Home > Decision-Making in Unsubstantiated Child Protective Services Cases: Synthesis of Recent Research > Decision-Making in Unsubstantiated Child Protective Services Cases: Synthesis of Recent Research: 2 - Findings

Decision-Making in Unsubstantiated Child Protective Services Cases: Synthesis of Recent Research
Grantee Lessons Learned

Author(s):  Children's Bureau (DHHS)

Year Published:  2003

Email Order (Free)  Print  (PDF 265 KB)  Share

Rate This    4.3/5, 3 Reviews

previous

You are in section: Table of Contents 1 - Introduction 2 - Findings 3 - Implications 4 - Recommendations for Future… 5 - Conclusion

next

2 - Findings

The research studies are complex and touch on many issues. The following section summarizes some of the key findings. In some cases, the studies' findings agree; occasionally they conflict. Taken together, these findings help develop a better understanding of the dynamics of unsubstantiated CPS cases. They do not, however, represent a national sample. Further exploration of these issues will be needed.

In this section, findings regarding factors that influence CPS decision-making are presented first. Then the researchers' findings regarding the impact of those decisions on outcomes for children are discussed, using re-referral as an indicator for outcomes. "Re-referral" indicates any situation in which a case (whether initially substantiated or not) returns to the system for a second or subsequent referral. In most cases in the research, re-referral has been identified from the child's perspective (often identified as "recurrence"). In some cases, re-referral has been identified from the perpetrator's perspective (often identified as "recidivism").

Figure 1: Research Studies on Unsubstantiated CPS Cases

2.1 Factors That Influence CPS Decision-Making

In general, the researchers agree that the decision-making process is complex, involving many layers and factors beyond the facts of the case itself. Factors found to influence CPS decision-making can be grouped into the following four categories:³

• Case factors are situations, events, or circumstances related to the child and family.
  
  
• Decision-maker factors include caseworker characteristics such as training, experience, beliefs, their interactions with each other, and their perceptions about the organization.
  
  
• Organizational factors are aspects of the CPS work environment such as the structure of the agency; its resources, tools, and training; workloads; and supervision.
  
  
• External factors include State laws and other policies that govern the CPS system.
  
  

2.1.1 Case factors

Characteristics of the case, child, and family were found to have the greatest influence on decision-making. These include many factors often used by CPS caseworkers to assess risk to the child, such as the child's age and development, the parent's ability to care for the child adequately, and the severity of the alleged incident.

One study examined how caseworkers' assessments of risk impacted their decisions regarding substantiation. These caseworkers used a risk assessment matrix that grouped 37 risk factors into 7 "domains."⁴ The domains included:

• Child characteristics (e.g., age of child; physical, mental, and social development).
  
  
• Severity of child abuse or neglect (e.g., dangerousness of the act, provision of basic needs, adequacy of supervision).
  
  
• Chronicity (frequency of occurrence).
  
  
• Caretaker characteristics (e.g., substance abuse, history of domestic violence, recognition of the problem).
  
  
• Caretaker/child relationship (attachment and bonding issues).
  
  
• Social and economic factors (e.g., employment status, social support systems).
  
  
• Perpetrator access to the child.
  
  

Table 1 summarizes how particular risk factors were found to increase the likelihood of a particular finding decision. It should be noted that the relative importance of particular risk factors to the finding decision varied in some cases according to maltreatment type. Some of the more interesting findings are discussed in detail below.

Multiple risk factors. When risk factors of significant severity were found in six to seven different domains, the likelihood of a substantiated finding was significantly increased. Unsubstantiated cases were more likely to have risk found in only one to five domains.⁹ However, if any risk was identified using the risk assessment matrix, the case was significantly less likely to be unsubstantiated than if no risk was identified.

Prior referrals. A finding of unsubstantiated was more likely if there was no history of child abuse or neglect referrals. In one study, cases with any prior referrals were found to be 21 percent less likely to be unsubstantiated.¹⁰ In a second study, when there was a history of 1 or 2 prior referrals, the investigation was more likely to be found inconclusive. (In Washington State, "inconclusive" is a third option, in addition to substantiated or unsubstantiated, indicating there is insufficient evidence for the caseworker to conclude either that the child has or has not been maltreated.) Three or more prior referrals resulted in a high likelihood the case would be substantiated.¹¹

Referral source. One study found that referrals from "community" sources (e.g., family, friends, neighbors) were more likely to be unsubstantiated,¹² while referrals from professionals were more likely to be substantiated.¹³ In the same study, interviews with caseworkers suggested the issue of referrer credibility was less influential than other factors in the decision-making process.¹⁴ In another study, cases referred by professionals were found to be 45 percent less likely to be unsubstantiated.¹⁵

