Search Type Title Author/Investigator Search Advanced Search

Table of Contents | Previous | Next

9. Service Needs and Receipt

The preceding chapters have provided broad evidence that children coming to the attention of child welfare services are, generally, substantially more likely than the general population of children to experience a social or cognitive deficit. Child welfare services are intended to be a mechanism to help address safety and permanency issues for children and, increasingly, to be a place to address issues of child well-being (Goerge, 2002). The well-being of children is a cornerstone of the federal Child and Family Services Reviews, which assess the extent to which (1) families have enhanced capacity to provide for their children’s needs, (2) children receive appropriate services to meet their educational needs, and (3) children receive adequate services to meet their physical and mental health needs (U.S. DHHS, 2002).

This chapter presents information on the need for and use of physical health, mental health, and special education services by children in the NSCAW sample. Many studies have reported on the high rates of physical, mental and developmental problems among children entering out-of-home care in CWS (Simms, Dubowitz, & Szilagyi, 2000; Simms & Halfon, 1994), but data on children receiving in-home child welfare services are limited. NSCAW allows for an examination of the needs and service use for all children involved in CWS; results take into account the background and characteristics of the children.

Because “a stable and highly skilled workforce is necessary to effectively provide child welfare services” (U.S. GAO, 2003; p. 1), this chapter also examines characteristics of child welfare workers who work with the children and families involved with the CWS. Although much more data are available about the child welfare workers who completed the surveys about the children in the study, we have focused our analyses on their demographic characteristics, backgrounds, and training. Presented here is information on the age, race/ethnicity, length of experience, and education of these important players in the delivery of child welfare services. Additional information about how agency and PSU characteristics differ and may relate to child and child welfare worker characteristics is another piece of the puzzle. The chapter concludes with a description of how agency administration and PSU urbanicity and poverty rates relate to service delivery.

9.1 Children’s Problems and Services

This chapter begins with an analysis of children’s health status and use of medical services. Whether children involved in CWS are receiving appropriate levels of medical care is an important part of assessing their overall well-being. NSCAW data do not provide any independent information on the need for health care, so in the health analyses no distinction between the need for and the receipt of services can be made. An analysis of mental health and special education services is also provided. Both of these topics are of particular interest for this population, as the background of children involved with CWS presumably contributes to their being more prone to involvement with these services as well.

9.1.1 Health

General health characteristics of children involved with CWS were obtained by report from the child’s caregiver. The vast majority of caregivers (94%) report that their children are in good, very good, or excellent health (Table 9-1). This is lower, however, than the most recently published results from the National Health Interview Survey (NHIS) (Bloom & Tonthat, 2002a), in which 98% of children aged 17 and younger are reported to have good, very good, or excellent overall health status. Comparison of these studies shows that children involved with CWS are three times more likely to be in fair or poor health than children in the general population. Among children in NHIS living below the federal poverty line, a population that may be more similar to a sample of children involved with CWS, the rate of good or higher health status is 96%. For NSCAW children, there are no significant differences in overall health status by age, gender, race/ethnicity, or setting.

Table 9-1. Health Status of Children Involved With the Child Welfare System, by Caregiver Report

		TOTAL	In-Home			Out-of-Home
			No Services	Services	Total In-Home	Foster Care	Kinship Care	Group Care	Total Out-Of-Home
		Percent/ (SE)
Overall Health (good, very good, excellent)		93.6 (0.7)	94.4 (0.9)	93.4 (0.8)	94.2 (0.7)	87.7 (2.8)	89.2 (2.9)	92.6 (3.7)	88.9 (2.3)
Chronic Health Problem		27.5 (1.4)	27.1 (1.5)	32.3 (2.3)	28.0 (1.4)	25.3 (2.3)	23.2 (4.2)	17.1 (6.5)	23.6 (2.8)
Up-to-Date Immunizations		95.8 (0.5)	95.7 (0.6)	96.5 (0.7)	95.9 (0.5)	91.6 (4.0)	95.5 (1.3)	99.6 (0.3)	94.2 (1.9)
Dentist or Dental Hygienist	Ever seen	N/A	82.2 (2.0)	76.6 (2.0)	80.7 (1.6)	N/A	N/A	N/A	N/A
	Seen since CWS investigation date	57.9 (2.3)	60.7 (3.0)	52.4 (2.4)	58.4 (2.4)	48.8 (5.3)	49.4 (6.7)	80.7 (14.1)	52.4 (3.8)
Vision Testing	Ever seen	N/A	72.9 (2.0)	72.0 (2.1)	72.7 (1.8)	N/A	N/A	N/A	N/A
	Seen since CWS investigation date	N/A	N/A	N/A	N/A	50.7 (4.2)	52.3 (5.1)	73.2 (13.2)	53.4 (3.9)
Admitted to Hospital Overnight for Illness/Injury, Since CWS Investigation		5.7 (0.7)	5.4 (0.9)	6.9 (0.9)	5.8 (0.7)	7.4 (1.3)	3.3 (0.8)	1.3 (1.0)	4.8 (0.7)
Emergency Room for Illness/injury	Last 12 months	N/A	37.4 (2.7)	35.6 (2.3)	36.9 (2.3)	N/A	N/A	N/A	N/A
	Since CWS investigation date	N/A	N/A	N/A	N/A	26.7 (3.9)	16.4 (2.3)	21.4 (11.3)	21.0 (2.2)
Serious Injury/Accident Requiring Medical Attention	Past 12 months	N/A	10.7 (1.3)	9.4 (1.4)	10.4 (1.1)	N/A	N/A	N/A	N/A
	Since CWS investigation date	6.1 (0.7)	6.8 (1.0)	5.6 (0.8)	6.5 (0.8)	4.6 (1.2)	1.8 (0.8)	4.5 (2.4)	3.2 (0.7)

Although caregivers reported good, very good, or excellent health for 94% of the children, over one-quarter (28%) reported that their child has a health problem that “lasts a long time or comes back again and again.” The most common chronic conditions are other health problems (11%), asthma (8%), and repeated ear infections (4%). Five percent of caregivers report that their child has more than one chronic health problem. There are no significant differences in report of a chronic health problem by age, gender, race/ethnicity, or setting. Comparisons of overall health status with reports of chronic health problems showed that only 17% of children identified as having a chronic health problem are described by their caregiver as being in fair or poor overall health.

With regard to immunization status, 96% of children are up to date with their scheduled immunizations according to caregiver report; children aged 0-2 years are less likely to be up to date on their immunizations than older children (p<.001). Other health indicators were assessed by caregiver report, with the time frame varying by placement setting. Caregivers of children in their home of origin were asked about lifetime or past-year history, whereas the questions for out-of-home caregivers focused on the period since the start of the CWS investigation. For children remaining at home, 81% have been to a dentist or dental hygienist at least once and 73% have had their vision tested. Children receiving in-home services are less likely to have ever visited a dentist or dental hygienist than those children not receiving services (p< .05). For children in out-of-home care, over half (52%) have visited a dentist or dental hygienist and 53% have received vision testing since the investigation date. There are no significant differences in these health status indicators by gender, race/ethnicity, or setting, although there is a trend toward children in group care reporting higher rates of dental care since the investigation date (p< .03) than children in other out-of-home settings.

Caregivers also reported on serious injuries and illnesses among children in their care. Six percent of children involved with CWS have been admitted to a hospital overnight for illness or injury since the start of the investigation; a trend is present with children in group care being less likely to have been admitted to a hospital than children in other out-of-home care settings (p<.02). Similarly, 6% of all children involved with CWS have experienced a serious injury or accident requiring a visit with a physician since the investigation date. By setting, this is reported for 7% of children remaining at home and 3% of children in out-of-home care. Finally, 21% of children in out-of-home care have visited the emergency room due to an illness or injury since the investigation date, whereas over one-third (37%) of children remaining at home have been to the emergency room in the 12 months prior to the interview. This rate is much higher than the 12-month rate reported by the 2000 NHIS for all children under 18 years of age (20%) or for children in families below the poverty line (26%) (Bloom & Thonthat, 2002).

9.1.2 Mental Health

Mental health services are examined in three categories: specialty outpatient, nonspecialty outpatient, and inpatient. To assess the use of specialty mental health services by children involved with CWS, current caregivers were asked several questions. If the child was in the home of origin, the caregiver was asked if the child had ever been to any of the following for emotional, behavioral, learning, attention, or substance abuse problems: a mental health or community health center; a therapeutic nursery (for children under age 6); day treatment (for children aged 6 and older); or a psychiatrist, psychologist, social worker, or psychiatric nurse for private professional help. The same information was obtained from the current caregiver of children in out-of-home care, although the time frame was changed to ask if the child had utilized any of these services since the date that the child welfare agency investigation began. Therefore, the permanent caregiver was asked if the child had ever received these services, whereas the out-of-home caregiver was asked if the child had received these services since the start of the investigation.

Additional questions were asked of caregivers regarding use of other, nonspecialty mental health services for emotional, behavioral, learning, attention, or substance abuse problems, including in-home counseling or in-home crisis services; visits with a family doctor or other medical doctor; visits with a school guidance counselor, school psychologist, or school social worker (for children aged 6 and older); or attendance at an outpatient drug or alcohol clinic (for children aged 10 and older). Again, caregivers of children in their homes of origin were asked if the child had ever used these services for emotional, behavioral, learning, attention, or substance abuse problems, whereas caregivers of children in out-of-home care were asked if the child had received these services since the start of the investigation.

For both specialty and nonspecialty mental health service use questions, if the caregiver indicated that a service had been used, follow-up questions examined the starting and ending dates as well as the frequency of service use. Given the difference in the reporting time frame between in-home and out-of-home caregivers, it is not possible to compare frequencies on these items directly. Instead, using information on the CWS investigation date, it was determined which children were receiving mental health services at the time of the interview. The results are presented in Table 9-2.

Overall, 11% of children involved with CWS are receiving outpatient mental health services for emotional, behavioral, learning, attention, or substance abuse problems. A total of 7% are receiving specialty mental health services, with children most often receiving private professional help from mental health clinicians (5%), followed by mental health or community health center services (2%). Use of day treatment and therapeutic nurseries are reported less frequently (0.8% and 0.3%, respectively). Nonspecialty mental health services are also being used, with 8% reportedly receiving services from a school guidance counselor, followed by in-home counseling (3%) and services from a family doctor (1%).

Table 9-2 also indicates the proportion of children currently receiving mental health services by various child and case characteristics. Not surprisingly, age appears to be a significant factor with regard to whether a child is receiving either category (specialty or nonspecialty) of mental health services, with children aged 11 years and older more likely to be receiving services than younger children (p<.001). There are no significant differences in current use of mental health services by the child’s race/ethnicity, although a trend is present by most serious abuse type: children with a most serious abuse type of “failure to provide” are less likely than those in other abuse categories to be receiving specialty outpatient mental health services (p<.04).

Use of mental health services varies greatly by setting, with children in out-of-home care being more likely to receive specialty and nonspecialty mental health services (p<. 001) than children remaining at home. Twenty-one percent of children in out-of-home care are receiving specialty mental health services, as compared with 5% of children remaining at home. A similar pattern was seen for receipt of any type of outpatient mental health service, specialty or nonspecialty (31% vs. 9% respectively). Among children in out-of-home care, there are no significant differences in use of mental health services between children in foster or kinship care, while children in group care settings are significantly more likely to use specialty or nonspecialty mental health services (63%) than children in foster (27%) or kinship (29%) care.

