urban institute nonprofit social and economic policy research

The Decision to Investigate

Understanding State Child Welfare Screening Policies and Practices

Read complete document: PDF


PrintPrint this page
Share:
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: May 01, 2000
Released online: May 01, 2000

Number A-38 in Series, "New Federalism: Issues and Options for States"

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Note: Those wishing to print this report may find it easier to use the PDF Version.


About the Series

This series is a product of Assessing the New Federalism, a multiyear project to monitor and assess the devolution of social programs from the federal to the state and local levels. Alan Weil is the project director. The project analyzes changes in income support, social services, and health programs. In collaboration with Child Trends, the project studies child and family well-being.

The project has received funding from The Annie E. Casey Foundation, the W.K. Kellogg Foundation, The Robert Wood Johnson Foundation, The Henry J. Kaiser Family Foundation, The Ford Foundation, The John D. and Catherine T. MacArthur Foundation, the Charles Stewart Mott Foundation, The David and Lucile Packard Foundation, The McKnight Foundation, The Commonwealth Fund, the Stuart Foundation, the Weingart Foundation, The Fund for New Jersey, The Lynde and Harry Bradley Foundation, the Joyce Foundation, and The Rockefeller Foundation.

This series is dedicated to the memory of Steven D. Gold, who was codirector of Assessing the New Federalism until his death in August 1996.


State child welfare agencies investigated reports of abuse and neglect involving more than 3 million children in 1996. Many more children, however, were and continue to be initially referred to child welfare agencies for child maltreatment1 allegations. State agencies assess each such referral to determine how best to respond and if the referral merits investigation. Child welfare agencies receive many referrals that clearly do not constitute abuse or neglect or that provide so little information that the agency has no way to locate the alleged perpetrator and conduct an investigation. Referrals are screened, and such screening serves a gate-keeping function, allowing child welfare agencies to determine which families are investigated and, by extension, which families may eventually receive more intensive agency services. In this process, some child maltreatment allegations are screened out before an in-person investigation is conducted.

There has been considerable debate about the growing number of reports investigated by child welfare workers and the declining proportion of these reports that are substantiated. Some have argued that these numbers reflect an increase in inappropriate referrals; others suggest that child welfare agencies have raised the threshold of what constitutes maltreatment to compensate for inadequate agency budgets (Besharov 1998; Downing, Wells, and Fluke 1990). Although virtually every state screens child maltreatment allegations prior to investigation, little is known about the number of referrals screened out before an investigation is conducted or about the criteria used to screen these referrals.

Screening in the Context of Federal Welfare Reform

Despite the lack of explicit guidelines, states screen out a substantial portion of initial child maltreatment allegations.

In the wake of landmark legislation that reshaped the nation's welfare system, public officials and researchers have voiced concerns about its potential effects. Competing theories have emerged on how federal welfare reform could alter the demand for child welfare services (Allen 1996; Child Welfare League of America 1998; Courtney 1997; Knitzer and Bernard 1997). Many of these theories are tied to how well families fare economically under the new rules. Since studies have shown that poverty status is correlated with higher reporting of child abuse and neglect (Pelton 1978; Sedlak and Broadhurst 1996), some argue that if welfare reform is successful in moving recipients into the labor force and increasing family income, the family stresses caused by economic conditions will lessen and fewer children may be reported for child maltreatment. Conversely, if families' economic well-being worsens, more children may be reported. In addition, policymakers have raised the concern that if parents cannot find adequate child care, they may leave young children inadequately supervised in order to meet work requirements.

Any attempt to measure the effect of welfare reform on child maltreatment presents sizable challenges. It is extremely difficult to track changes in the incidence of abuse and neglect, since many incidents are never reported to child welfare agencies. Examining only changes in the number of child maltreatment reports investigated ignores the influence that changes in screening practices may have. An alternative measure, the number of referrals for alleged maltreatment, is not affected by screening practices but may be subject to broader social trends and changing policies with respect to reporting. Regardless of the measure used, significant methodological challenges emerge with any attempt to link changes in child maltreatment rates to particular social policies.