Neglect type. In one study, caseworkers indicated they were least likely to substantiate referrals for neglect.¹⁶ Among types of neglect cases, allegations of medical neglect (defined as failure to obtain or maintain medical and dental services necessary to a child's continued health, welfare, and development) were found to result in the highest rate of substantiation, while allegations of lack of supervision resulted in the lowest. The researchers suggest this may be because medical neglect referrals often come from professional sources (which, as discussed above, are associated with an increased likelihood of substantiation). Meanwhile, referrals for lack of supervision often come from neighbors and relatives (community sources, which are associated with a decreased likelihood of substantiation).¹⁷

Age of the child. One study found the likelihood of unsubstantiation increased incrementally with the age of the child.¹⁸ In another study, children ages 0 to 2 and older than 13 were more likely to have their cases substantiated, while children ages 3 to 5 were more likely to have their cases unsubstantiated.¹⁹

Types of information used in decision-making. One study found that caseworkers emphasized different types of information in making each case decision (substantiated, inconclusive, or unsubstantiated). For example, issues of proof were emphasized in decisions to substantiate. Other factors—such as whether the child's basic needs were met, whether a caregiver offered a plausible explanation, and supervisory input—were more significant in inconclusive cases.

Findings regarding the decision not to substantiate a case were more complex. The types of information used varied according to the type of maltreatment alleged. In physical abuse cases, issues of proof and evidence (especially evidence of physical harm or medical evidence) were most important. In sexual abuse cases, caseworkers were more likely to rely on testimonial information and credibility issues in deciding not to substantiate. For physical neglect cases, observational information (such as condition of the home) was most significant.²⁰

Factors found to be insignificant. All other things being equal, no significant differences in rates of substantiation were found between children who were male or female, of an ethnic minority or majority, or from high- or low-income families.²¹

2.1.2 Decision-maker factors

Certain caseworker characteristics were found to impact decision-making. These include both individual characteristics (such as a worker's inherent tendency to decide one way or another in particular kinds of cases) and the context of the decision (including factors ranging from the caseworker's level of experience to his or her relationship with co-workers).

Caseworker tendencies. Caseworkers' individual tendencies were found to have a direct impact on their decisions, which in some cases outweighed other factors or evidence. For example, a worker who observed a child with a severe behavior problem or a family under extreme stress might consider those to be "good excuses," or acceptable reasons not to substantiate the referral, even if the caseworker believed that abuse or neglect had occurred.²²

Context. Other caseworker characteristics were found to affect decision-making in more complex ways. Characteristics such as more experience, higher self-assessments of skills, and more supportive relationships with co-workers were found to increase the likelihood of unsubstantiation.²³ Caseworkers who agreed State policy was important to their decision-making also were found to be more likely not to substantiate a case.²⁴ This may indicate that these caseworkers substantiate only in the most clear-cut cases.

2.1.3 Organizational factors

Aspects of the organizational environment also were found to be significant factors in the decision-making context discussed above. Workload stress and supervisor behavior were two of the most significant organizational factors in decision-making.

Workload stress. In one study, workers noted that having adequate time to do a thorough investigation was a significant factor to both substantiated and unsubstantiated case decisions.²⁵

In another study, increased stress resulting from heavy workloads tended to decrease the likelihood of unsubstantiation. The researchers noted several possible explanations for this seemingly contradictory finding. Because these workers feel overworked already, they may not have time to investigate to the extent they would like and, as a result, make a finding that is neither substantiated or unsubstantiated (such as "unable to determine"—a third decision-making category available in the State of Texas). Another possibility is that these caseworkers may be doing more in-depth investigations, which may tend to result in a higher percentage of substantiated findings. Or, these caseworkers may focus on a few complicated cases that involve more time and labor to substantiate. Since these are not cleared as quickly, they build up over time and become a greater proportion of a caseworker's workload. (Thus those workers will show a greater proportion of substantiated cases over time.)²⁶

Role of supervisors. Supervisor behavior tended to buffer caseworkers from some effects of the organizational environment, such as high stress. For example, when supervisors perceived their own work units as cohesive and themselves as supportive, there was an increased likelihood of unsubstantiation, suggesting that workers in units where the supervisors see themselves as supportive and work units as cohesive are more likely to define cases as unsubstantiated. These feelings of nurturance and support may offset the effect of a higher stress environment (noted above).²⁷

2.1.4 External factors

Differences among State policies were found to affect rates of unsubstantiation from State to State. Whether State policies allowed for caseworker uncertainty regarding the occurrence of maltreatment (by offering options for case decisions that acknowledged indeterminacy) was found to be significant, as was the degree of evidence required by State policy to substantiate a case.

