (0.3) |
| History of domestic violence against caregiver | 1.7 (0.4) |
| History of abuse and neglect of SCG | 1.3 (0.4) |
| Intellectual or cognitive impairments-PCG | 1.2 (0.4) |
| Recent history of arrests or detention in jail or prison-PCG | 0.8 (0.3) |
| Physical impairments-PCG | 0.2 (0.1) |
|
^ 5,047 child welfare workers identified at least one critical factor; 4,962 child welfare workers identified two critical factors. (back) |
| Factor | Setting | ||||||
|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | ||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home | |
| Percent^/(SE) | |||||||
| Reasonable level of caregiver cooperation | 41.8 a (2.3) |
22.2 (2.2) |
36.5 b (1.9) |
7.0 (2.1) |
11.3 (2.9) |
14.9 (9.9) |
13.6 (2.3) |
| Child's inability to self-protect | 22.5 (2.0) |
25.7 (1.9) |
23.3 (1.6) |
27.8 (4.5) |
27.0 (5.7) |
18.6 (11.9) |
25.5 (3.6) |
| Another supportive caregiver present in the home | 16.2 c (1.6) |
6.9 (1.4) |
13.7 d (1.3) |
3.3 (1.6) |
4.4 (2.0) |
0 | 4.4 (1.5) |
| Prior investigation of abuse or neglect | 13.2 (1.4) |
12.3 (2.0) |
13.0 (1.2) |
6.5 (1.3) |
13.1 (2.9) |
6.0 (4.2) |
9.3 (1.6) |
| High stress on the family | 12.0 (1.2) |
13.6 (1.5) |
12.4 (1.0) |
12.4 (2.2) |
11.6 (3.7) |
3.2 (2.1) |
10.3 (2.0) |
| Child's major special needs or behavior problems | 11.5 (1.7) |
13.4 (1.8) |
12.0 (1.3) |
10.5 (1.9) |
11.0 (4.0) |
37.9 (11.4) |
12.7 (2.8) |
|
a Children remaining at home with no services are significantly more likely than children remaining at home with services to have “reasonable level of caregiver cooperation” identified as a critical factor in determining how to proceed with their case (X2 = 34.0, p < .001). (back) b Children remaining at home are significantly more likely than children in out-of-home care to have “reasonable level of caregiver cooperation” identified as a critical factor in determining how to proceed with their case (X2 = 30.2, p < .001). (back) c Children remaining at home with no services are significantly more likely than children remaining at home with services to have “another supportive caregiver present in the home” identified as a critical factor in determining how to proceed with their case (X2 = 17.2, p < .001). (back) |
The entire list of risk factors was again examined to determine the three that were most often identified as most critical in determining how the child welfare worker would proceed with the case within each of the categories of child setting, child age, and child race/ethnicity (i.e., as opposed to the overall top six as presented in Table 4-16). This analysis was undertaken in an effort to further understand how child demographics and setting affect case decision-making for child welfare workers. As shown in Table 4-17, “child's inability to self-protect” was a critical factor mentioned among the three most frequent by child welfare workers of children in each of the settings. “Reasonable level of caregiver cooperation” was a top critical factor for child welfare workers of both categories of children remaining at home but not for child welfare workers of children in any of the out-of-home placement types. “Child's major special needs or behavior problems” and “prior incident of substantiated abuse or neglect” were the top two critical factors for child welfare workers of children in group care, but neither was one of the three most frequently mentioned factors by child welfare workers of children in any of the other setting subgroups. Drug abuse by the primary caregiver was one of the top three most frequently mentioned critical factors for child welfare workers of children in both foster care and kinship foster care (it was the most frequently mentioned for children in kinship foster care). This was not so for either of the in-home subgroups, indicating that children who are in out-of-home care are more likely to be perceived by child welfare workers as having this risk factor. This finding is consistent with much child welfare research indicating the centrality of substance abuse in child welfare services dynamics (e.g., Semidei, Feig-Radel, & Nolan, 2001) and replicates Karski's (1996) analysis, which found substance abuse to be one of the top four critical factors.