Nonetheless, given the potential for great variation across jurisdictions in child maltreatment screening practices, there is value in monitoring changes in referrals. In order to measure the magnitude and direction of the change in referrals (if any) experienced by child welfare agencies following federal welfare reform, the Urban Institute conducted a baseline survey of state child welfare agencies in 1997. This survey collected data on both the nature of states' formal screening policies and the number of initial allegations of child maltreatment that states screened out.2  Before this study, screening data were not collected systematically across all states.3  The results of this study are presented here.4

General Flow of Child Maltreatment Allegations

Child welfare systems are complex. Allegations of child maltreatment made to state child welfare agencies may result in a variety of outcomes (see figure 1). Most reports of child maltreatment find their way to a local child welfare agency via state telephone hotlines. All states have hotlines staffed by child welfare intake workers who record allegations of child maltreatment. In some states, workers also check agency records to determine if reported families are currently involved in open child welfare cases or have a history of substantiated child maltreatment reports. Workers then decide whether to refer the new reports for investigation.


The Mechanics of the State Screening Practices

Virtually all states screen child maltreatment referrals. (Only the District of Columbia reported that currently all child maltreatment allegations referred to child welfare are investigated.5) Despite having screening policies, few states have explicit guidelines delineating the types of reports that should be screened out; even fewer states use formal instruments to guide the screening process (Downing et al. 1990). Therefore, the actual process varies widely by state and even by counties within states.

In most states, the responsibility for making a screening decision rests with one worker. However, in some states more than one staff member takes part in a multiple-review process for each initial allegation. For example, a hotline worker may initially screen out reports that are later reviewed by a supervisor who may elect to investigate a case that the first reviewer screened out or reject a recommendation to investigate.

Of the 40 states that provided information, the majority (26) had single-review screening processes, while one-third (14) used multiple-review screening. About half of the states (12 out of 26) that relied on only one worker to make screening decisions used a supervisor instead of a hotline worker, an investigator, or a social worker to make these decisions. In states with multiple-review screening, the most common combination (for 8 out of 14) was an intake worker making an initial screening decision that was reviewed by a supervisor.

The survey also found that training requirements for screeners varied by state. Some states had strict educational requirements for their screeners; others did not. In some states, screeners were required to have several years of experience in child protection, while other states had no such requirements. Workers who answered state hotlines made screening decisions in some states, while in others these workers simply recorded information and left screening decisions to intake supervisors.

Percentage of Initial Allegations Screened Out

The survey attempted to quantify the number of initial child maltreatment allegations made to state agencies in 1996.6  The results show that state child welfare agencies screened out and closed without investigation about one-third of the initial referrals they received. Based on the 31 states that could provide complete data, states screened out a median of 36 percent of all initial allegations of child abuse and neglect. However, the percentage of initial allegations screened out varied widely—from a low of 5 percent (New Jersey) to a high of 78 percent (Vermont). A significant number of states had screened-out rates substantially over or under the national median. While seven states screened out at least half of their initial allegations, five states screened out 10 percent or less (table 1).