Table 9-2. Caregiver Report of Mental Health Outpatient Services Currently Being Received by Children Involved With the Child Welfare System

		Specialty Mental Health					In-Home Counseling	Family Doctor	School Guidance Counselor	Alcohol & Drug Clinic	Any Out-patient Mental Health Services
		Mental Health Clinician	Community Mental Health Center	Day Treatment	Thera-peutic Nursery	Any Specialty Mental Health Out-patient Services
		Percent/ (SE)
Child Age	0-2	0.2 (0.1)	0.1 (0.03)	N/A	0.1 (0.1)	0.3 a (0.1)	0.9 (0.2)	0.3 (0.1)	N/A	N/A	1.0 c (0.2)
	3-5	1.6 (0.5)	0.4 (0.2)	N/A	0.4 (0.3)	2.3 b (0.6)	2.0 (0.8)	0.2 (0.1)	N/A	N/A	4.1 d (1.0)
	6-10	5.5 (1.2)	2.3 (0.7)	0.4 (0.2)	N/A	7.8 (1.4)	3.3 (0.9)	0.6 (0.3)	8.4 (1.6)	0.0 (0.0)	15.3 (2.3)
	11+	9.2 (1.6)	5.7 (1.3)	1.3 (0.6)	N/A	13.5 (1.9)	5.5 (1.3)	1.8 (0.7)	7.7 (1.4)	0.1 (0.1)	19.9 (2.4)
Child Race/ Ethnicity	African American	3.4 (1.1)	1.6 (0.6)	0.1 (0.1)	0.2 (0.1)	4.8 (1.4)	1.7 (0.4)	0.3 (0.2)	7.6 (1.6)	0.0 (0.0)	9.5 (1.4)
	White	5.4 (0.9)	2.9 (0.7)	1.0 (0.3)	0.1 (0.1)	7.8 (1.1)	3.9 (1.0)	1.1 (0.4)	8.4 (1.6)	0.2 (0.1)	12.8 (1.7)
	Hispanic	3.7 (1.5)	1.7 (0.9)	1.3 (1.0)	0.1 (0.04)	5.8 (1.9)	3.4 (1.5)	0.6 (0.5)	7.9 (3.4)	0.1 (0.1)	10.1 (2.7)
	Other	5.4 (2.5)	2.3 (1.3)	0.7 (0.4)	2.6 (2.4)	8.3 (3.0)	2.4 (1.1)	0.1 (0.1)	8.8 (4.3)	0.2 (0.2)	12.5 (3.6)
Child Setting/ Services	Total in-home	3.6 (0.5)	1.6 (0.4)	0.6 (0.3)	0.2 (0.2)	4.9 e (0.7)	2.5 (0.6)	0.6 (0.2)	6.5 (1.0)	0.1 (0.1)	9.1 (1.0)
	In-home, no services	3.0 (0.7)	1.3 (0.4)	0.4 (0.3)	0.2 (0.2)	4.4 (0.9)	1.3 (0.4)	0.6 (0.3)	5.5 (1.1)	0.03 (0.03)	7.3 (1.1)
	In-home, services	4.9 (1.3)	2.3 (0.6)	1.2 (0.6)	0.04 (0.03)	6.4 (1.3)	5.9 (2.0)	0.5 (0.2)	9.4 (2.0)	0.2 (0.2)	13.8 (2.2)
	Total out-of-home	13.2 (2.0)	8.4 (2.4)	2.0 (0.5)	1.2 (0.5)	21.1 (2.4)	8.2 (1.4)	2.2 (1.1)	23.0 (4.5)	0.4 (0.3)	31.2 f (3.5)
	Foster care	13.4 (2.5)	5.3 (1.8)	2.0 (1.0)	1.5 (0.9)	18.9 (2.9)	9.1 (2.2)	0.9 (0.3)	17.7 (4.5)	0.8 (0.7)	26.8 (4.1)
	Kinship foster care	8.9 (2.5)	9.0 (4.0)	1.6 (0.8)	0.6 (0.4)	18.0 (4.1)	2.8 (0.8)	1.6 (1.0)	27.9 (8.3)	0.0 (0.0)	29.1 (5.4)
	Group care	35.4 1 (12.1)	20.2 (9.9)	3.6 (1.9)	17.5 (15.7)	47.6 (11.6)	34.1 2 (10.0)	12.1 (9.8)	22.0 (7.7)	0.5 (0.5)	62.6 (11.9)
Most Serious Abuse Type	Physical abuse	5.4 (1.4)	2.6 (0.8)	1.3 (0.7)	0.1 (0.1)	8.3 (1.7)	3.3 (1.0)	0.9 (0.5)	8.2 (2.0)	0.1 (0.1)	14.3 (2.5)
	Sexual abuse	4.7 (1.5)	3.7 (1.6)	0.5 (0.3)	0	7.4 (2.0)	2.6 (1.2)	1.7 (1.1)	5.3 (1.5)	0.5 (0.4)	9.6 (2.2)
	Failure to provide	2.9 (1.0)	1.5 (0.6)	1.2 (0.8)	0.3 (0.1)	3.8 (1.1)	1.7 (0.5)	0.4 (0.1)	11.7 (3.7)	0.1 (0.1)	9.2 (1.9)
	Failure to supervise	3.6 (1.1)	1.5 (0.5)	0.3 (0.1)	0.5 (0.4)	4.9 (1.1)	3.4 (1.2)	0.1 (0.1)	8.5 (2.1)	0	10.4 (1.6)
	Other abuse	7.9 (2.3)	1.6 (1.2)	0.4 (0.2)	0.7 (0.6)	8.6 (2.4)	3.8 (1.7)	2.5 (1.5)	3.6 (1.2)	0	11.1 (2.6)
TOTAL		4.6 (0.5)	2.3 (0.4)	0.8 (0.3)	0.3 (0.1)	6.6 (0.7)	3.1 (0.6)	0.7 (0.2)	8.1 (1.1)	0.1 (0.1)	11.4 (1.1)

N/A = Not Applicable (Caregivers of children in this age group were not asked regarding use of this service). a Children aged 0-2 are less likely than children aged 11 and older to be currently receiving specialty outpatient mental health services, by caregiver report (X2 = 33.4, p<.001). (back) b Children aged 3-5 are less likely than children aged 11 and older to be currently receiving specialty outpatient mental health services, by caregiver report (X2 = 28.3, p< .001). (back) c Children aged 0-2 are less likely than children aged 11 and older to be currently receiving outpatient mental health services, by caregiver report (X2 = 47.3, p<.001). (back) d Children aged 3-5 are less likely than children aged 11 and older to be currently receiving outpatient mental health services, by caregiver report (X2 = 45.7, p<.001). (back) e Children remaining at home are less likely than children in out-of-home care to be currently receiving specialty outpatient mental health services, by caregiver report (X2 = 27.8, p<.001). (back) f Children remaining at home are less likely than children in out-of-home care to be currently receiving outpatient mental health services, by caregiver report (X2 = 25.5, p<.001). (back) 1 Caregivers were asked if child had received “private professional help from a psychiatrist, psychologist, social worker, or psychiatric nurse.” It is not possible to determine if these services were provided in the community or in the group care setting. (back) 2 Caregiver was asked whether child had received “any in-home counseling or in-home crises services.” The caregiver may be interpreting services received in the group care setting as in-home services. (back)

9.1.3 Specialty Outpatient Mental Health Services for Children

To further examine the relationship between various child and case characteristics and the receipt of outpatient specialty mental health services by children involved with CWS, a logistic regression was performed that modeled current receipt of any of the four specialty mental health services as reported by caregivers, controlling for race/ethnicity, age, gender, service setting, and most serious abuse type. 32 The results are summarized in Table 9-3. Younger children (aged 5 years and under) continue to be less likely to receive specialty mental health services than older children (p<.001), whereas those in out-of-home settings are more likely than those remaining at home and not receiving services to be receiving specialty mental health services (p<.001 for both foster and kinship foster care settings).

Table 9-3. Logistic Regression Modeling Current Receipt of Any Specialty Mental Health Outpatient Service by Children Involved with the Child Welfare System

		Any Specialty	Mental Health Service
		OR	95% CI
Age (continuous)		1.20**_i	1.14, 1.27
Gender	Male	0.89	0.43, 1.83
	Female	(reference group)
Race/Ethnicity	White	(reference group)
	African American	0.52	0.22, 1.22
	Hispanic	1.01	0.41, 2.45
	Other	1.32	0.58, 3.01
Child Setting/ Services	In home, no services	(reference group)
	In home, services	1.93	0.97, 3.84
	Foster care	8.89**_ii	4.72, 16.75
	Kinship foster care	5.90**_iii	3.27, 10.62
	Group care	#_i	#_ii
Most Serious Abuse Type	Physical abuse	(reference group)
	Sexual abuse	0.66	0.33, 1.33
	Failure to provide	0.50	0.21, 1.16
	Failure to supervise	0.52	0.25, 1.10
	Other	0.93	0.40, 2.18

Cox and Snell pseudo-R2 is NA ** p<.001 (back: **_i, **_ii, **_iii) # Due to small cell size, the odds ratio for Group care could not be calculated. (back: #_i, #_ii)

A similar logistic regression, adding the child’s CBCL score as a potential predictor variable, was run for children aged 2 and older (CBCL scores were not obtained on children under 2). Increasing age (p<.001) and out-of-home placement continued to be significantly related to increased likelihood of mental health service use (p<.001) for each of the three out-of-home placement settings), whereas children with CBCL scores above the borderline/clinical cut-point of 60 are almost four times more likely to be receiving mental health services than those below the cut-point (p<.001) (Table 9-4). Children receiving in-home CWS services were no more likely to also be receiving outpatient mental health services than those children in their home of origin who were not receiving CWS services. The involvement of child welfare services does not, at least in the short term, appear to increase access to children’s mental health services, unless the children are in out-of-home care. It should be noted, however, that system policies and procedures may impact who receives services. For instance, a policy may dictate that children placed in out of home care routinely receive a mental health assessment.

Table 9-4. Logistic Regression Modeling Current Receipt of Any Specialty Mental Health Outpatient Service by Children Involved with the Child Welfare System, Aged 2 Years and Older

		Any Specialty	Mental Health Service
		OR	95% CI
Age (continuous)		1.15**_i	1.08, 1.22
Gender	Male	0.80	0.39, 1.65
	Female	(reference group)
Race/ Ethnicity	White	(reference group)
	African American	0.54	0.23, 1.25
	Hispanic	1.11	0.44, 2.80
	Other	1.26	0.52, 3.04
Child Setting/ Services	In home, no services	(reference group)
	In home, services	1.76	0.86, 3.58
	Foster care	8.26**_ii	4.27, 15.97
	Kinship foster care	6.62**_iii	3.52, 12.46
	Group care	10.84**_iv	3.63, 32.34
CBCL Scores	Nonclinical score	(reference group)
	Borderline or clinical score	3.34**_v	2.02, 5.53
Most Serious Abuse Type	Physical abuse	(reference group)
	Sexual abuse	0.66	0.33, 1.34
	Failure to provide	0.54	0.25, 1.18
	Failure to supervise	0.57	0.28, 1.18
	Other	0.94	0.41, 2.13

Cox and Snell pseudo-R2 is .08 ** p<.001 (back: **_i, **_ii, **_iii, **_iv, **_v)

9.1.4 Inpatient Mental Health Services for Children

Use of inpatient services was examined through caregiver report. Lifetime use was reported for children remaining at home, whereas use since the CWS investigation date was reported for children in out-of-home care. Approximately 7% of children remaining at home had a lifetime history of inpatient service use (Table 9-5). There is no significant difference in inpatient mental health service use between children remaining at home who are receiving services and those who are not receiving services.