| Factor | In-Home | Out-of-Home | |||
|---|---|---|---|---|---|
| No Services | Services | Foster Care | Kinship Foster Care | Group Care | |
| Percent/ (SE) | |||||
| Reasonable level of caregiver cooperation^_i | 41.8 (2.3) |
22.2 (2.2) |
|||
| Child's inability to self-protect^_ii | 22.5 (2.0) |
25.7 (1.9) |
27.8 (4.5) |
27.0 (5.7) |
18.6 (11.9) |
| Another supportive caregiver present in the home^_iii | 16.2 (1.6) |
||||
| Prior investigation of abuse or neglect^_iv | 13.1 (2.9) |
||||
| Child's major special needs or behavior problems^_v | 37.9 (11.4) |
||||
| Poor parenting skills-PCG^^ | 14.4 (1.6) |
15.6 (4.8) |
|||
| Prior incident of substantiated abuse or neglect | 21.5 (10.9) |
||||
| Drug abuse-PCG | 19.0 (2.6) |
28.8 (4.7) |
|||
|
|
Examined from the perspective of age, “reasonable level of caregiver cooperation” was one of the three factors most frequently cited as critical by child welfare workers of children in all age groups, while “child's inability to self-protect” was one of the top three for all but the oldest age group (Table 4-18).
| Factor | Child's Age in Years | |||
|---|---|---|---|---|
| 0-2 | 3-5 | 6-10 | 11+ | |
| Reasonable level of caregiver cooperation^_i | 27.6 (2.8) |
37.5 (3.7) |
35.2 (2.7) |
33.9 (2.8) |
| Child's inability to self-protect^_ii | 37.4 (2.8) |
30.7 (3.4) |
19.4 (1.8) |
|
| Another supportive caregiver present in the home^_iii | 14.5 (2.0) |
|||
| Prior investigation of abuse or neglect^_iv | 18.9 (3.6) |
|||
| High stress on the family^_v | 14.1 (1.7) |
|||
| Child's major special needs or behavior problems^_vi | 18.0 (2.5) |
|||
| Prior reports of maltreatment to the agency | 14.1 (2.4) |
|||
There was less divergence in the top three critical factors when examined by race/ethnicity subgroups. As shown in Table 4-19, “reasonable level of caregiver cooperation” and “child's inability to self-protect” were the two most frequently mentioned critical factors (and in that order) by child welfare workers of children in all racial/ethnic groups, which is consistent with the overall findings.
| Factor | Child's Race/Ethnicity | |||
|---|---|---|---|---|
| African American | White | Hispanic | Other | |
| Reasonable level of caregiver cooperation^_i | 30.6 (3.7) |
35.3 (2.2) |
34.0 (3.1) |
37.8 (5.9) |
| Child's inability to self-protect^_ii | 26.6 (3.3) |
24.3 (1.7) |
18.7 (2.9) |
19.2 (3.5) |
| Another supportive caregiver present in the home^_iii | 17.7 (2.9) |
17.5 (5.6) |
||
| Prior investigation of abuse or neglect^_iv | 14.8 (2.0) |
|||
| Child's major special needs or behavior problems^_v | 13.3 (1.9) |
|||
In general, child welfare workers heavily weigh the caregiver's level of cooperation and the child's ability to self-protect in determining how they will proceed with a case. The caregiver's level of cooperation (as well as whether or not there is a second supportive caregiver in the home) appears to be particularly important in cases in which the child is ultimately allowed to remain at home. A positive caregiver attitude toward the involvement of the child welfare agency in his or her life apparently allows the agency to provide safety services with more confidence, whereas the presence of an uncooperative caregiver may require the child's removal from the home in order for the agency to meet its child protection mandate. The child's ability to self-protect, on the other hand, is particularly important when the child is younger, whereas the child's special needs and behaviors are more influential with regard to case decision-making for older children. The needs and abilities of the child often take precedence regardless of the other circumstances present in the home. The diversity of reasons for service decisions vary considerably by child's age—reflecting a developmental perspective in the way decisions are made—even if there is little corresponding developmental differentiation in child welfare policy. Substance abuse by the primary caregiver seems most significantly related to decision-making when children are then placed into kinship care. Critical factors in decision-making do not appear to vary much based on the race/ethnicity of the child.
