Home > Statistics & Research > Child Maltreatment Report 2000 > Chapter Seven

 

Chapter 7
Additional Research Based on NCANDS
and State Administrative Data
Child Maltreatment 2000

In this chapter, several examples of additional analyses that examine child maltreatment are discussed.

REPORTS RESEARCH

The goal of the National Study of CPS Systems and Reform Efforts is to describe the overall status of child protective services (CPS) systems and of reform efforts that are underway. The study is funded by the Children's Bureau and managed by the Office of the Assistant Secretary for Planning and Evaluation, both within the U.S. Department of Health and Human Services. The three main areas of the study are the State policies that define CPS functions and specify how these functions are carried out; local CPS agency organization and the practices that implement CPS functions; and innovative reform efforts that seek to restructure, redefine, or reformulate the purposes and functions of CPS.

As a part of the study, National Child Abuse and Neglect Data System (NCANDS) data will be used in conjunction with local agency-level survey data. The purpose of the analysis will be to explore the relationship of organizational features and practices on the age, race, and sex of rates of children reported and the impact of these local agency characteristics on rates of reporting, victimization, postinvestigation service provision, types of maltreatment, and recurrence. County-level data from the NCANDS will be used to support these analyses. A report will be produced to discuss commonalties and differences in how CPS work is carried out and thereby provide a comprehensive picture of the CPS system nationwide.

For further information about the National Study of CPS Systems and Reform Efforts, contact:
Laura Radel
Office of the Assistant Secretary for Planning and Evaluation
Department of Health and Human Services
200 Independence Avenue SW, Room 450-G
Washington, DC 20201
Laura.Radel@hhs.gov

In 1998, the Children's Bureau funded three studies to increase the knowledge on "Unsubstantiated" reports. One study The Dynamics of Unsubstantiated Reports: A Multi-State Study conducted by the American Humane Association in collaboration with the American Public Human Services Association and Walter R. McDonald & Associates, Inc., used data from the NCANDS to test several different hypotheses.¹ Some of the key findings are:

• Differences in rates of unsubstantiation are due in part to differences in State policy. The inclusion of explicit disposition options that address uncertainty at the end of the investigation reduces the proportion of reports that are unsubstantiated. Higher levels of evidence required to substantiate a report were associated with higher rates of unsubstantiation.
• There is some evidence that personal and organizational factors influence disposition decision-making even in the presence of key case factors.
• Decisionmaker characteristics—The study suggests that there may be two independent types of worker characteristics—proclivity and contextual. Proclivity is the worker's tendency to lean toward substantiating an ambiguous case, or to want to err on the side of child safety. By contrast, the contextual decision-maker characteristic appears to be much more complex. Thus, experience as a caseworker, assessment of one's own casework skills, positive views of policy, and supportive relationships with co-workers tend to increase the level of unsubstantiation. However, other organizationally-based contextual characteristics-increased workload, and bureaucratic distractions interact with these other contextual characteristics such that when they are present, the level of of unsubstantiation is lower.
• Increases in worker-perceived supervisory support are associated with lower levels of substantiation.

For further information about The Dynamics of Unsubstantiated Reports: A Multi-State Study, contact:
American Humane Association
63 Inverness Drive East
Englewood, CO 80112-5117
303-792-9900
or
John D. Fluke, Ph.D.
Walter R. McDonald & Associates, Inc.
15884 East Crestridge Circle
Aurora, CO 80015
jfluke@wrma.com

Another study, Recidivism in Child Protective Services Among Substantiated and Unsubstantiated Cases, addressed the rates of recidivism of substantiated and unsubstantiated reports.² Drake, et al., from the George Warren Brown School of Social Work, used the administrative data of the Missouri Department of Social Services. Subjects were followed for 4.5 years. The main findings included the following:

• Children with initial "Substantiated" dispositions were slightly more likely than children with initial "Unsubstantiated" dispositions to have another report if the initial report concerned neglect, emotional maltreatment, or "Other" abuse.
• A statistically significant difference (p<0.003) was found between the percentages of "Substantiated" reports compared to "Unsubstantiated" reports that had a subsequent "Substantiated" report. (This did not hold true for victims of sexual, physical, or "mixed" abuse.)
• The risk of recidivism did not differ between "Substantiated" reports that received services and "Unsubstantiated" reports.
• Recidivism was most likely to be associated with neglect.