Table 9-5. Caregiver Report of Lifetime Receipt of Inpatient Mental Health Services for Children Living in Their Home of Origin

		In-Home
		No Services	Services	TOTAL In-Home
		Percent / (SE)
Institutional/ Inpatient	Psychiatric hospital	4.7 (0.9)	5.2 (0.8)	4.8 (0.7)
	Medical hospital, psychiatric unit	1.9 (0.6)	3.2 (0.7)	2.3 (0.5)
	Residential treatment center or group home	1.9 (0.6)	2.1 (0.6)	1.9 (0.5)
	Emergency shelter	1.6 (0.6)	1.4 (0.5)	1.5 (0.5)
	Hospital emergency room	2.2 (0.7)	3.3 (0.8)	2.5 (0.6)
	Detox or inpatient drug or alcohol services	0.2 (0.2)	1.2 (0.7)	0.5 (0.2)
Any Inpatient Services		7.1 (1.0)	7.6 (0.9)	7.2 (0.8)

Among children in out-of-home care, 13% have used inpatient mental health services since the investigation date (Table 9-6). There are significant differences in inpatient mental health service use among children in the three out-of-home placement settings, with those in group care being more likely than those in either foster care or kinship foster care to have used any inpatient services, as well as residential treatment center or group home services, since the CWS investigation date. It is important to note, however, that in the analyses examining inpatient services for children in out-of-home care, group care is considered both a placement setting and a mental health treatment, leading to higher overall rates of inpatient mental health service use by children in group care settings.

Logistic regressions, including the child’s CBCL score as a potential predictor variable, were run for two categories of inpatient service use: (1) receipt of any inpatient mental health services (Table 9-7) and (2) receipt of psychiatric hospital services (a psychiatric hospital or a psychiatric unit of medical hospital) (Table 9-8). For children remaining at home, increasing age and borderline or clinical CBCL score predicts lifetime use of inpatient services for both categories (p<.001); male children are also more likely than female children to report psychiatric hospitalization during their lifetime (p<.01).

The results of the logistic regression varied for children in out-of-home care by the type of inpatient service examined. When “any use of inpatient mental health service since CWS investigation date” was modeled, several variables were significant (out-of-home caregivers were asked only about services since the investigation date because they would have limited knowledge of the child’s lifetime history). Increasing age and borderline or clinical CBCL scores continued to be associated with use of inpatient services since the investigation date (p<.001 for each) (Table 9-9). In addition, African American children in out-of-home care are more likely than White children in out-of-home care to have utilized inpatient mental health services since the start of the investigation. Type of out-of home-placement is also a factor, with children in kinship care less likely (p<.05) and children in group homes (p<.01) more likely than children in foster care to have accessed inpatient mental health services. This may be somewhat misleading, however, as a group home may be both a placement setting and an inpatient mental health service, as noted earlier.

Table 9-6. Caregiver Report of Receipt of Inpatient Mental Health Services Since the CWS Investigation Date by Children in Out-of-Home Care

		Out-of-Home
		Foster Care	Kinship Foster Care	Group Care	TOTAL Out-of-Home
		Percent /(SE)
Institutional/ Inpatient	Psychiatric hospital	5.2 (2.2)	2.0 (0.8)	25.2 (10.6)	5.9 (1.6)
	Medical hospital, psychiatric unit	1.9 (1.2)	0.2 (0.2)	1.9 (1.2)	1.1 (0.5)
	Residential treatmentcenter or group home	3.9 a (1.6)	1.9 b (1.3)	54.8 (12.8)	8.9 (1.8)
	Emergency shelter	2.4 (0.9)	1.1 (0.6)	27.4 (10.7)	5.5 (1.8)
	Hospital emergency room	2.3 (1.3)	0.2 (0.1)	1.4 (1.0)	1.2 (0.5)
	Detox or inpatient drug or alcohol services	0.0 (0.0)	0.0 (0.0)	1.4 (1.1)	0.3 (0.2)
Any Inpatient Services		10.6 c (2.7)	4.1 d (1.4)	59.5 (13.1)	13.1 (2.0)

a Children in foster care are less likely than children in group care to have received residential treatment center or group home care since the CWS investigation date, by caregiver report (X2 = 13.9, p<.001). (back) b Children in kinship foster care are less likely than children in group care to have received residential treatment center or group home care since the CWS investigation date, by caregiver report (X2 = 14.6, p<.001). (back) c Children in foster care are less likely than children in group care to have received any inpatient mental health service since the CWS investigation date, by caregiver report (X2 = 13.1, p<.001). (back) d Children in kinship foster care are less likely than children in group care to have received any inpatient mental health service since the CWS investigation date, by caregiver report (X2 = 15.9, p<.001). (back)

When the regression model is run to examine use of inpatient psychiatric hospital services (both psychiatric hospital and psychiatric unit of medical hospital), the impact of race/ethnicity and placement setting are not evident. However, the associations of increasing age and borderline or clinical CBCL score (p<.001) with service use continue to be present (Table 9-10).

9.1.5 Special Education

Caregiver report of need for and use of special education services was also obtained for children involved with CWS. Over one-quarter of the caregivers report that they have ever been told by an education or health professional that their child had learning problems, special needs, or developmental disabilities. Caregivers also report that 17% of the children have been tested for learning problems since the start of the CWS investigation (the mean period from investigation start to interview date was 7.1 months) (Table 9-11). Almost one in five (19%) children is classified as needing special education services, defined as having the caregiver report that the child has been given an individualized education program (IEP) or individualized family service plan (IFSP). Fifteen percent are currently receiving special education services, as compared with 11% of children nationwide (U.S. Department of Education, 2001).

Table 9-7. Logistic Regression Modeling Lifetime Receipt of Any Inpatient Mental Health Services by Children in their Home of Origin, Aged 2 Years and Older

		Any Inpatient Mental Health Service Use
		OR	95% CI
Age (continuous)		1.29**_i	1.20, 1.39
Gender	Male	1.89	1.07, 3.34
	Female	(reference group)
Race/Ethnicity	White	(reference group)
	African American	0.73	.36, 1.48
	Hispanic	1.11	.51, 2.42
	Other	0.65	.26, 1.62
Child Setting/ Services	In home, no services	(reference group)
	In home, services	0.98	.61, 1.59
CBCL Scores	Non-clinical score	(reference group)
	Borderline or clinical score	3.34**_ii	1.82, 6.16

Cox and Snell pseudo-R2 is .08 n=2874 ** p<.001 (back: **_i, **_ii)

Table 9-8. Logistic Regression Modeling Lifetime Receipt of Psychiatric Hospital Services by Children in their Home of Origin, Aged 2 Years and Older

		Psychiatric Hospital Service Use
		OR	95% CI
Age (continuous)		1.28**	1.18, 1.38
Gender	Male	2.33*_i	1.25, 4.35
	Female	(reference group)
Race/Ethnicity	White	(reference group)
	African American	0.62	.27, 1.42
	Hispanic	1.06	.44, 2.54
	Other	0.70	.25, 1.96
Child Setting/ Services	In home, no CWS	(reference group)
	In home, CWS	0.98	0.56, 1.69
CBCL Scores	Nonclinical score	(reference group)
	Borderline or clinical score	2.88**_ii	1.41, 5.89

Cox and Snell pseudo-R2 is .06 n=2873 * p<.01 (back: *_i, **_ii) ** p<.001 (back)

Table 9-9. Logistic Regression Modeling Any Inpatient Mental Health Services Since the CWS Investigation Date by Children Living in Out-of-Home Care, Aged 2 Years and Older

		Any Inpatient Mental	Health Service Use
		OR	95% CI
Age (continuous)		1.39**_i	1.23, 1.56
Gender	Male	1.16	.46, 2.96
	Female	(reference group)
Race/ Ethnicity	White	(reference group)
	African American	4.83**_ii	2.11, 11.08
	Hispanic	1.43	.39, 5.27
	Other	1.10	.32, 3.82
Child Setting/ Services	Foster care	(reference group)
	Kinship foster care	0.30	.11, .82
	Group care	7.35*	2.25, 24.00
CBCL Scores	Nonclinical score	(reference group)
	Borderline or clinical score	8.25**_iii	3.58, 19.00

Cox and Snell pseudo-R2 is .27 * p<.01; (back) ** p<.001 (back: **_i, **_ii, **_iii)

Younger children (aged 0-2 and 3-5 years) are less likely than older children (aged 6-10 and 11 years and older) to have been identified by an education or health professional of having learning problems, special needs, or developmental disabilities (p<.001). Female children and African American or Hispanic children are also less likely than male children (p<.001) and White or “other” racial/ethnic category children (p<.01), respectively, to have a professional report a special education need. In addition, younger children (aged 0-2 and 3-5 years) are also less likely than children aged 11 years and older to have been tested for special education needs, to have an IEP/IFSP (p<.001 for both), or to be receiving special education services (p<.001 for both). Finally, children in their homes of origin are significantly less likely than those in out-of-home care to have been assessed for special education needs since the start of the CWS investigation (p<.001).

For those children who have not been tested for or classified as needing special education services, 24% of caregivers report that the child needs testing (Table 9-12). A majority of caregivers reporting an unmet need (62%) indicate that the testing has not been received for other reasons.

Table 9-10. Logistic Regression Modeling Any Psychiatric Hospital Services Since CWS Investigation Date by Children Living in Out-of-Home Care (Aged 2+)

		Psychiatric	Hospital Service Use
		OR	95% CI
Age (continuous)		1.26**_i	1.11, 1.44
Gender	Male	1.15	.36, 3.66
	Female	(reference group)
Race/ Ethnicity	White	(reference group)
	African American	1.94	.60, 6.34
	Hispanic	0.73	.15, 3.55
	Other	0.44	.10, 1.94
Child Setting/ Services	Foster care	(reference group)
	Kinship foster care	0.28	.08, 1.05
	Group care	2.06	.67, 6.34
CBCL Scores	Nonclinical score	(reference group)
	Borderline or clinical score	9.11**_ii	3.14, 26.44

Cox and Snell pseudo-R2 is .12 ** p<.001 (back: **_i, **_ii)

Caregivers were asked to report on the specific category of learning problem or special need that their child had. The most common special education classifications for children are specific learning disability (10%), speech or language impairment (7%), and emotional disturbance (7%) (Table 9-13). These rates are higher than those reported for children aged 6-17 years in the United States during the 1999-2000 school year (U.S. Dept. of Education, 2001). This report showed that 6% of children nationwide were receiving special education services for a specific learning disability, 2% for a speech or language impairment, and 1% for emotional disturbance.

The relationship between special education categories and age, race/ethnicity, gender, and setting of the child was also examined. When examining overall classification in a special education category, there were significant differences by age, race/ethnicity, and gender. Young children (aged 0-2) are less likely than older children to be classified in any category (p<.001). African American and Hispanic children are less likely than White children to be classified in any special education category (p<.01 for both). Finally, male children are more likely than female children to be classified in a special education category (p<.001). There are no significant differences in the reported disability categories by the child’s service setting.

Differences in the individual special education categories were also examined. Younger children (aged 0-2 and 3-5 years) have significantly lower rates of disabilities than older children (aged 11 years and older) for the emotional disturbance (p<.001 for aged 0-2, p<.01 for aged 3-5), specific learning disability (p<.001 for both), and other (p<.001 for aged 0-2, not significant for aged 3-5) categories. In addition, children aged 0-2 years, as well as those aged 11 years and older, are significantly less likely than children aged 3-5 years to be classified as speech- or language-impaired. The only significant difference between categories by racial/ethnic group occurred for the specific learning disability category, in which Hispanic children were less likely than White children to be classified (p<.01). For gender, a trend was present in which males (9%) were more likely to be classified as emotionally disturbed than females (5%) (p=.01).