Table 1 State Screening Rates, 1996
States
Single or Multiple-Review Screening Process
Percentage of Reports Screened Out before Investigation
Percentage of Investigated Reports Substantiated
Percentage of All Reports Substantiated
District of Columbia
None
0.0
51.4
51.4
New Jersey
Single
5.0
28.3
26.9
California
Multiple
5.4
N/A
N/A
Arizona
Single
5.5
55.0
52.0
New Mexico
Single
10.0
34.2
30.8
Delaware
Single
11.0
33.9
30.2
Connecticut
Single
14.5
44.1
37.7
Kentucky
Single
15.0
40.1
34.1
Idaho
Single
25.6
33.5
24.9
Georgia
Single
26.9
55.8
40.8
Texas
Single
28.0
36.8
26.5
Oklahoma
Multiple
31.2
34.8
23.9
Arkansas
Multiple
33.0
36.4
24.4
Alaska
Single
33.4
93.4
62.2
Wyoming
Single
35.1
N/A
N/A
Rhode Island
Single
35.8
30.0
19.3
Colorado
Single
36.2
16.5
10.5
Iowa
Single
36.6
33.4
21.2
Oregon
Multiple
38.4
59.7
36.7
Maine
Single
38.7
68.4
41.9
Maryland
Multiple
40.6
52.9
31.4
Massachusetts
Single
40.8
50.0
29.6
New York
Multiple
42.7
32.3
18.5
Missouri
Single
45.0
34.3
18.9
Florida
Single
49.6
45.6
23.0
Michigan
Single
52.7
15.9
7.5
New Hampshire
Single
53.8
13.6
6.3
Washington
Multiple
56.3
69.4
30.3
Nevada
Single
71.1
39.2
11.3
Illinois
Single
75.8
35.5
8.6
Vermont
Single
78.1
45.1
9.9
Median
N/A
35.8
38.0
26.7
Sources: The Urban Institute Child Welfare Survey; reports from state child welfare officials. Hawaii and Virginia did not respond to the survey. All other missing states (18) could not provide the necessary data to calculate screening rates; these states are excluded from the table.

Aggregate state screening data may mask divergent county trends. For example, in 1994 California reported that statewide, 32 percent of all initial allegations of child maltreatment were screened out before investigation. The screening rates ranged from a low of 19 percent in Los Angeles County to a high of 55 percent in Contra Costa County (California Legislative Analyst's Office 1996). California is not alone; other states reported similar cross-county variation in their screening rates (Geen and Tumlin 1999). Therefore, the diversity of screening practices is likely much greater than the state variation revealed by the survey.

Changes in State Screening Policies

Prior research has shown that states often use stricter screening standards during times of high demand or staff reductions. For example, the Child Welfare League of America reported that 45 percent of state administrators surveyed in 1995 said that state triage policies dismissed some reports that would have been investigated five years earlier (Curtis et al. 1995).7  States were asked to report any changes in their official screening policies over the previous two years as well as changes in practices resulting in the screening out of certain types of allegations that previously would have been investigated. Many state agencies reported that they had recently tightened the criteria by which they judge whether a referral should be investigated. While only one state—Connecticut— reported that its official state policy changed to exclude certain types of reports from investigation,8 many states reported unofficial changes in practice that produced the same result. Only one state—Indiana— reported an expansion in the types of cases investigated.9  Several states, however, reported expansions in who can be considered a perpetrator of child maltreatment. Prior to these expansions, allegations involving such perpetrators would have been screened out before investigation. For example, several states reported that they now investigate allegations that a parent's live-in partner has maltreated his or her child.

The Relationship between Screening and Substantiation

How do the wide range of state screening rates and practices affect states' caseloads and substantiation rates? Over the past decade, state child welfare agencies have been criticized for investigating more allegations of child maltreatment but substantiating fewer cases. In 1996, for example, states substantiated on average two cases out of every five referred for investigation (DHHS 1998).10  Since one of the primary goals of state screening policies is to reduce the number of inappropriate investigations, it seemed that states with higher screening percentages at intake might have higher substantiation rates. No such pattern was found (figure 211).  For example, even outlying states that each screen out over 70 percent of all referrals (Illinois and Nevada) substantiated less than 40 percent of investigated reports—rates similar to or lower than the national median. Meanwhile, two states with small or nonexistent screening rates (Arizona and the District of Columbia) substantiated investigated reports at rates over 50 percent—well above the national median.


What explains the apparent lack of relationship between aggregate screening and substantiation rates? While there is no evidence to suggest that the proportion of inappropriate referrals varies greatly by state, it does appear that some states maintain a higher threshold of not only what they will investigate but also what they will substantiate. Public officials have reported that screening decisions are influenced by agency capacity, and it seems reasonable that substantiation decisions are similarly affected. Thus, severely overloaded states may screen out a large percentage of referrals and substantiate few investigated reports in an attempt to mitigate these overwhelming demands.