For further information about the Recidivism in Child Protective Services Among Substantiated and Unsubstantiated Cases study, contact:
Brett Drake, Ph.D.
George Warren Brown School of Social Work
Washington University
Campus Box 1196
One Brookings Drive
St. Louis, MO 63113
Brettd@gwbssw.wustl.edu

Return to Table of Contents

VICTIMS RESEARCH

In the Child Abuse and Neglect Among American Indian/Alaska Native Children: An Analysis of Existing Data study funded by Casey Family Programs, data from the Detailed Case Data Component (DCDC) were used to compare White and American Indian victims of abuse or neglect.³ Individual case data from 1995 to 1999 were combined into one file and sorted to select only the first case for each child. Each American Indian child was matched to a White child by age, gender, State, year of incident, and Hispanic ethnicity. Hispanic ethnicity was found to be a confounding variable, and all Hispanic cases were dropped from the final analysis, which included 9,031 American Indian and 9,027 White children. Statistically significant differences between these groups were found on the following variables:

• American Indian children were more likely than White children to have foster care services provided and to have a juvenile court petition filed.
• American Indian children were more likely than White children to be victims of neglect and less likely to be victims of physical abuse or sexual abuse.
• Both the American Indian children and the American Indian caretakers were more likely than White children and White caretakers to have problems with alcohol.
• There was a higher likelihood of violence between caretakers of an American Indian child than between caretakers of a White child.
• American Indian families were more likely than White families to receive public assistance.

For further information about the Child Abuse and Neglect Among American Indian/Alaska Native Children: An Analysis of Existing Data study, contact:
Kathleen A. Earle, Ph.D.
Institute for Child and Family Policy
Muskie School of Public Service
University of Southern Maine
400 Congress Street
Portland, ME 04112-5010
kearle@usm.maine.edu

Another study, the Occurrence and Recurrence of Physical Abuse in Young Children: A Secondary Analysis of NCANDS undertook to examine the occurrence and recurrence of physical abuse in young children. Child maltreatment information for 1995-1999 from the Detailed Case Data Component were examined. The first occurrence of physical abuse was compared to other types of maltreatment in children younger than 3 years, and the recurrence of physical abuse after a first occurrence before the child's first birthday was compared to children with no recurrence or with recurrence of other forms of maltreatment. Critical findings of the study included:

• Among 577,655 reports concerning children younger than 3 years old, there were 30,202 confirmed physical abuse reports in 16 States during 1995-1999.
• Reports coming from medical sources (Odds Ratio=5.2), for male children (OR=1.1), children with drug exposure (OR=5.2), behavior problems (OR=1.6), in families with drug problems (OR=1.6), or other violence (OR=1.7) had increased risk for physical abuse occurrence (p<0.01).
• Older children (OR=0.70) and those with other medical (OR=0.54) or family medical problems (OR=0.71), inadequate housing (OR=0.31), or who received public assistance (OR=0.70) had less risk for physical abuse.
• The overall incidence of first confirmed reports for physical abuse was 2.5 cases per 1,000 children younger than 3 years. Physical abuse recurrence occurred in less than 1 percent of children.
• Child learning disability (OR=20.9) and prior provision of family support services (OR=3.20) were associated with increased recurrence before age 3 (p<0.05).

Only eight States contributed data for at least three consecutive years to the DCDC during 1995-1999, and substantial data are missing. The subtype of physical abuse is also not specified in the DCDC. Different types of injuries resulting from physical abuse are postulated to have different rates of occurrence and recurrence, and the impact of these specific injuries cannot be assessed. In addition, the archived DCDC data do not include confidential child data on child fatalities, an important outcome of physical abuse. Future research will benefit from improved consistency of State reporting and additional information about the subtypes of abuse, including fatalities.

For further information about the Occurrence and Recurrence of Physical Abuse in Young Children: A Secondary Analysis of NCANDS study, contact:
Vincent John Palusci, M.D.
Director
Child Protection Team
DeVos Children's Hospital
100 Michigan Street, NE
MC-98
Grand Rapids, MI 49503
Vincent.Palusci@Spectrum-Health.org

The Evidence Base for Child Welfare Policy study conducted in conjunction with The Chapin Hall Center for Children at the University of Chicago examined the patterns of reports to and response by CPS agencies with regards to the age of child victims. The key findings based on 1999 data include those listed below:

• The youngest children alleged to have been maltreated are the most likely to be the subject of a "Substantiated" or "Indicated" report and found to be victims of maltreatment.
• The highest rates of victimization were for children less than 1 year of age.
• Maltreatment of the very young was most likely to have been reported by medical and health professionals.
• Of victims removed from the home as a consequence of a report alleging maltreatment, the youngest children were the most likely to be placed in foster care.
• The recurrence of maltreatment was lowest for the very young-those less than 1 year old.