Table 9-11. Caregiver Report of Testing, Classification, and Receipt of Special Education Services, by Child’s Placement Setting

		Ever told by professional that child has needs	Tested since investigation date	Has an IEP/IFSP	Currently receiving special education services
		Percent/ (SE)
Child Age	0-2	7.9 a b c (1.4)	11.9 k l (1.6)	3.8 n o p (1.2)	2.9 u v w (1.2)
	3-5	18.7 d e (2.4)	13.1 m_i (2.4)	9.9 q r (1.5)	7.8 x y (1.3)
	6-10	36.6 (3.1)	20.7 (2.5)	22.9 (2.7)	17.5 (2.2)
	11+	41.5 (3.7)	19.7 (1.9)	31.8 (3.1)	25.0 (2.8)
Gender	Male	33.7 f (2.3)	20.3 (1.8)	22.1 (1.9)	16.4 (1.5)
	Female	23.5 (1.8)	14.2 (1.4)	15.2 (1.3)	12.6 (1.4)
Child Race/ Ethnicity	African American	23.2 g h (2.3)	18.6 (2.0)	15.7 (2.0)	12.6 (1.8)
	White	33.0 i (2.2)	17.9 (1.6)	22.1 (1.9)	16.3 (1.5)
	Hispanic	21.2 j (3.2)	13.6 (2.6)	11.4 s t (1.9)	10.4 (1.8)
	Other	40.2 (5.4)	17.0 (4.1)	27.3 (5.2)	21.1 (4.9)
Child Setting/ Services	Total In-home	28.4 (1.7)	15.2 m_ii (1.2)	18.4 (1.3)	14.4 (1.2)
	In home, no services	27.7 (1.8)	14.0 (1.5)	17.4 (1.6)	13.7 (1.4)
	In home, services	30.3 (2.6)	18.6 (2.1)	21.1 (2.0)	16.3 (1.7)
	Total Out of Home	30.3 (3.4)	34.7 (2.7)	21.2 (2.9)	15.2 (2.5)
	Foster care	32.2 (3.4)	38.8 (4.4)	20.4 (3.4)	15.4 (3.1)
	Kinship foster care	27.5 (5.1)	27.6 (4.2)	18.3 (4.5)	11.6 (3.7)
	Group care	36.6 (9.7)	55.7 (12.6)	40.7 (11.2)	33.7 (9.4)
TOTAL		28.6 (1.5)	17.2 (1.1)	18.7 (1.3)	14.5 (1.0)

a Children aged 0-2 years are less likely than children aged 3-5 years to have been identified by a professional as having special education needs, by caregiver report (X2=13.6, p<.001). (back) b Children aged 0-2 years are less likely than children aged 6-10 years to have been identified by a professional as having special education needs, by caregiver report (X2=65.1, p <.001). (back) c Children aged 0-2 years are less likely than children aged 11 years and older to have been identified by a professional as having special education needs, by caregiver report (X2=55.6, p <.001). (back) d Children aged 3-5 years are less likely than children aged 6-10 years to have been identified by a professional as having special education needs, by caregiver report (X2=23.0, p <.001). (back) e Children aged 3-5 years are less likely than children aged 11 years and older to have been identified by a professional as having special education needs, by caregiver report (X2=29.4, p <.001). (back) f Female children are less likely than male children to have been identified by a professional as having special education needs, by caregiver report (X2=13.0, p <.001). (back) g African American children are less likely than White children to have been identified by a professional as having special education needs, by caregiver report (X2=8.4, p <.01). (back) h African American children are less likely than children in the “other” racial/ethnic category to have been identified by a professional as having special education needs, by caregiver report (X2=7.4, p <.01). (back) i Hispanic children are less likely than White children to have been identified by a professional as having special education needs, by caregiver report (X2=7.6, p <.01). (back) j Hispanic children are less likely than children in the “other” racial/ethnic category to have been identified by a professional as having special education needs, by caregiver report (X2=7.3, p <.01). (back) k Children aged 0-2 years are less likely than children aged6-10 years to have been tested for special education needs, by caregiver report (X2=8.0, p <.01). (back) l Children aged 0-2 years are less likely than children aged 11 years and older to have been tested for special education needs, by caregiver report (X2=10.3, p <.01). (back) m Children remaining at home are less likely than children in out-of-home care to have been tested for special education needs, by caregiver report (X2=36.3, p <0. 001). (back: m_i, m_ii) n Children aged 0-2 years are less likely than children aged 3-5 years to have an IEP/IFSP, by caregiver report (X2=13.2, p <0. 001). (back) o Children aged 0-2 years are less likely than children aged 6-10 years to have an IEP/IFSP, by caregiver report (X2=44.3, p <0. 001). (back) p Children aged 0-2 years are less likely than children aged 11 years and older to have an IEP/IFSP, by caregiver report (X2=58.4, p <0. 001). (back) q Children aged 3-5 years are less likely than children aged 6-10 years to have an IEP/IFSP, by caregiver report (X2=18.6, p <0. 001). (back) r Children aged 3-5 years are less likely than children aged 11 years and older to have an IEP/IFSP, by caregiver report (X2=34.9, p <0. 001). (back) s Hispanic children are less likely than White children to have an IEP/IFSP, by caregiver report (X2=13.2, p <0. 001). (back) t Hispanic children are less likely than children in the other racial/ethnic category to have an IEP/IFSP, by caregiver report (X2=8.8, p <0. 01). (back) u Children aged 0-2 years are less likely than children aged 3-5 years to be currently receiving special education services, by caregiver report (X2=11.9, p <0. 001). (back) v Children aged 0-2 years are less likely than children aged 6-10 years to be currently receiving special education services, by caregiver report (X2=42.8, p <0. 001). (back) w Children aged 0-2 years are less likely than children aged 11 years and older to be currently receiving special education services, by caregiver report (X2=45.9, p <0. 001). (back) x Children aged 3-5 years are less likely than children aged 6-10 years to be currently receiving special education services, by caregiver report (X2=17.3, p <0. 001). (back) y Children aged 3-5 years are less likely than children aged 11 years and older to be currently receiving special education services, by caregiver report (X2=32.3, p <0. 001). (back)

Table 9-12. Caregiver Report of Unmet Need for Special Education Services

		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)
If Child has not been tested or classified as “special needs”	Caregiver expresses “any” need for testing	24.3 (2.1)	22.1 (2.3)	26.2 (3.5)	23.2 (2.3)	40.3 (6.7)	27.9 (5.9)	53.8 (22.1)	34.8 (4.4)
Of caregivers who expressed unmet need, reasons for not getting tested:	Not available in area	4.1 (1.8)	6.0 (2.9)	0.9 (0.5)	4.5 (2.1)	3.3 (2.3)	0	0	1.6 (1.1)
	Wait-listed	4.6 (2.0)	5.3 (3.4)	3.9 (1.7)	4.9 (2.3)	4.9 (2.3)	1.5 (1.0)	0	3.0 (1.2)
	Ineligible	4.0 (1.8)	3.2 (2.2)	3.2 (1.6)	3.2 (1.7)	0.0 (0.0)	20.3 (15.7)	4.6 (4.6)	8.9 (7.3)
	Couldn’t be financed	2.8 (1.0)	2.0 (1.1)	4.8 (2.1)	2.8 (1.1)	0.0 (0.0)	7.0 (6.1)	0	2.9 (2.6)
	Scheduling/child care problem	2.0 (0.7)	2.3 (1.1)	1.1 (0.4)	1.9 (0.8)	2.7 (1.9)	2.2 (1.7)	0	2.2 (1.2)
	Transportation problem	5.8 (2.2)	8.6 (3.2)	2.2 (1.1)	6.8 (2.5)	0.1 (0.1)	0	0	0.03 (0.03)
	Other reason	62.6 (4.3)	60.1 (5.6)	66.9 (4.6)	62.6 (4.7)	75.2 (7.5)	59.2 (14.2)	13.9 (10.6)	62.4 (8.6)

Table 9-13. Caregiver Report of Special Education Categories for Children Involved with the Child Welfare System, by Age, Race/Ethnicity, Gender, and Child Setting

		Special Education Category
		TOTAL	Autism	Emotional Disturb.	Mental Retard-ation	Multiple Dis-abilities	Ortho. Impair.	Specific learning Dis-ability	Speech or language Impair.	Traumatic Brain Injury	Vision or Hearing	Other^
		Percent/ (SE)
Child Age	0-2	7.7 a b c (1.4)	0.0 (0.0)	1.1 i j (0.9)	0.1 (0.1)	0.4 (0.2)	0.9 (0.3)	0.8 n o p (0.4)	3.9 t (1.3)	0.6 (0.4)	0.5 (0.2)	2.8 v w (0.6)
	3-5	18.5 d (2.4)	1.0 (0.8)	3.9 k l (1.3)	1.8 (1.1)	0.9 (0.7)	1.1 (0.7)	5.1 q r (1.4)	12.1 (2.3)	0.2 (0.2)	2.9 (1.2)	6.5 (1.4)
	6-10	34.8 (3.2)	0.6 (0.4)	9.0 (1.6)	1.0 (0.3)	0.9 (0.4)	0.7 (0.3)	14.4 (2.1)	7.7 (1.3)	0.1 (0.1)	2.4 (0.9)	13.6 (1.9)
	11+	38.8 (3.8)	0.1 (0.1)	12.5 (2.1)	2.2 (0.9)	0.9 (0.4)	0.1 (0.1)	16.1 (2.1)	4.5 u (0.8)	0.7 (0.5)	1.3 (0.5)	11.8 (2.0)
Child Race/ Ethnicity	African American	22.1 e (2.2)	0.04 (0.04)	6.1 (1.2)	0.8 (0.3)	0.3 (0.1)	0.4 (0.2)	8.3 (1.6)	5.3 (0.9)	0.9 (0.5)	0.8 (0.3)	6.7 (1.4)
	White	32.3 (2.4)	0.2 (0.1)	9.1 (1.1)	1.5 (0.5)	0.7 (0.3)	0.6 (0.2)	12.2 (1.5)	8.2 (1.0)	0.1 (0.04)	2.4 (0.7)	10.8 (1.8)
	Hispanic	18.0 f g (3.7)	0.9 (0.9)	4.8 (1.7)	1.9 (1.0)	1.4 (0.8)	1.3 (0.8)	5.7 s (1.6)	7.1 (2.5)	0.3 (0.3)	2.6 (1.5)	9.0 (2.5)
	Other	38.3 (5.3)	2.7 (2.0)	6.7 (2.2)	0.1 m (0.1)	2.0 (1.1)	0.2 (0.1)	17.4 (4.7)	7.7 (2.4)	0.1 (0.1)	1.2 (0.7)	15.8 (4.6)
Gender	Male	33.0 h (2.3)	0.8 (0.4)	9.2 (1.3)	1.5 (0.5)	0.9 (0.3)	0.7 (0.2)	11.2 (1.3)	8.2 (1.2)	0.6 (0.3)	2.4 (0.7)	11.6 (1.4)
	Female	21.4 (1.9)	0.1 (0.04)	5.3 (0.8)	1.0 (0.4)	0.7 (0.3)	0.6 (0.3)	9.4 (1.2)	6.0 (0.9)	0.2 (0.1)	1.4 (0.4)	7.6 (1.2)
Child Setting /Services	In home, no services	26.0 (2.0)	0.4 (0.3)	5.8 (0.9)	1.3 (0.4)	0.7 (0.3)	0.4 (0.2)	10.3 (1.5)	6.7 (1.1)	0.3 (0.2)	1.1 (0.4)	9.9 (1.4)
	In-home, services	29.8 (2.6)	0.8 (0.5)	9.6 (1.7)	1.2 (0.3)	1.1 (0.3)	1.0 (0.4)	10.4 (1.3)	8.5 (1.1)	0.7 (0.3)	4.2 (1.4)	8.1 (1.3)
	Total In-Home	27.1 (1.9)	0.5 (0.3)	6.8 (0.9)	1.3 (0.3)	0.8 (0.3)	0.6 (0.2)	10.4 (1.2)	7.2 (0.9)	0.4 (0.2)	1.9 (0.5)	9.4 (1.1)
	Foster care	31.4 (3.4)	0.1 (0.1)	11.1 (2.3)	2.7 (0.8)	1.4 (0.5)	2.1 (0.7)	11.5 (2.0)	8.8 (1.7)	0.3 (0.2)	3.6 (1.0)	8.2 (1.5)
	Kinship care	24.9 (5.2)	0	8.7 (3.2)	0.3 (0.2)	0.3 (0.2)	0.4 (0.3)	7.4 (3.3)	5.4 (1.5)	0.04 (0.04)	0.6 (0.4)	14.9 (4.5)
	Group care	36.6 (9.7)	0.5 (0.5)	25.5 (8.6)	2.4 (1.9)	1.0 (1.0)	0	11.3 (4.8)	4.0 (3.0)	0.9 (1.0)	0	9.9 (3.7)
	Total out-of-home	28.7 (3.3)	0.1 (0.1)	11.3 (2.1)	1.5 (0.4)	0.9 (0.3)	1.1 (0.3)	9.5 (1.9)	6.7 (1.1)	0.2 (0.1)	1.8 (0.5)	11.6 (2.4)
TOTAL		27.2 (1.7)	0.5 (0.2)	7.3 (0.8)	1.3 (0.3)	0.8 (0.2)	0.6 (0.2)	10.3 (1.0)	7.1 (0.8)	0.4 (0.2)	1.9 (0.4)	9.6 (1.1)