The Benefits and Risks of Screening

Screening can benefit child welfare agencies as well as families reported for abuse and neglect. Child welfare agencies can save valuable staff time and agency funds. While data are limited, experts agree that children and families were reported to state child welfare agencies in record numbers over the past decade (American Public Welfare Association 1996; DHHS 1998; English 1998; Sedlak and Broadhurst 1996). In addition, child advocates, researchers, and state officials alike have reported that funding for child welfare services has not kept pace with increases in demand. (Courtney 1998; Meyers 1994; Schorr 1997). Screening enables child welfare agencies to respond to this increased work burden in a prioritized fashion. In theory, this type of screening should reduce caseloads to a manageable number so that investigators have time to conduct thorough investigations of accepted cases and make the best use of limited resources.

Screening also frees up agency resources for substantiated cases of child maltreatment. Across states, such cases often receive no ongoing services. Based on data reported by 31 states to the National Child Abuse and Neglect Data System, 40 percent of identified victims of abuse and neglect in 1997 received no services following a child protective services investigation (DHHS 1999). These data, coupled with screening data, show that only a small fraction of all families initially reported to child welfare agencies receive services following an investigation. For example, in Florida, only 21 percent of all initial allegations in 1996 resulted in children or families receiving services after an investigation was conducted.12  Similarly, a study conducted in Alameda County, California, found that in 1993-1994 only 12 percent of families referred to child welfare received services following an investigation (Karski, Gilbert, and Frame 1996). Viewed from this perspective, states that increase their screened-out rates may be able to reserve precious funds to better serve families with substantiated allegations.

For inappropriate reports of maltreatment, screening can also reduce unnecessary intrusion into families' private lives. Many argue that current child maltreatment statutes are vague and overly broad and, coupled with expansions of mandatory reporting laws, have led to an increase in the number of families investigated but a decrease in the percentage of substantiated reports (Besharov 1998). These critics argue that the declining percentage of substantiated reports is a result of investigation of inappropriate referrals, which causes an unnecessary intrusion into families' privacy. Screening is a mechanism state child welfare agencies have been using to reduce such intrusions and to ensure that only appropriate cases are investigated. Cases that are screened out generally involve no contact with the reported family and therefore could represent effective triaging without subjecting families to intrusion or shame.

While effective screening of maltreatment reports can be beneficial, there are also risks. Child welfare agencies may miss cases of true maltreatment, subjecting children to further abuse or neglect. Unfortunately, there is little expert consensus about the causes of child maltreatment or how best to predict which children are at risk of future abuse. For example, research on the effectiveness of risk assessment models used by child welfare agencies indicates that these models do not reliably predict future harm to children (Wald and Woolverton 1990). Furthermore, screening occurs at the initial intake phase, where it is inherently most difficult for child welfare staff to render accurate decisions about maltreatment incidents, since these decisions are often based on skeletal information and little or no contact with the reported family.

A related concern is the lack of information about what becomes of screened-out cases. How often are screened-out cases re-referred to child welfare agencies? And if these cases do return to the system, what percentage eventually result in a substantiated allegation of child maltreatment? The Alameda County study shows that the majority of cases (62 percent) that are initially screened out are re-reported to child welfare (Gilbert et al. 1997). Further research is needed to assess the frequency of re-reporting and eventual substantiation of allegations initially screened out by child welfare agencies. Such data are necessary to determine the effectiveness of current state screening practices.

Conclusions

Child protection investigators must often make difficult and highly subjective decisions in determining whether to substantiate a report of maltreatment once an investigation has been conducted. But there are sometimes other equally challenging decisions that child welfare staff must make before a case is investigated—decisions that, if made in error, put children's safety at risk. Unless a referral is clearly inappropriate, child welfare staff must balance the desire to use limited agency resources efficiently and minimize unnecessary intrusions into families' lives with the possibility of failing to investigate a report on a child truly in need of protection. Despite the lack of explicit screening guidelines, states screen out a substantial proportion of initial child maltreatment allegations. Research shows that since few states have explicit guidelines, workers use their own discretion and biases when making screening decisions (Gilbert et al. 1997). Moreover, since screening decisions may be influenced by agency demand and capacity, informal criteria for investigating a child maltreatment referral may fluctuate over time. This fluidity and cross-county variation may send mixed messages to potential reporters about what constitutes abusive or neglectful behavior.