For further information about The Evidence Base for Child Welfare Policy study, contact:
Ying-Ying T. Yuan, Ph.D.
Walter R. McDonald & Associates, Inc.
12300 Twinbrook Parkway, Suite 310
Rockville, MD 20852
yyuan@wrma.com

The Child Welfare Outcomes 1999: Annual Report examined certain features of the child welfare system in conjunction with demographic conditions.⁴ The findings included the following:

• The correlation between the percentages of child victims experiencing neglect and of children living in poverty was not significant.
• State variation in recurrence was not correlated with the percentage of child victims who were neglected. Recurrence was significantly correlated with the rate of child victims per 1,000 children in the population.

For further information about the Child Welfare Outcomes 1999: Annual Report, contact:
Marianne Rufty
Children's Bureau
330 C Street SW
Washington, DC 20447
mrufty@acf.dhhs.gov

Return to Table of Contents

SERVICES RESEARCH

The National Survey of Child and Adolescent Well-Being (NSCAW) study, will follow a nationally representative sample of children who come into contact with the child welfare system, assessing their social, emotional, cognitive, and functional status, as well as services needed and provided for children and their families. More than 6,000 children in 36 States are included in the study. The sample includes both children who are receiving services from the child welfare system, and children who have been the subject of investigations by child protective services, but who have not come into the system.

The study includes direct assessments and interviews of children, their primary caregivers (either their parents or substitute caregivers, such as foster parents), and their caseworkers. Information from teachers is obtained, via questionnaires, for school-aged children. The sampling was dependent on being able to estimate the size of the child welfare population in the U.S. at the county level. To obtain these estimates for the approximately 3,000 counties in the U.S., the NSCAW statisticians used NCANDS data along with Census data, natality statistics, and State child welfare statistics, where available. The NCANDS data provided State and county level statistics on reporting and substantiation.

For further information about the National Survey of Child and Adolescent Well-Being (NSCAW) study, contact:
Mary Bruce Webb
Office of Policy, Research and Evaluation
ACF/DHHS
370 L'Enfant Promenade SW
Washington, DC 20447
mbwebb@acf.dhhs.gov

Return to Table of Contents

SUGGESTIONS FOR FUTURE RESEARCH

Some topics of interest for future research or program planning and review are briefly discussed below.

Characteristics of Victims

While it is hypothesized that certain children are more at risk for maltreatment and recurrence, there remains additional research to be conducted. The NCANDS data include items on factors such as financial assistance, child and adult alcohol and drug abuse, and racial and ethnic background of children and perpetrators. These factors can be examined further in terms of disposition rates and recurrence rates.

Services

The linkages between service provision and the types of reports that are investigated and assessed continue to be an issue for research. Characteristics of children who receive services as a consequence of reports alleging abuse or neglect should be examined. Of special interest, perhaps, are the characteristics of children in reports that receive "Unsubstantiated" dispositions as related to service provision. Starting with data for 2000, some States have included service data on "Unsubstantiated" reports. This new data will help to address this question.

Perpetrators

The characteristics of perpetrators, including risk factors associated with the family or the perpetrator, can be further studied to determine what, if any, characteristics relate to multiple maltreatments by a perpetrator.

Researchers interested in pursuing these or other topics based on the NCANDS data should contact the National Data Archive on Child Abuse and Neglect, which is maintained by the Family Life Development Center, New York State College of Human Ecology, Cornell University, under a cooperative agreement with the Children's Bureau. The Archive provides technical assistance and training on the use of child welfare data. It maintains case-level data and the multiyear data set of NCANDS. Intensive training is provided at a 1-week Summer Institute in June of each year. Contact information is provided on the inside cover of this report.

Notes

¹ Fluke, J., Parry, C., Shapiro, P., Hollinshead, D., Bollenbacher, V., Baumann, D., Davis-Brown, K., (2001). The Dynamics of Unsubstantiated Reports: A Multi-State Study. Englewood, CO: The American Humane Association. Back
² Drake, B., Jonson-Reid, M., Way, I., Chung, S., (2001). Recidivism in Child Protective Services Among Substantiated and Unsubstantiated Cases. Unpublished Final Report, Grant #90CA159101. Back
³ Earle, K.A., & Cross, A. (2001) Child Abuse and Neglect Among American Indian/Alaska Native Children: An Analysis of Existing Data. Seattle, WA: Casey Family Programs. Back
⁴ U.S. Department of Health and Human Services, Children's Bureau. Child Welfare Outcomes 1999: Annual Report (Washington, DC: U.S. Government Printing Office, 2002). Back

 

Return to Table of Contents