^ The proportion of “other” is high because this includes “other health impaired”—a large category because it includes children with ADHD. (back) a Children aged 0-2 years are less likely than children aged 3-5 years to be classified in a special education category, by caregiver report (X2=13.8 p<.001). (back) b Children aged 0-2 years are less likely than children aged 6-10 years to be classified in a special education category, by caregiver report (X2=55.3 p<.001). (back) c Children aged 0-2 years are less likely than children aged 11 years and older to be classified in a special education category, by caregiver report (X2=52.5 p<.001). (back) d Children aged 3-5 years are less likely than children aged 11 years and older to be classified in a special education category, by caregiver report (X2 = 26.7 p<.001). (back) e African American children are less likely than White children category to be classified in a special education category, by caregiver report (X2=8.7 p<.01). (back) f Hispanic children are less likely than children in the other racial/ethnic category to be classified in a special education category, by caregiver report (X2=7.0, p=.01). (back) g Hispanic children are less likely than White children to be classified in a special education category, by caregiver report (X2=8.1, p=.01). (back) h Males are more likely than females to be classified in a special education category, by caregiver report (X2=17.9, p<.001). (back) i Children aged 0-2 years are less likely than children aged 6-10 years to be classified as emotionally disturbed, by caregiver report (X2=21.0 p<.001). (back) j Children aged 0-2 years are less likely than children aged 11 years and older to be classified as emotionally disturbed, by caregiver report (X2=23.1 p<.001). (back) k Children aged 3-5 years are less likely than children aged 6-10 years to be classified as emotionally disturbed, by caregiver report (X2=8.3, p<.01). (back) l Children aged 3-5 years are less likely than children aged 11 years and older to be classified as emotionally disturbed, by caregiver report (X2=11.0, p<.01). (back) m Children in the other racial/ethnic category are less likely than White children to be classified as mentally retarded, by caregiver report (X2=7.1, p<.01). (back) n Children aged 0-2 years are less likely than children aged 3-5 years to be classified with a specific learning disability, by caregiver report (X2=8.0, p<.01). (back) o Children aged 0-2 years are less likely than children aged 6-10 years to be classified with a specific learning disability, by caregiver report (X2=42.0, p<.001). (back) p Children aged 0-2 years are less likely than children aged 11 years and older to be classified with a specific learning disability, by caregiver report (X2=42.9, p<.001). (back) q Children aged 3-5 years are less likely than children aged 6-10 years to be classified with a specific learning disability, by caregiver report (X2=15.3 p<.001). (back) r Children aged 3-5 years are less likely than children aged 11 years and older to be classified with a specific learning disability, by caregiver report (X2=21.7 p<.001). (back) s Hispanic children are less likely than White children to be classified with a specific learning disability, by caregiver report (X2=7.0 p<.01). (back) t Children aged 0-2 years are less likely than children aged 3-5 years to be classified with a speech or language impairment, by caregiver report (X2=10.4, p<.01). (back) u Children aged 11 years and older are less likely than children aged 3-5 years to be classified with a speech or language impairment, by caregiver report (X2=9.1, p<.01). (back) v Children aged 0-2 years are less likely than children aged 6-10 years to be classified in the other special education category, by caregiver report (X2=28.7, p<.001). (back) w Children aged 0-2 years are less likely than children aged 11 years and older to be classified in the other special education category, by caregiver report (X2=17.9, p<.001). (back)

Children with at least one clinical or borderline score on the child social and cognitive development measures described in Chapter 5 were further examined. The relationship between child characteristics (age, gender, race/ethnicity, setting, and most serious abuse type) and the receipt of special education services for these children is presented in Table 9-14. Overall, 21% of children with one or more clinical scores are receiving special education services. There are significant differences in the percentage of children receiving special education services by age, with younger children (aged 0-2 and 3-5 years) being less likely than older children (aged 11 years and older) to receive services (p<.001 for both). This relationship continues when children with clinical cognitive and social scores are examined: younger children with clinical cognitive or social scores are less likely than older children in these groups to be receiving special education services. Finally, children in kinship care settings with at least one clinical cognitive score are less likely than children in group care settings or in-home children without or without CWS services to report receiving special education services (p<.01 for all).

Many trends are indicated when the receipt of special education services is compared with the child’s clinical cognitive and social scores. Overall, a trend is present among children with any clinical scores for whom failure to supervise is the most serious abuse type—these children are less likely to receive special education services than children with physical abuse (p<.5). White, African American, and Hispanic children with clinical cognitive scores are less likely than children with clinical cognitive scores in the “other” racial/ethnic category to be receiving special education services (p<.05 for all). When setting is examined, children with clinical cognitive scores in group care settings are more likely than those in foster care or those in their own homes, regardless of receipt of services, to be receiving special education (p<.03 for all).

Current use of special education services by children with clinical scores was then examined separately for two age groups: children under 6 years of age and those 6 years of age and older; this allows for school-aged children, who are more likely to be receiving services, to be examined separately. Logistic regression models were used to further explore the differences between children receiving special education services and those who were not receiving services, while controlling for age, race/ethnicity, gender, child setting, and most serious abuse type. The data in Table 9-15 indicate that, for children aged 0-5 years, there were significant differences by age and race ethnicity. Children aged 0-2 with a clinical cognitive score were less likely than children aged 3-5 to be receiving special education services (p<.01). African American and Hispanic children with clinical cognitive scores were less likely than White children with clinical cognitive scores to be receiving special education services; this pattern continued for African American children with clinical social scores (p<.01 for all).

When the same model was used to analyze the same variables for children aged 6 years and older, a significant difference surfaced by gender (Table 9-16). Male children are significantly more likely to be currently receiving special education services than female children (p<.01). Also among this older group of children, two trends continue to be present, with children aged 6-10 years being less likely (p=.02) to receive special education services than those in the older age group (aged 11 years and older), and children with “failure to supervise” as the most serious abuse type who have at least one clinical score being less likely to be currently receiving special education services than children in the “physical abuse” category (p<.03).

Table 9-14. Percent of Children Receiving Special Education Services, by Clinical Score

		At least one clinical score^	At least one clinical cognitive score^^	At least one clinical social score^^^
		Percent (SE)
TOTAL		20.8 (1.6)	40.8 (4.2)	21.7 (1.7)
Age	0-2	5.3 (2.5) a b	6.9 (3.7) e f	13.3 (7.9)
	3-5	11.8 (2.0) c d	16.8 (6.5) g h	12.2 (2.1) l m
	6-10	24.0 (3.2)	52.1 (6.1)	23.2 (3.0)
	11+	31.4 (3.5)	57.9 (6.6)	28.2 (3.6)
Gender	Male	23.1 (1.9)	40.9 (4.6)	25.0 (2.4)
	Female	18.5 (2.1)	40.8 (5.5)	18.5 (2.5)
Race/ Ethnicity	African American	17.4 (2.4)	30.8 (5.0)	19.8 (2.9)
	White	23.6 (2.3)	45.9 (6.4)	24.5 (2.6)
	Hispanic	15.3 (2.9)	33.6 (9.5)	13.0 (3.5)
	Other	31.1 (7.0)	76.9 (10.0)	33.8 (6.9)
Child Setting/ Services	In-home, no services	19.9 (2.1)	43.3 (5.1)	19.9 (2.2)
	In-home, services	22.0 (2.4)	38.2 (6.4)	24.1 (2.7)
	Foster care	23.2 (5.1)	38.4 (9.3)	26.9 (6.1)
	Kinship foster care	20.5 (6.4)	13.4 (4.4) i j k	24.4 (7.7)
	Group care	37.9 (11.7)	86.0 (8.9)	38.1 (12.0)
Most Serious Abuse Type	Physical	25.3 (4.0)	53.7 (8.2)	23.7 (3.7)
	Sexual	17.5 (3.5)	55.2 (13.3)	17.4 (3.5)
	Failure to provide	25.3 (3.8)	54.5 (7.7)	25.7 (4.1)
	Failure to supervise	16.1 (2.3)	26.3 (5.6)	17.9 (2.7)
	Other	18.2 (4.4)	26.6 (12.9)	20.9 (4.7)

^ Potential measures include BDI, BINS, CBCL, K-BIT, MBA, PLS-3, SSRS, TRF, and VABS. (back) ^^ Potential cognitive measures include K-BIT, MBA, and PLS-3. (back) ^^^ Potential social measures include CBCL, SSRS, TRF, and VABS. (back) a Children aged 0-2 years with at least one clinical score are less likely than children aged 6-10 years with at least one clinical score to be currently receiving special education services, by caregiver report (X2=22.1, p<.001). (back) b Children aged 0-2 years with at least one clinical score are less likely than children aged 11 years and older with at least one clinical score to be currently receiving special education services, by caregiver report (X2=29.9, p<.001). (back) c Children aged 3-5 years with at least one clinical score are less likely than children aged 6-10 years with at least one clinical score to be currently receiving special education services, by caregiver report (X2=10.6, p<0.01). (back) d Children aged 3-5 years with at least one clinical score are less likely than children aged 11 years and older with at least one clinical score to be currently receiving special education services, by caregiver report (X2=29.8, p<.001). (back) e Children aged 0-2 years with at least one clinical cognitive score are less likely than children aged 6-10 years with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=23.0, p<.001). (back) f Children aged 0-2 years with at least one clinical cognitive score are less likely than children aged 11 years and older with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=15.1, p<.001). (back) g Children aged 3-5 years with at least one clinical cognitive score are less likely than children aged 6-10 years with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=11.1, p<.01). (back) h Children aged 3-5 years with at least one clinical cognitive score are less likely than children aged 11 years and older with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=21.5, p<.01). (back) i Children in kinship foster care settings with at least one clinical cognitive score are less likely than in-home, no CWS services children with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=10.2,p<.01). (back) j Children in kinship foster care settings with at least one clinical cognitive score are less likely than in-home with CWS services children with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=8.6,p<.01). (back) k Children in kinship foster care settings with at least one clinical cognitive score are less likely than children in group care settings with at least one clinical cognitive score to be currently receiving special education services, by caregiver report (X2=8.9, p<.01). (back) l Children aged 3-5 years with at least one clinical social score are less likely than children aged 6-10 years with at least one clinical social score to be currently receiving special education services, by caregiver report (X2=8.8, p<.01). (back) m Children aged 3-5 years with at least one clinical social score are less likely than children aged 11 years and older with at least one clinical social score to be currently receiving special education services, by caregiver report (X2=14.3, p<.001). (back)