Options for indeterminacy. "Two-tiered" States (where substantiated or unsubstantiated are the only two decision-making options) were found to have higher proportions of unsubstantiated cases than States where caseworkers have options that allow for some uncertainty, such as "unable to determine" or "insufficient evidence." States that allow for an alternative response track that focuses more on identifying needs and providing services than on gathering evidence also have lower proportions of unsubstantiated cases.²⁸ Researchers suggest this may indicate that in States with more decision-making categories, ambiguous cases are isolated in those distinct categories. In two-tiered States, on the other hand, ambiguous cases may be more likely to be unsubstantiated whether or not unsubstantiation is explicitly defined in State policy to allow for ambiguity.²⁹

Standard of evidence. States' standards of evidence required to substantiate a referral were found to fit one of three categories:

• "High" (e.g., requiring a "preponderance" of or "clear and convincing" evidence)
• "Low" (e.g., requiring "some credible evidence" or "credible evidence")
• "Vague or no clear evidentiary standard"

Not surprisingly, in States where the evidence threshold to substantiate a case was higher, the rate of unsubstantiated cases also tended to be higher.³⁰

Factors found not to influence substantiation rates. All other factors being equal, States that used a central registry, formal screening criteria, or a formalized risk assessment were not found to substantiate cases at significantly different rates than those that did not. Whether a State involves law enforcement in the investigation process also was found to be an insignificant factor in determining substantiation rates.³¹

2.2 How Decision-Making Impacts Outcomes

The rate at which cases re-refer to the CPS system is one way to measure outcomes. The studies synthesized here examined the connection between decision-making and re-referral in two ways:

• Directly—by looking at the correlation between particular case decisions and re-referral rates.
• Indirectly—by looking at how CPS decision-making impacts service provision and how services impact re-referral.

2.2.1 Re-referral

Re-referral rates for substantiated and unsubstantiated cases. Two studies found substantiated cases were more likely than unsubstantiated cases to re-refer, although the degree to which this was found varied. One of these studies found families whose cases were substantiated or inconclusive were "significantly more likely" to re-refer than those determined to be unsubstantiated.³²

The second study categorized subsequent CPS referrals into three groups, when comparing the likelihood of initially substantiated and unsubstantiated cases to re-refer:

• All subsequent referrals (substantiated or not)
• Subsequent referrals that were substantiated
• Subsequent referrals resulting in placement

For most maltreatment types, researchers in the second study found rates of re-referral to be only slightly higher for initially substantiated cases, in the categories of all subsequent referrals or substantiated subsequent referrals.³³ This differential was found to be greater when looking at subsequent referrals resulting in placement.³⁴

It is important to note that in the second study, despite the higher rates of re-referral among initially substantiated cases, the sheer numbers of subsequent referrals from initially unsubstantiated cases was found to be higher. While this may appear contradictory, the discrepancy is due to the much higher number of cases that were initially unsubstantiated.³⁵

Neglect and re-referral. Neglect cases that were initially substantiated were found to be highly likely to re-refer.³⁶ Regardless of the type of maltreatment alleged in the initial referral, subsequent referrals were found to be most often due to allegations of neglect.³⁷

Case factors affecting re-referral. Younger children were more likely to be re-referred to the system, for all types of maltreatment except sexual abuse.³⁸ On average, a case in which the youngest child was age 15 would be 50 percent less likely to re-refer than a case in which the youngest child was age 5. Younger perpetrators were also found to be at higher risk for re-referral.³⁹ Researchers suggest two possible explanations for this: older parents may be more likely to have their children age out of the study, or younger parents may experience a greater lack of knowledge and resources.

A parent as perpetrator increased the risk of re-referral dramatically in sexual abuse cases, moderately in physical abuse cases, and seemed to have little impact on re-referral rates for emotional abuse cases. Lower community income-levels were associated with a moderate increase in the likelihood of re-referral for all types of maltreatment except emotional abuse.⁴⁰ Families with fewer resources were also more likely to be re-referred.⁴¹

Decision-maker factors affecting re-referral. Caseworkers' perceptions that State policy was important to their decision-making significantly increased the likelihood of re-referral.⁴² This coincides with the finding that caseworkers who felt State policy was important to their decision-making were more likely to unsubstantiate, and may indicate the tendency to substantiate only the most clear-cut cases results in less protective decisions.