4.1.4 Likelihood of Future Reports of Abuse or Neglect
Finally, child welfare workers were asked to give their opinion on the likelihood that the child would be reported for abuse or neglect in the next 24 months based on three levels of service provision: no services, in-home services, and placement into out-of-home care. The four possible responses were: 1 = very low likelihood, 2 = low likelihood, 3 = high likelihood, and 4 = very high likelihood. These data are useful in examining how child welfare workers perceive the impact of various levels of services and whether or not the level of service deemed necessary is related to the decision to leave the child at home or place the child in out-of-home care. Ultimately, the predictions will be tested against the follow-up findings. Mean responses overall and by child setting are presented in Table 4-20.
Analysis results indicate that if no services were provided, child welfare workers believe that children in out-of-home care would have a greater likelihood of re-report than children remaining at home (p < .001), and children remaining at home with services would have a greater likelihood of re-report than children remaining at home with no services (p < .001). The same patterns are present under the scenario of service provision. Further, child welfare workers believe that if no services were provided, children in group care would have a significantly greater likelihood of re-report than children in kinship foster care (p < .001) and tend to believe that children in foster care would also have a greater likelihood of re-report than children in kinship foster care (p = .04). All of these results indicate that child welfare workers' decisions regarding whether to place a child in out-of-home care and whether to provide services to a child remaining at home, as well as in what type of out-of-home care to place a child, are generally consistent with the level of risk for re-report that they believe the child faces.
4.1.5 Discussion of Risk Assessment
The risk assessment data show a relatively consistent and logical connection between the child welfare worker's perception of a child's level of risk at home at the time of the investigation and whether or not the child and his or her family receive services from the child welfare agency and/or the child is placed in out-of-home care. Although some of this logic could be influenced by the post-hoc completion of the instruments, we doubt that this is the whole reason for this consistency. The risk factors that appear to be most predominant in the households of families being investigated for maltreatment—that is, high family stress and no other supportive caregiver—are not the ones identified by the child welfare worker as most critical in the case decision-making process. Rather, the child welfare worker places more importance on the cooperation of the caregiver (which could certainly be affected by other family risk factors) and the ability of the child to self-protect. This indicates that the relationship of the caregiver and child is being examined with regard to the family's potential to manage the current safety concerns. The decision-making influences also vary by the ages of children. The special behavioral problems of children gain substantially greater salience when the children are older.
| Characteristic | TOTAL | Setting | ||||||
|---|---|---|---|---|---|---|---|---|
| In-Home | Out-of-Home | |||||||
| No Services | Services | TOTAL In-Home | Foster Care | Kinship Foster Care | Group Care | TOTAL Out-of-Home^^ | ||
| Mean^/ (SE) | ||||||||
| Likelihood that child would be reported for abuse or neglect in the next 24 months if. | ||||||||
| ...no services were provided | 2.4 (0.03) |
2.1 a (0.04) |
2.8 (0.1) |
2.3 b (0.03) |
3.4 (0.1) |
3.1 c (0.1) |
3.6 (0.1) |
3.3 (0.1) |
| ...in-home services were provided | 1.7 (0.03) |
1.6 d (0.03) |
1.8 (0.04) |
1.7 e (0.03) |
2.4 (0.1) |
2.3 (0.1) |
2.5 (0.2) |
2.4 (0.1) |
| ...placement into out-of-home care was provided | 1.4 (0.02) |
1.4 (0.03) |
1.4 (0.03) |
1.4 (0.02) |
1.4 (0.1) |
1.4 (0.1) |
1.5 (0.1) |
1.4 (0.1) |
|
^ Mean based on the following responses: 1 = very low likelihood of report in next 24 months, 2 = low likelihood, 3 = high likelihood, 4 = very high likelihood. (back) a Children remaining at home with services have a greater likelihood of re-report within 24 months if no services were provided than children remaining at home with no services, according to child welfare worker report (t = -8.7, p < .001). (back) b Children in out-of-home placement have a greater likelihood of re-report within 24 months if no services were provided than children remaining at home, according to child welfare worker report (t = -13.9, p < .001). (back) c Children in group care have a greater likelihood of re-report within 24 months if no services were provided than children In kinship foster care, according to child welfare worker report (t = -3.8, p < .001). (back) d Children remaining at home with services have a greater likelihood of re-report within 24 months if in-home services were provided than children remaining at home with no services, according to child welfare worker report (t = -5.7, p < .001). (back) |
4.2 Summary of Findings for Risk Assessment at Time of Investigation
4.2.1 History of Child Welfare Services
-
Among all families, 51% had a prior report of maltreatment (more than half of these had a prior incident of substantiated maltreatment) and about 30% had previously received services.