Table 9-15. Logistic Regression Modeling Current Receipt of Special Education Services by Children By Clinical Score Aged 0-5 Years

		At Least One Clinical Score^		At Least One Clinical Cognitive Score±		At Least One Clinical Social Score+
		OR	95% CI	OR	95% CI	OR	95% CI
Age	0-2	0.35	.12, 1.04	0.07*_i	.01, 0.44	0.80	.16, 3.98
	3-5	(reference group)		(reference group)		(reference group)
Race/ Ethnicity	White	(reference group)		(reference group)		(reference group)
	African American	0.27	.10, 0.70	0.10*_ii	.02, 0.58	0.22*_iii	.07, 0.67
	Hispanic	0.40	.10, 1.60	0.04*_iv	.00, 0.48	0.45	.10, 1.96
	Other	0.80	.17, 3.79	3.33	0.46, 23.96	1.00	.20, 4.86
Gender	Male	0.71	.23, 2.22	2.23	.56, 8.89	0.63	.20, 2.02
	Female	(reference group)		(reference group)		(reference group)
Child Setting/ Services	In-home, no services	(reference group)		(reference group)		(reference group)
	In-home, services	1.01	.42, 2.43	0.42	.07, 2.58	0.76	.27, 2.10
	Foster care	1.93	.62, 6.01	4.85	.92, 25.73	1.35	.29, 6.41
	Kinship foster care	0.53	.15, 1.91	0.55	.03, 11.30	0.37	.05, 2.81
	Group care	3.67	.20, 68.60	---	---	---	---
Most Serious Abuse Type	Physical	(reference group)		(reference group)		(reference group)
	Sexual	0.62	.08, 4.64	1.40	.16, 12.39	0.31	.05, 2.12
	Failure to provide	0.71	.28, 1.79	0.21	.02, 1.93	0.65	.22, 1.92
	Failure to supervise	1.02	.26, 4.08	0.55	.08, 3.79	1.13	.31, 4.16
	Other	0.13	.02, 0.96	---	---	.24	.04, 1.41

^ Potential measures include BDI, BINS, CBCL, K-BIT, MBA, PLS-3, SSRS, TRF, and VABS. (back) ± Potential cognitive measures include K-BIT, MBA, and PLS-3. (back) + Potential social measures include CBCL, SSRS, TRF, and VAB (back) Cox and Snell pseudo-R2 is .05 (for at least one Clinical score), 0.25 (for at least one Cognitive score), and 0.05 (for at least one Social score) * p<.01 (back: *_i, *_ii, *_iii, *_iv)

When children aged 6 years and older with at least one clinical cognitive score are examined (Table 9-16), the bivariate findings that children in kinship foster care are less likely than in-home children not receiving CWS services to report the current receipt of special education services were continued (p<.01). Two trends are also present: children aged 6-10 years and Hispanic children are less likely to receive services than children aged 11 years and older and White children, respectively (p<.05 for both). Two trends are present for children aged 6 years and older who have at least one clinical social score: (1) Hispanic children are less likely than White children to be currently receiving special education services (p<.03) and (2) children whose most serious abuse type is “failure to supervise” are less likely than those in the “physical abuse” category to report current receipt of special education services (p<.02).

9.1.6 Summary of Children’s Problems and Services

Children involved with CWS have lower cognitive and academic abilities and more problem behaviors than children in the general population (see Chapter 5), which would indicate that children involved with CWS experience high levels of need for mental health and special education services. These analyses also show that, while children placed in out-of-home care have higher levels of need than those remaining at home, both groups have levels of need greater than the general population. This would indicate that children involved with the child welfare system have elevated needs for physical, mental, and developmental services, regardless of whether they are placed in out-of-home care. This is an important finding, as higher levels of need have been established for children in foster care, but no previous national study has examined the needs of children who remained in their own homes.

Examination of mental health service use showed that children in out-of-home care are more likely than those remaining at home to be receiving outpatient services. These results could not be replicated for inpatient services, as the NSCAW survey examined different service time frames for children remaining at home and those in out-of-home care. The child’s level of need, as measured by CBCL, is significantly associated with the receipt of both outpatient and inpatient services. Receipt of services is clearly driven by need: as mental health and behavior problems increase, so does the likelihood that the child is receiving services. Yet a sizable group of children in need of services does not receive them, despite their involvement with CWS, which could catalyze such service receipt. For example, approximately 40% of children scored in the borderline/clinical range on CBCL, an accepted measure of children’s mental health and behavioral and emotional functioning, while only 11% of children are currently receiving mental health services.

Special education services are being used at higher rates among children involved with CWS, especially those experiencing neglect, than in the general population. Children in out-of-home care are more likely to have been tested for special education needs but are not more likely to have an IEP/IFSP, been identified by a professional as needing special education services, or to be receiving services, perhaps indicating that children in out-of-home care are targeted for assessment on the basis of the placement itself. When need for special education services was examined in relation to service receipt, younger children (aged 0-5 years) were shown to be less likely than older children to be receiving services. Among children aged 6 years and older with a need for services, males are more likely to be receiving services than females, while for those with clinical cognitive scores, children in kinship foster care settings are less likely to report current receipt of services than those in group care settings or in their home of origin.

Table 9-16. Results of Logistic Regression Modeling Current Receipt of Special Education Services by Children by Clinical Score (Aged 6+)

		At Least One Clinical Score^		At Least One Clinical Cognitive Score±		At Least One Clinical Social Score+
		OR	95% CI	OR	95% CI	OR	95% CI
Age	6-10	0.53	.31, .90	0.45	.20, 1.01	0.60	.34, 1.03
	11+	(reference group)		(reference group)		(reference group)
Race/ Ethnicity	White	(reference group)		(reference group)		(reference group)
	African American	0.73	.46, 1.16	0.56	.22, 1.38	0.86	.54, 1.37
	Hispanic	0.57	.33, 0.99	0.35	.12, 0.98	0.39	.17, .90
	Other	1.58	.63, 3.97	4.35	.67, 28.09	1.45	.62, 3.38
Gender	Male	1.79*_i	1.27, 2.53	2.03	1.02, 4.04	2.03	1.28, 3.20
	Female	(reference group)		(reference group)		(reference group)
Child Setting/ Services	In-home, no services	(reference group)		(reference group)		(reference group)
	In-home, services	1.00	.64, 1.57	0.61	.30, 1.24	1.29	0.84, 1.98
	Foster care	1.16	.46, 2.89	0.38	.06, 2.27	1.65	.70, 3.88
	Kinship foster care	0.87	.40, 1.89	0.13*_ii	.03, .48	1.08	.48, 2.44
	Group care	1.21	.34, 4.36	2.17	.55, 8.58	1.51	0.41, 5.63
Most Serious Abuse Type	Physical	(reference group)		(reference group)		(reference group)
	Sexual	0.79	.45, 1.40	3.09	.58, 16.45	0.95	.51, 1.75
	Failure to provide	1.42	.65, 3.10	1.67	.53, 5.21	1.27	.59, 2.76
	Failure to supervise	0.55	.34, .91	0.36	.12, 1.10	0.59	.35, 1.00
	Other	0.88	.46, 1.66	0.74	.14, 3.79	1.12	.59, 2.14

^ Potential measures include BDI, BINS, CBCL, K-BIT, MBA, PLS-3, SSRS, TRF, and VABS. (back) ± Potential cognitive measures include K-BIT, MBA, and PLS-3. (back) + Potential social measures include CBCL, SSRS, TRF, and VAB (back) Cox and Snell pseudo-R2 is .06 * p<.01 (back: *_i, *_ii)

9.2 Child Welfare Worker Characteristics

No national survey of child welfare workers has ever been undertaken. However, because they play such a direct and important role with children involved with CWS, taking a preliminary look at who these workers are is logical and worthwhile. Although the child welfare workers in this study were not randomly selected from all child welfare workers, they are a reasonable approximation of a random sample of child welfare workers who become involved with children and families at intake into CWS or who follow their cases in the first months after entering CWS. We do not assume that these child welfare workers represent all child welfare workers, as they are primarily assigned to intake and investigations units—although some are carrying the cases of the children in foster care—and their general characteristics could, for example, be quite different from workers in the adoption program. Nonetheless, given the import role of intake and investigation child welfare workers—who are involved with decisions about more than 50,000 referrals per week alleging that children were abused (U.S. DHHS, 2003)—this is an important group to understand.

This section contains a brief summary of the characteristics of child welfare workers involved with children in the NSCAW sample. Although children may have had more than one child welfare worker, we ascertained the characteristics of a single child welfare worker at the time of the baseline data collection. Child welfare workers were selected to describe the case if they knew the most about the case at the investigation stage and had ready access to case materials.

Child welfare workers are a diverse group, with a broad distribution of ages, races, and ethnicities, educational types, and experiences. On average, the child welfare workers in this sample have 7 years of experience (Table 9-17). A logistic regression (controlling for child welfare worker age, race/ethnicity, and education level) showed no significant difference in experience between child welfare workers who work with children remaining at home and those who work with children living out of the home.

The range of experience is approximately the same for each education group, ranging from less than 1 year to 30 years or more (Table 9-18). Overall, 9% of child welfare workers have less than 1 year of experience, 17% have between 1 and 2 years of experience, 24% have between 2 and 5 years of experience, 23% have between 5 and 10 years of experience, 19% have 10 to 20 years of experience, and almost 8% have 20 years or more of experience.

The average length of experience of child welfare workers varies somewhat by race/ethnicity, with White workers having significantly more experience than Hispanic workers (Table 9-19).

A regression of child welfare worker age, race/ethnicity, and education level on length of experience showed that only age is significantly associated with length of experience (Table 9-20). Not surprisingly, older child welfare workers have more experience than younger child welfare workers. Race/ethnicity and education are not significantly associated with length of experience. However, when age is not included in the model, Hispanic workers have significantly less experience than White workers (p<.001).

We also looked to see whether the race/ethnicity of the child matched the race/ethnicity of the child welfare worker. We found that about 34% of African American children have an African American child welfare worker, 74% of White children have a White child welfare worker, 27% of Hispanic children have an Hispanic child welfare worker, and about 12% of children of other races/ethnicities have child welfare workers of other races/ethnicities (Table 9-21).