In determining the effects of welfare reform or any other social policy change on child maltreatment, it is essential to measure changes in how child welfare agencies respond to initial allegations. If, for example, the number of children reported to child welfare agencies increases dramatically, but in response agencies simply screen out a greater proportion of reports, this increase will go unnoticed by conventional measures. Yet, to date, most policymakers and researchers have relied on data on the number of child maltreatment reports investigated or substantiated to assess changes in the demand for child welfare services. In fact, many states do not maintain or examine child maltreatment screening data. While the benefits, risks, and best practices for effective screening are still open to debate, there should be no doubt that policymakers and researchers must include screening data in any assessment of changes in child welfare caseloads.


Notes

1. In this brief, the term "child maltreatment" is used to refer collectively to incidents of child abuse and neglect.

2. The survey was mailed to state officials in all 50 states and the District of Columbia. Urban Institute staff had follow-up phone conversations with state officials to ensure state data were interpreted accurately.

3. The closest proxy for the number of initial allegations of child maltreatment reported to child welfare agencies is the number of reports of child abuse and neglect referred for investigation collected by the U.S. Department of Health and Human Services from the National Child Abuse and Neglect Data System (NCANDS). Beginning in 1998, The NCANDS collected data similar to that collected by the Urban Institute on the number of reports of child abuse and neglect screened out before investigation in its 1997 survey.

4. Two states, Hawaii and Virginia, did not respond to the Urban Institute Child Welfare Survey. Of the 49 states that responded, not all were able to provide all of the screening information requested on the survey. For the purposes of this brief, the District of Columbia is treated as a state.

5. However, the general receiver for the District's child welfare agency is currently in the process of implementing a formal screening process (Jones 1999). In addition, since Hawaii and Virginia did not complete the screening survey it is unknown whether these states have a formal screening process.

6. We worked with states to ensure that the counts of initial allegations were comparable. Specifically, we asked states to exclude crank calls, wrong numbers, and allegations that do not involve a child's caregiver. Some states also excluded allegations that, if true, would not meet state definitions of child abuse or neglect.

7. Several state officials interviewed by Urban Institute staff for a separate study in 1996 and 1997 also noted that they currently rejected certain types of allegations that were investigated in times of less demand. See Geen and Tumlin, (1999).

8. In Connecticut, an allegation that a child is "at risk of neglect" is no longer sufficient to warrant an investigation.

9. Indiana is now investigating allegations of statutory rape even when consent of the minor can be established and the minor is over 12 years old.

10. The national substantiation rate was calculated by counting reports that were "substantiated" and "indicated" under state criteria as substantiated reports.

11. While there is no clear relationship between state screening rates and substantiation rates, a pattern might emerge if one were able to disaggregate state screening and substantiation data by type of maltreatment. State data show that substantiation rates vary greatly by the three major types of child maltreatment—physical abuse, sexual abuse, and neglect. It is reasonable to expect that screening may be more effective for certain types of maltreatment and that aggregate screening data would mask these trends.

12. Data provided by the Florida Department of Children and Families.


References

Allen, Mary Lee. 1996. "The Implications of the Welfare Act for Child Protection." Bad News/Good News for Children: The Need to Look Ahead. Washington, D.C.: Children's Defense Fund.

American Public Welfare Association. 1996. Voluntary Cooperative Information Systems for Child Welfare. Washington, D.C.: American Public Welfare Association.

Besharov, Douglas J. 1998. "How We Can Better Protect Children from Abuse and Neglect." The Future of Children: Protecting Children from Abuse and Neglect 8 (1): 120-23.

California Legislative Analyst's Office. 1996. Child Abuse and Neglect in California. Sacramento: California Legislative Analyst's Office. January 10.