Organizational factors affecting re-referral. Some findings regarding organizational factors that predicted re-referral contradicted researchers' expectations. Although researchers originally assumed high workloads for caseworkers might lead to increased re-referral rates due to what they called "inadequate decisions," this was not found to be the case. High job satisfaction among supervisors, on the other hand, was found to increase re-referral—also contrary to what researchers expected, having assumed that job satisfaction would equate to "better performance" and therefore lower re-referral rates.⁴³

2.2.2 Services

Impact of CPS decision-making on service provision. Decisions regarding substantiation impact both the degree to which workers are able to provide services and the level of investment families are likely to feel in the services offered. Several studies suggest there may be children and families who are at risk but underserved because of a decision to unsubstantiate.⁴⁴ One study found families whose referrals were substantiated and whose children were offered in-home services were more likely to take advantage of services than were families whose referrals were classified as "inconclusive" but who were also offered services.⁴⁵

Impact of services on re-referral. When caseworkers reported a lack of agency resources to provide services for clients, re-referral rates increased.⁴⁶ Although neglect cases were found to be highly likely to re-refer, any form of services was found to reduce that risk.⁴⁷ Though there was not statistical significance, one study did find all cases that later returned as child fatalities received no services after the initial finding.⁴⁸

Type of services. One study examined the impact of the following three types of services on re-referral rates:

• Family preservation services were defined as brief, intensive in-home services provided to those families whose children were at greatest risk for foster care placement.
  
  
• Family centered services were less intensive in-home services provided over several months to families whose children were at lower risk for foster care placement.
  
  
• Foster care was defined as out-of-home services provided to children who needed to be removed from their families.
  
  

When the family in an initially substantiated case received family centered services or foster care, the risk of a subsequent referral was frequently reduced to a level no higher than for an initially unsubstantiated case. This despite the fact that substantiation, on its own, has been found to increase the likelihood of re-referral.⁴⁹ By contrast, it was found that family preservation services, which are intended to keep high-risk families together and avoid foster care placement, were associated with an increased risk of re-referral involving eventual out-of-home placement.⁵⁰

Length of stay in foster care. Longer stays in out-of-home care resulted in decreased re-referral rates over time. Children remaining in foster care fewer than 3 months experienced the highest rates of re-referral among those children who spent time in out-of-home care. Those in care between 7 and 11 months experienced the lowest rates of re-referral over time.⁵¹

³ CPS Decision-Making Ecology (Baumann, 1997), cited in Fluke, p. 6-7. back
⁴ English I, p. 133-134. back
⁹ English I, p. 115. back
¹⁰ Fluke, p. 109, 115. back
¹¹ English I, p. 118. back
¹² English I, p. 122. back
¹³ English I, p. 39. back
¹⁴ English II, p. 102. back
¹⁵ Fluke, p. 109. back
¹⁶ English II, p. 98. back
¹⁷ English I, p. 39. back
¹⁸ Fluke, p. 109. back
¹⁹ English I, p. 32. back
²⁰ English II, p. 105-106. back
²¹ English I, p. 119. back
²² English I, p. 121. back
²³ Fluke, p. 89-90. back
²⁴ Fluke, p. 72. back
²⁵ English II, p. i. back
²⁶ Fluke, p. 72, 75. back
²⁷ Fluke, p. 115. back
²⁸ Fluke, p. 32. back
²⁹ Fluke, p. 45. back
³⁰ Fluke, p. 33-34. back
³¹ Fluke, p. 37-42. back
³² English I, p. 42. back
³³ Drake, p. 174. back
³⁴ Drake, p. 142. back
³⁵ Drake, p. 169. back
³⁶ Drake, p. 134. back
³⁷ Drake, p. 154. back
³⁸ Drake, p. 142. back
³⁹ Drake, p. 185. back
⁴⁰ Drake, p. 143. back
⁴¹ English III, p. 63. back
⁴² Fluke, p. 111. back
⁴³ Fluke, p. 111-112. back
⁴⁴ Fluke, p. 120; English III, p. 68. back
⁴⁵ English I, p. 42. back
⁴⁶ Fluke, p. 126. back
⁴⁷ Drake, p. 134. back
⁴⁸ Drake, p. 168. back
⁴⁹ Drake, p. 141. back
⁵⁰ Drake, p. 142. back
⁵¹ Drake, p. 151. back

previous

You are in section: Table of Contents 1 - Introduction 2 - Findings 3 - Implications 4 - Recommendations for Future… 5 - Conclusion

next

This material may be freely reproduced and distributed. However, when doing so, please credit Child Welfare Information Gateway.

Email Order (Free)  Print  (PDF 265 KB)  Share

Updated on October 9, 2009