Table 9-17. Child Welfare Worker Characteristics, by Child Service Setting

		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)
Child Welfare Worker Age	<30	25.9 (2.8)	63.6 (2.6)	22.9 (2.0)	86.5 (2.1)	5.2 (1.1)	5.9 (1.3)	1.2 (0.5)	13.5 (2.1)
	30-39	34.0 (2.1)	66.8 (3.0)	22.1 (2.3)	88.9 (1.7)	4.7 (1.0)	4.4 (0.8)	1.2 (0.4)	11.2 (1.7)
	40-49	21.2 (1.9)	65.4 (2.8)	25.6 (2.7)	91.1 (1.6)	3.2 (0.5)	4.8 (1.2)	0.5 (0.3)	9.0 (1.6)
	≥ 50	18.9 (2.3)	62.4 (3.6)	26.9 (3.4)	89.3 (1.7)	4.7 (0.9)	3.7 (0.8)	1.3 (0.8)	10.7 (1.7)
Child Welfare Worker Race/ Ethnicity	African American	20.4 (2.4)	68.3 (3.6)	21.6 (2.5)	89.9 (1.8)	3.8 (0.7)	4.3 (0.9)	0.5 (0.2)	10.1 (1.8)
	White	65.4 (2.9)	64.2 (1.9)	25.0 (1.6)	89.2 (1.5)	4.6 (0.8)	4.2 (0.6)	1.3 (0.3)	10.8 (1.5)
	Hispanic	8.8 (1.9)	62.6 (5.7)	25.5 (5.5)	88.1 (2.1)	4.1 (1.4)	7.1 (1.9)	0.4 (0.2)	12.0 (2.1)
	Other	5.4 (1.0)	69.2 (5.1)	12.8 (2.5)	82.0 (4.3)	6.5 (3.2)	9.8 (3.7)	1.0 (0.5)	18.0 (4.3)
Length of Experience, in Years	Mean(SE)	6.9 (0.4)	7.1 (0.5)	6.8 (0.5)	7.1 (0.5)	5.4 (0.4)	5.5 (0.4)	8.5 (2.9)	5.7 (0.4)
Child Welfare Worker Education	<BA	2.6 (0.7)	51.4 (14.8)	41.6 (15.5)	93.0 (3.7)	3.0 (0.8)	---	---	7.0 (3.7)
	BA/BSW	71.4 (3.2)	65.6 (2.1)	23.6 (1.6)	89.2 (1.1)	4.2 (0.6)	4.8 (0.7)	0.8 (0.2)	10.8 (1.1)
	MSW	12.1 (1.7)	64.3 (3.7)	22.6 (2.5)	86.9 (2.9)	5.3 (1.5)	5.3 (1.4)	2.1 (1.2)	13.1 (2.9)
	Other	14.0 (3.1)	67.4 (3.6)	21.2 (2.9)	88.6 (3.2)	4.9 (1.8)	4.8 (1.1)	1.4 (0.7)	11.4 (3.2)
TOTAL		100	65.3 (1.6)	23.6 (1.4)	88.9 (1.3)	4.4 (0.6)	4.8 (0.7)	1.0 (0.2)	11.1 (1.3)

^ Except where indicated (back) ^^ Includes children in other types of out-of-home care. (back)

Table 9-18. Length of Experience as Child Welfare Worker by Education

	Less than a BA	BA or BSW	MSW	Other
Mean years of experience (SE)	8.1 (1.1)	6.5 (0.5)	8.3 (0.8)	7.7 (0.6)
Range of years of experience	0.25 – 31.0	0.083 – 36.0	0.083 – 36.0	0.167 – 50.0

Table 9-19. Length of Experience as Child Welfare Worker by Race/Ethnicity

	African American	White	Hispanic	Other
Mean years of experience (SE)	6.5 (0.8)	7.2a(0.5)	5.5 (0.5)	7.5 (1.4)
Range of years experience	0.083 – 30.0	0.083 – 36.0	0.083 – 28.0	0.083 – 50.0

a White child welfare workers have significantly more experience than Hispanic child welfare workers (t=3.1, p<.01) (back)

Table 9-20. Regression Results for Explaining Length of Child Welfare Worker Experience, in Years

		Beta Coefficient (SE)
Age	<30 years	(reference group)
	30-39 years	3.76 (.42)*_i
	40-49 years	7.55 (.85)*_ii
	≥ 50 years	9.5 (1.5)*_iii
Race/ Ethnicity	African American	-.51 (.68)
	White	(reference group)
	Hispanic	-1.39 (.55)
	Other	.37 (1.04)
Education	Less than a BA	-.79 (1.74)
	BA/BSW	(reference group)
	MSW	.80 (.85)
	Other	-.59 (.89)

* p<.001 (back: *_i, *_ii, *_iii)

Table 9-21. A Comparison of the Child Welfare Worker’s Race/Ethnicity with the Child’s Race/Ethnicity

Child Race/Ethnicity	Child Welfare Worker Race/Ethnicity
	African American	White	Hispanic	Other
	Percent(SE)
African American	34.5 (4.2)	58.3 (4.0)	3.5 (1.0)	3.7 (1.1)
White	15.6 (2.1)	74.3 (2.6)	5.6 (1.5)	4.5 (1.2)
Hispanic	15.6 (2.7)	49.6 (4.2)	27.3 (4.3)	7.5 (1.6)
Other	9.5 (3.0)	75.0 (4.7)	3.0 (1.3)	12.5 (3.9)

A comparison of the child welfare worker’s race/ethnicity by education level showed that the majority of child welfare workers received a bachelor’s degree regardless of their race/ethnicity (Table 9-22). Almost three-quarters of African American and White child welfare workers hold a BA or BSW and about two-thirds of Hispanic child welfare workers and child welfare workers of other races/ethnicities hold a BA or BSW. African American child welfare workers are the least likely to have an MSW, Hispanic and White child welfare workers are equally likely to hold an MSW, and those of other races/ethnicities are most likely to hold an MSW.

Table 9-22. Child Welfare Worker Education by Race/Ethnicity

	Less than a BA	BA or BSW	MSW	Other^
	Percent (SE)
African American	2.9 (1.4)	74.8 (3.0)	8.0 (1.9)	14.3 (2.7)
White	2.4 (1.1)	72.0 (3.8)	13.2 (2.1)	12.5 (3.1)
Hispanic	4.5 (2.0)	60.9 (6.4)	12.0 (4.3)	22.6 (6.3)
Other	---	65.3 (10.7)	15.8 (6.2)	18.1 (8.5)
Total	2.6 (0.7)	71.4 (3.2)	12.1 (1.7)	14.0 (3.1)

^ Includes child welfare workers who held either another type of master’s degree (besides an MSW) or doctoral degree. (back)

The races/ethnicities of child welfare workers who hold an MSW do not differ significantly from the races/ethnicities of child welfare workers who do not hold an MSW. About one-fifth of child welfare workers with an MSW are less than 30 years old, another three-fifths are between 30 and 49 years old, and the final one-fifth are 50 years old or more.

9.2.1 Summary of Child Welfare Worker Characteristics

Child welfare workers are a diverse group in terms of age, race/ethnicity, and education level. Overall, the largest proportions of child welfare workers are between 30 and 39 years of age, White, and hold a bachelor’s degree. The average length of experience for child welfare workers is 7 years. Our analyses do not indicate any clear predictors of length of experience (other than age), as race/ethnicity and educational level are not significantly associated with length of experience.

Nearly three-quarters of White children have a child welfare worker of the same race/ethnicity; only one-quarter to one-third of Hispanic and African American children have child welfare workers of the same race/ethnicity. This reflects the predominance of White child welfare workers. With regard to education, about two-thirds to three-quarters of the child welfare workers have a bachelor’s degree but no advanced degree. Only 12% of child welfare workers have an MSW, with an additional 14% holding another master’s degree or a doctorate (which could be in addition to a degree in social work). The races/ethnicities of child welfare workers who hold an MSW do not differ significantly from the races/ethnicities of child welfare workers who do not hold an MSW.

9.3 PSU (County) Characteristics and Service Delivery

Although federal policy has a major role in governing the delivery of child welfare services, many local conditions also influence their delivery (Mitchell et al., in press). To examine the extent to which these local conditions may affect service delivery, we tested for differences in select case characteristics by PSU (county) characteristics. 33 Granting that the relationship between many such conditions and child welfare services delivery could be tested, just three PSU characteristics were compared in this analysis: administration auspices of the child welfare agency (state or county), urbanicity (rural or urban), and poverty level (poor or non-poor). Consistent with Census Bureau definitions, urban was defined as greater than 50% of the population living in an urban area, whereas rural was defined as all areas that did not meet this requirement. Poverty level was defined as either (a) non-poor, 5% or less of families with children living below the 50% poverty level, or (b) poor, greater than 5% of families with children living below the 50% poverty level.

We identified a limited set of possible relationships between PSU characteristics and CWS processes for testing. Many other questions could have been tested, but these were selected because we had some prior evidence that they varied by locality. One case characteristic was the child’s race/ethnicity, which has been shown to vary considerably with regard to entrance into CWS across counties (Barth, in press; Needell, Brookhart, & Lee, 2002; Ards et al., 2002). We also know that there are different policies and practices that govern the use of foster care, kinship foster care, and group care (Berrick et al., 1998; Geen, 2001), so we examined the child’s service setting. Previous research suggests that children with different numbers of risks present in the household at the time of placement may have different likelihoods of being placed (Shlonsky & Gambrill, 2001). Further, because of the evidence that children’s involvement with mental health services may be associated with their degree of mental health problems (Garland & Landsverk, in press), we tested PSU characteristics against mean number of risks present at the time of placement and the proportion of children with a borderline or clinical CBCL score. Because hiring and training practices may also vary by locality (Perry & Dickinson, in press), we examined four child welfare worker characteristics: length of service as a child welfare worker, highest child welfare worker degree, race/ethnicity, and age.

Table 9-23 presents the results of analyses that compared these case characteristics by PSU dimension, which is different than comparing these relationships by location, per se. That is, practices in two different counties may vary substantially, but this variation may not be explained by any of the three PSU characteristics examined. For the majority of the variables related to child characteristics (i.e., setting, risks, and CBCL score), there were no significant differences by agency administration, urbanicity, or poverty level, although there were some significant findings related to the child’s race/ethnicity, as described below.

Some significant associations between PSU characteristics and service or client characteristics do exist. With regard to the length of service as a child welfare worker, child welfare workers of children in urban PSUs had significantly longer lengths of service than did those in rural PSUs, with an average of 2.5 years’ more experience.

Both the race/ethnicity of the child welfare worker and the race/ethnicity of the child exhibited notable differences with regard to PSU type, with urban PSUs appearing more racially and ethnically diverse. The child welfare workers of children in rural PSUs are significantly more likely to be White (p<.001) and significantly less likely to be Hispanic (p<.001) than those of children in urban PSUs. On average, almost three-quarters or more of the child welfare workers of children in these counties are White, with Hispanic child welfare workers being a rare occurrence (less than 1% in rural PSUs). In addition, there is a trend indicating that child welfare workers of children in rural PSUs are less likely to be of some other race/ethnicity than those of children in urban PSUs (X2=5.5, p=.02). An additional trend indicates that child welfare workers of children in poor PSUs are more likely to be African American than those of children in non-poor PSUs (X2=5.2, p=.03).