Child Welfare League of America. 1998. "Child Welfare League of America Decries Results of New Welfare Law: Predicts Foster Care Population Will Double." http//www.cwla.org/cwla/communic/resultsofnewwelfarelaw.html. March 23.

Courtney, Mark E. 1997. "Welfare Reform and Child Welfare Services." In Child Welfare in the Context of Welfare "Reform," Report V, edited by Sheila B. Kamerman and Alfred J. Kahn (1-35). New York: Columbia University School of Social Work.

———. 1998. "The Costs of Child Protection in the Context of Welfare Reform." The Future of Children: Protecting Children from Abuse and Neglect 8 (1): 88-103.

Curtis, Patrick A., Jennifer D. Boyd, Mary Liepold, and Michael Petit. 1995. Child Abuse and Neglect: A Look at the States-the CWLA Stat Book. Washington, D.C.: Child Welfare League of America.

DHHS. See U.S. Department of Health and Human Services.

Downing, Jane D., Susan J. Wells, and John Fluke. 1990. "Gatekeeping in Child Protective Services: A Survey of Screening Policies." Child Welfare 69 (4): 357-67.

English, Diana J. 1998. "The Extent and Consequences of Child Maltreatment." The Future of Children: Protecting Children from Abuse and Neglect (8) 1: 39-53.

Geen, Rob, and Karen C. Tumlin. 1999. State Efforts to Remake Child Welfare: Responses to New Challenges and Increased Scrutiny. Washington, D.C.: The Urban Institute. Assessing the New Federalism Occasional Paper No. 29.

Jones, Ernestine F. 1999. Child and Family Services Agency Strategic Plan: 1998-1999. Washington, D.C.: Child and Family Services Agency. September 4.

Karski, Ruth L., Neil Gilbert, and Laura Frame. 1996. The Emergency Response System: Screening and Assessment of Child Abuse Reports. Berkeley, Calif.: University of California at Berkeley School of Social Welfare.

Knitzer, Jane, and Stanley Bernard. 1997. The New Welfare Law and Vulnerable Families: Implications for Child Welfare/Child Protection Systems. Children and Welfare Reform Issue Brief 3. New York: National Center for Children in Poverty.

Meyers, J.E.B. 1994. The Backlash: Child Protection under Fire. Thousand Oaks, Calif.: Sage Publications.

Pelton, L.H. 1978. "Child Abuse and Neglect: The Myth of Classlessness." American Journal of Orthopsychiatry 48 (4): 608-17.

Sedlak, Andrea, and Diane Broadhurst. 1996. Third National Incidence Study of Child Abuse and Neglect: Final Report. Washington, D.C.: National Center on Child Abuse and Neglect (DHHS).

Schorr, L.B. 1997. Common Purpose: Strengthening Families and Neighborhoods to Rebuild America. New York: Anchor Books, Doubleday.

U.S. Department of Health and Human Services, Children's Bureau. 1998. Child Maltreatment 1996: Reports from the States to the National Child Abuse and Neglect Data System. Washington, D.C.: U.S. Government Printing Office.

———. 1999. Child Maltreatment 1997: Reports from the States to the National Child Abuse and Neglect Data System. Washington, D.C.: U.S. Government Printing Office.

Wald, Michael S., and Maria Woolverton. 1990. "Risk Assessment: The Emperor's New Clothes?" Child Welfare 69: 483-511.


About the Authors

Karen C. Tumlin was a research associate in the Urban Institute's Population Studies Center. Her career has concentrated on child welfare and immigrant policy issues. Ms. Tumlin is currently working on issues of family violence and immigration in Thailand as a part of the Luce Scholars Program.

Rob Geen is a senior research associate in the Urban Institute's Population Studies Center, specializing in child welfare and related child, youth, and family issues. He directed the Child Welfare Survey, the data from which are described in this brief, and is currently directing a national study assessing the effects of welfare reform on the child welfare system.



Topics/Tags: | Children and Youth | Poverty, Assets and Safety Net


Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page