-
Child age and race/ethnicity were not factors in whether or not children and families with previous service receipt had higher levels of current service receipt—this relationship held across the board.
4.2.2 Caregiver Risk Factors at Time of Investigation
-
Agencies are very concerned about active substance abuse and serious mental health problems. Still, poor parenting—and the related concepts of motivation to change and cooperation with CWS—is the most significant factor influencing placement decisions.
-
Risk scores were predictive of placement decisions in that high parenting risk scores predicted placement in out-of-home care or receipt of services at home over having a case closed at home. In addition, high substance abuse risk scores predicted placement in out-of-home care as opposed to remaining at home with no services, and low cooperation scores predicted receipt of services at home as opposed to no receipt of services.
4.2.3 Critical Factors in Case Decision-Making
-
“Reasonable level of caregiver cooperation” was mentioned most often by child welfare workers as one of the two most critical factors influencing their case decision-making; “Child's inability to self-protect” was the second most common factor cited.
-
“Reasonable level of caregiver cooperation” was most influential in cases where the child ultimately remained at home.
-
In terms of child age, child welfare workers paid close attention to younger children's ability to self-protect and the special needs and behaviors of older children when making case decisions.
4.2.4 Likelihood of Future Reports of Abuse or Neglect
- Child welfare workers indicated large differences in the likelihood of re-reports for children currently receiving services over those not receiving services, for children placed into out-of-home care versus those served at home, and for those placed into group care versus those placed into kinship foster care.
4.3 Summary of Findings and Conclusions
The risk assessment items include items about the child and family's prior experience with CWS—in this case, summary measures. On a day-to-day basis, the families of children who are investigated for maltreatment in America's child welfare agencies are very likely to be known to the agency. Although this study is not definitive in assessing the impact of prior child welfare involvement on the current case plan, the relationship between the number of prior reports the family had and the child's current setting shows meaningful variation. Specifically, children in kinship care have significantly fewer prior family reports than children remaining at home with services and tend to have fewer prior family reports than children in nonkinship foster care and children in group care.
The risk assessment analysis has obvious limitations because the data were collected after the decision was made. These data do, nonetheless, offer valuable insights into the kinds of families that are being investigated and the child welfare worker's view of what key factors best reflect child and family characteristics. These characteristics can then be used to better understand who receives different types of services.
Child welfare workers assessed the likelihood of recurrence of maltreatment in ways that were highly supportive of the decisions made. Child welfare workers indicated large differences in the likelihood of re-reports for children currently receiving services over those not receiving services, for children placed into out-of-home care versus those served at home, and for those placed into group care versus those placed into kinship foster care. Child welfare workers clearly have considerable confidence in the ability of services opened at home to reduce risk, despite the evidence that the opening of in-home service cases is associated with greater supervision, which may increase the likelihood of re-reports (Johnson & L'Esperance, 1984; Fluke, Yuan, & Edwards, 1999).
There are some potentially meaningful findings, nonetheless, especially with regard to differences between kinship care and nonkinship care. Child welfare workers do not estimate the likelihood of re-reports for children in kinship care to be as great as they are for children in foster care. They also place children into kinship care when they have fewer reports against their parents than any other children who are receiving services. The precise reasons for this are not evident from these analyses, although these findings are consistent with evidence discussed in Chapter 5 and in other studies that show that children in kinship care have fewer behavioral problems.
| Table of Contents | Previous | Next |