Table 9-23. Select Case Characteristics by PSU Characteristics

		PSU Characteristics
		Percent^ (SE) (unless otherwise indicated)
		Agency Administration		Urbanicity		Families in Poverty
		State	County	Rural	Urban	Non-poor	Poor
Child Race/ Ethnicity	African American	24.6 (3.3)	32.8 (3.5)	17.3 (5.8)	31.6 (2.6)	23.0 (3.5)	32.3 (3.4)
	White	53.9 (3.7)	37.8 (5.6)	74.4 (5.9) a	38.3 (3.6)	51.9 (5.7)	42.9 (5.1)
	Hispanic	14.5 (2.1)	22.6 (5.8)	3.4 (1.1) b	22.7 (3.5)	17.0 (4.7)	18.9 (4.5)
	Other	6.9 (1.3)	6.8 (0.7)	5.0 (1.2)	7.5 (1.0)	8.2 (1.3)	5.9 (1.0)
Child Setting^^	In-home, no services	63.2 (2.2)	66.5 (2.5)	70.5 (3.3)	62.8 (1.8)	63.4 (2.5)	65.7 (2.4)
	In-home, services	23.4 (1.5)	24.8 (2.9)	21.7 (2.5)	24.8 (1.8)	26.8 (2.2)	21.8 (2.1)
	Foster care	5.2 (0.7)	3.5 (1.0)	3.6 (1.0)	4.7 (0.7)	3.6 (0.7)	5.1 (0.9)
	Kinship foster care	5.8 (0.8)	4.1 (1.0)	3.2 (0.7)	5.6 (0.8)	4.8 (0.9)	5.3 (0.8)
	Group care	1.1 (0.3)	0.8 (0.3)	0.7 (0.4)	1.1 (0.3)	0.9 (0.3)	1.0 (0.3)
Number of Risks Present at Time of Placement^^^(Mean) (SE)		6.6 (0.2)	6.1 (0.2)	6.7 (0.3)	6.3 (0.2)	6.5 (0.2)	6.3 (0.2)
Borderline/ Clinical CBCL Score		51.8 (2.4)	48.7 (2.8)	57.4 (2.5)	48.2 (2.2)	50.4 (3.1)	50.5 (2.1)
Length of Service as a Child Welfare Worker in Months (Mean) (SE)		79.9 (6.8)	86.9 (6.9)	59.7 (7.9) c	90.4 (5.1)	87.0 (5.6)	79.6 (7.9)
Highest Degree of Child Welfare Worker	<Bachelor’s	1.7 (0.5)	3.7 (1.5)	3.9 (1.3)	2.1 (0.9)	3.6 (1.4)	1.7 (0.6)
	BSW	28.0 (3.3)	23.9 (3.9)	30.3 (5.3)	25.0 (2.9)	22.5 (3.8)	29.4 (3.6)
	Other Bachelor’s	51.0 (3.7)	37.3 (4.9)	55.0 (6.6)	41.9 (3.3)	44.7 (4.2)	45.5 (4.8)
	MSW	10.6 (1.9)	14.1 (3.2)	7.9 (2.8)	13.5 (2.1)	12.9 (3.1)	11.4 (1.8)
	Other Master’s	8.3 (2.0)	18.9 (5.2)	2.7 (1.0)	16.1 (3.4)	14.8 (4.0)	11.2 (4.6)
	Doctorate	0.4 (0.2)	2.1 (0.7)	0.3 (0.2)	1.4 (0.5)	1.5 (0.6)	0.8 (0.6)
Child Welfare Worker Race/Ethnicity	African American	19.9 (3.1)	21.0 (3.6)	14.4 (3.8)	22.3 (2.8)	14.8 (3.0)	25.0 (3.5)
	White	66.7 (3.3)	63.7 (5.0)	82.3 (4.2) d	60.0 (3.2)	71.5 (3.8)	60.3 (4.1)
	Hispanic	8.2 (2.7)	9.6 (2.7)	0.8 (0.5) e	11.4 (2.4)	7.2 (1.7)	10.2 (3.4)
	Other	5.2 (1.4)	5.6 (1.4)	2.4 (0.9)	6.4 (1.2)	6.5 (1.5)	4.5 (1.5)
Child Welfare Worker Age	< 30	28.7 (3.8)	22.1 (4.1)	35.2 (8.4)	22.8 (2.3)	27.3 (4.8)	24.8 (3.2)
	30-39	30.7 (2.5)	38.5 (3.5)	32.2 (4.6)	34.5 (2.2)	32.9 (3.4)	34.9 (2.5)
	40-49	21.2 (2.1)	21.3 (3.3)	19.1 (5.5)	22.0 (1.6)	21.0 (2.2)	21.5 (2.9)
	≥ 50	19.4 (2.7)	18.1 (4.1)	13.5 (5.5)	20.7 (2.5)	18.9 (2.5)	18.9 (3.9)

^ Percentages may not total to 100 due to rounding. (back) ^^ Other out-of-home placement setting omitted. (back) ^^^ Potential range is 0-29. (back) a Children in rural PSUs are significantly more likely than children in urban PSUs to be White (X2=17.8, p<.001). (back) b Children in rural PSUs are significantly less likely than children in urban PSUs to be Hispanic (X2=14.6, p<.001). (back) c Child welfare workers of children in rural PSUs have significantly shorter lengths of service than child welfare workers in urban PSUs (t=3.2, p<.01). (back) d Child welfare workers of children in rural PSUs are significantly more likely than those in urban PSUs to be White (X2=13.8, p<.001). (back) e Child welfare workers of children in rural PSUs are significantly less likely than those in urban PSUs to be Hispanic (X2=11.8, p<.001). (back)

In terms of the child’s race/ethnicity, very similar patterns emerge, with significantly larger proportions of White children (p<.001) and significantly smaller proportions of Hispanic children (p<.001) seen in rural PSUs. On average, the proportion of White children in rural PSUs is about twice that of urban PSUs. Hispanic children account for approximately one-quarter to one-third of the children in urban PSUs, while their proportions are in the single digits in rural PSUs. A trend indicates that rural PSUs also have smaller proportions of African American children (X2=5.2, p=.03).

Finally, trends exist indicating that children in rural PSUs may be more likely than their counterparts in urban PSUs to have borderline or clinical CBCL scores (X2=6.2, p=.02). This finding suggests that the smaller county child welfare systems are more likely to fulfill a role in the mental health system—by more often accepting children with mental health problems—than is true in more urban counties.

9.3.1 Summary of PSU (County) Characteristics and Service Delivery

Based on the above analyses, the strongest associations between the PSU characteristics and child welfare case characteristics are with regard to the urbanicity of the PSU and the race/ethnicity of the agency employees, as well as the race/ethnicity of the children that come to the attention of the child welfare agency. Specifically, rural PSUs do not appear to be as racially and ethnically diverse as their urban counterparts, as both their child welfare workers and clientele are mostly White and include few Hispanics. Child welfare workers in rural communities also appear to have less experience as child welfare workers than the child welfare workers in urban communities. Child setting, number of risks present at time of placement, proportion of children with a borderline or clinical CBCL score, and the child welfare worker age and education level do not appear to be tied to PSU characteristics. In addition, the proportion of children in poverty, as well as the administration of the child welfare agency, appears to have little relationship to the aspects of the functioning of the child welfare agency.

9.4 Conclusions

NSCAW is the first study to provide in-depth information about the physical, mental, and developmental needs of and service use by children involved with CWS nationally. Learning about the needs and service use patterns of the children is essential to achieving the goal of improving the safety, permanency, and well-being of this vulnerable group of children by providing valuable information about areas of need and, in particular, unmet need. Chapter 5 provided compelling evidence that children involved with CWS have elevated needs for mental health and special education services. The finding that the service needs of children remaining at home are comparable to those of children placed in out-of-home care is a new finding that indicates that even when children are not taken into custody by CWS, they should be given comparable attention with regard to health and education service provision as children who are removed from their homes. Although the findings here show that children involved with CWS are more likely than those in the general population to receive such services, future analyses are needed to better clarify the extent to which needs are being met and the factors that facilitate children’s access to necessary services.

Child welfare workers are generally older than 30 and reflect a diversity of ages. Although over one-third of child welfare workers are between 30 and 39 years, over one-quarter are younger than 30, with one-fifth between 40 and 49, and another fifth aged 50 years or older. That 40% of child welfare workers are at least 40 years of age is a positive finding, indicating dedication to the profession and a certain level of continuity for the children and families involved with CWS.

The race/ethnicity of child welfare workers also reflects considerable diversity, as almost two-thirds are White/non-Hispanic, about one-fifth are African American/non-Hispanic, and 9% are Hispanic. However, these proportions are not in keeping with the greater representation of African American and Hispanic children and families involved with CWS.

Although almost all child welfare workers have a college degree, only 12% have an MSW (although some of the 14% with a doctorate almost certainly have an MSW, we cannot estimate the size of this group). The benefits of child welfare workers having MSWs, as well as the reasons why most do not, should be examined further. Because we do not have information about all the child welfare workers involved with each case, this study does not allow for an analysis of the contribution of advanced training to case outcomes. We do have information, however, from the child welfare worker interviews about the agency climate and organizational characteristics—this information could be used, in future analyses, to better understand their contributions to the achievement of child welfare goals.

Although rural PSUs were shown in this analysis to have child welfare workers who are less racially and ethnically diverse than their urban counterparts, this tends to be consistent with the general racial and ethnic makeup of the PSU overall. That child welfare workers in rural PSUs have shorter mean lengths of service as child welfare workers than the child welfare workers in urban PSUs merits further examination to determine why workers in these communities may have higher turnover and thus provide less continuity for the children and families involved with CWS. Although none of the other PSU or service delivery characteristics showed significant associations with each other, as previously mentioned, the variables included in these analyses were limited to a few. More in-depth investigation utilizing other variables could be useful in determining how local conditions may otherwise affect service delivery.

The key findings for service needs and receipt includes the following:

Children’s Problems and Services

Physical Health

• About one-quarter of caregivers report that their child has a chronic health problem.

• More than one in five (21%) children in out-of-home care have visited the emergency room for an illness or injury since the start of the CWS investigation, and 37% of children remaining at home have been to the emergency room in the past year.

Mental Health

• 11% of children involved with CWS are using outpatient mental health services.

• The use of inpatient mental health services is high—7% of children in their homes of origin have used inpatient services in their lifetime, and 13% of children in out-of-home care have used these services since the CWS investigation date.

• Children in group care settings are more likely to use mental health services than those in kinship or foster care.

• Need for mental health services is related to use, as children with CBCL scores in the borderline or clinical range are more likely to be receiving services than those with lower scores.

Special Education

• Almost 20% of children have an IEP/IFSP, and 15% are currently receiving special education services.

• About one out of every five caregivers of children who had not been tested for special education needs reports that his or her child needs this service.

• The most common categories of special education need are for specific learning disability (10%), speech or language impairment (7%), and emotional disturbance (7%), all of which are much higher than the national rates for children in the U.S.

• Only 21% of children with at least one clinical score are currently receiving special education services.

Child Welfare Worker Characteristics

• Overall, the largest proportions of child welfare workers are between 30 and 39 years of age, White, and hold a bachelor’s degree.

• Child welfare workers have an average of 7 years of experience. Race/ethnicity and education level are not significantly associated with length of experience.

• White children are two to three times more likely than Hispanic and African American children to have a child welfare worker of the same race/ethnicity.

• The highest education level for the majority of child welfare workers is a bachelor’s degree (71%), although over one-quarter (26%) have a master’s degree or doctorate.

PSU (County) Characteristics and Service Delivery

• The child welfare workers in rural PSUs and the children with whom they work are significantly more likely than those in urban PSUs to be White and significantly less likely to be Hispanic.

• Child welfare workers in rural PSUs have significantly shorter lengths of service as child welfare workers than the child welfare workers in urban PSUs.

• Child setting, the number of risks present at time of placement, the proportion of children with a borderline or clinical CBCL score, and the child welfare worker’s age and education level do not differ significantly by agency administration, urbanicity of PSU, or poverty level of PSU.

• From the perspective of the PSU characteristics, the poverty level of the PSU and agency administration do not have significant associations with the service delivery variables examined.

---

 

Table of Contents | Previous | Next