Keywords

The stakes in dependency courts are extremely high, both for the welfare of involved children and for the fundamental rights of both children and families. These cases involve the state reaching into a family and, at the minimum, forcing parents and children to engage in activities and services not of their voluntary choosing. Further, state action often includes removing children from their parents’ care and placing them in the foster care system. Although, in the past, dependency court stakeholders may have viewed foster care as a benign alternative that held few risks when compared to a potentially unsafe family situation, recent research has illustrated the harms inherent in growing up in foster care (Fowler et al. 2009; Lawrence et al. 2006). Finally, in a small but significant percent of cases, the state moves on to terminate the parent’s rights to his or her children, and subsequently the children’s ability to have a relationship with his or her parent.

Given the high stakes, and the large number of involved citizens,Footnote 1 the dearth of social and behavioral science research that is relevant to dependency court (not surprisingly related to limited federal funding for this area of research) is notable. The previous chapters (Lederman and Gatowski in this volume) illustrate two significant, but different, ways that dependency courts might utilize social/behavioral science to improve their work. Gatowski et al. in this volume (Chap. 9) describe evaluation research that tests the effectiveness of dependency court accepted “best practices.” They wisely suggest that just because there is a consensus from professionals on a best practice that empirical validation of the effectiveness of the practice is needed before courts should be confident that the practice will likely result in desired outcomes. These researchers focus on court processes themselves, including, for example, timing of hearings, frequency of hearings, judicial involvement, and parent engagement in three court models: a traditional dependency court model, a family drug court model, and a therapeutic justice court model.

Judge Lederman, also in this volume (Chap. 9), does not focus on the court process itself, but instead makes an argument for judges to be mindful of the scientific basis for the evidence they consider (e.g., assessments, recommendations for interventions). She expresses legitimate concerns for the lack of scientific foundational support for most court-ordered services for parents and children. She describes evidence-based psychological assessments and interventions (Lederman and Osofsky 2008) that she is able to rely on in her court and how those interventions contribute to desired outcomes for children and parents. Judge Lederman calls upon judges to require scientific underpinnings for the services they require for parents and children.

As noted by Gatowski (Chap. 9), since 1980 dependency courts have had oversight responsibility over the state agency that is charged with caring for the children, providing rehabilitative services to the parents, and working to keep children in their homes, return them to their parents, or find another permanent family. Courts depend on others to bring them information, in particular caseworkers, attorneys, and treatment providers. In most states, the rules of evidence are relaxed for disposition and review hearings, so that information that comes in can be quite informal, hearsay ridden, and subjective. In some states, the courts have a fair amount of authority over the state agency. In many states, the courts have quite limited authority.

These two papers make strong arguments for the need for more research regarding court processes as well as research regarding the validity of assessments and the effectiveness of services that courts utilize in making far reaching decisions in the lives of parents and children. This chapter will briefly discuss some challenges that might be considered as courts and social–behavioral scientists work together to build a dependency court system that meets its expectations for a fair and efficient court process. Additionally, the chapter will address potential challenges and recommendations for judges who expect that court referred services have a sound scientific basis.

Challenges in Isolating Dependency Court Process Features

Gatowski et al. describe a wide range of broad features of best practice dependency courts that include judicial leadership, a problem-solving judicial orientation, a therapeutic jurisprudence approach, early and active engagement of parties, a collaborative approach, etc. They compared a general best practice dependency court with two specific problem-solving courts (family drug court and therapeutic jurisprudence court) and found the two specialized courts to have significantly better outcomes with higher reunification rates and lower re-entry (into the child welfare system) rates. There were no differences in case processing timeframes, which were all well within federal and state timeframes. This comparison of types of best practices is a good first step in assessing the effectiveness of various court procedures and Gatowski et al. do an admirable job in describing the aspirational features of these best practice courts. It is difficult, however, to tease out which factors may be contributing to the outcomes. Is the higher reunification rate and lower re-entry rate for family drug courts related to features of the court process or simply to the fact of prioritization for drug treatment? The relative similarity of outcomes for the drug court and therapeutic jurisprudence cases suggest that perhaps the group participation aspects of the drug courts are not critical. Then again, it may be that the group dynamic is critical for substance abusing parents but not for the general child welfare population that includes but is not limited to substance abuse problems. It is also not clear whether some of the effects are individual judge variables rather than court process variables (difficult to untangle in the best practices model emphasis on the significance of judicial leadership).

One approach that might help move the field along would be to reduce the number of features that are being tested. Gatowski et al. describe nine broad principles. Perhaps, the field leaders who have promulgated these best practice principles could identify a handful of important features that could be validly and reliably operationalized. Gatowski et al. have already made progress in this area. For example, a problem-solving orientation is described as asking questions, seeking more information, etc. Early and active engagement is described as including direct interaction with respondent parents in court hearings. A therapeutic jurisprudence approach includes frequent and proactive judicial review (frequent hearings, individualized court orders—not rubber stamping the agency’s recommendations). More foundational work seems needed to identify, operationalize, and validate measures of important best practice features that may (or may not) contribute to desired judicial practice outcomes. Further, researchers would need to then systematically study these factors either one or two at a time, with more sophisticated statistical models that can handle multiple variables, or in some combination of both approaches.

Court cases are complex. A multitude of factors can potentially affect outcomes for children and families, including features of the court process, activities of the child welfare system, and behaviors of the parents or children. A systematic study of the court variables is needed to identify which of them are the ones that contribute to the desired outcomes.

Challenges in Selecting and Measuring Outcomes for Dependency Court Process Reform

Gatowski et al. (Chap. 9) report on three sets of outcomes: case processing timeframes , percentage of reunifications, and re-entries into the system. Reducing case processing timeframes is an important management outcome for courts, many of those over the past decade have had to manage increasing caseloads with decreasing resources [US Department of Justice (DOJ) 2008]. Further, if one assumes that mandatory court involvement is an intrusion into family life, then limiting the length of that intrusion seems to be desirable (Sankaran 2010; Guggenheim 2005). More efficient resolution of child welfare cases so that they do not linger in the court system appears to be a reasonable outcome [National Council of Juvenile and Family Court Judges (NCJFCJ) 1995] and its measurement seems relatively straightforward (DOJ 2008). Similarly, the percentage of reunifications is relatively straightforward to measure and it also reflects the legal framework that reunification should be the goal in most cases (DOJ 2008; Sankaran 2010). Re-entry into the system, considered a proxy for child safety is, on the one hand, perhaps the most critical outcome and also the one that contains the most measurement challenges that will be discussed below.

The outcomes in the reported study are included in the broader array of court performance measures that are included in the Toolkit for Court Performance Measures in Child Abuse and Neglect Cases (DOJ 2008). The development of the Toolkitreflects an interest in a variety of national groups to develop uniform measures for court performance measures, many of which could be considered outcomes. The Toolkit identifies 30 measures of child safety, child permanency, due process, and judicial timeliness that are considered representative of desired outcomes.

The safety measures include (a) the percentage of children who are abused or neglected while under court jurisdiction and (b) the percentage of children who are abused or neglected within 12 months after the case is closed following a permanent placement. Unlike some of the other outcomes (e.g., there are differences of opinion as to whether it is better for children to take longer for permanency if they can eventually be reunified with their parents as compared to more speedy permanency that involves termination of their parent’s right and adoption; Guggenheim 2005), the goal of child safety is uncontroversial. Gatowski et al. measure of re-entry is an example of this safety outcome measure.

Although re-entry is one measurement that can be used to assess child safety, the Toolkit also suggests emergency removal orders, custody transfer orders, and written agency reports for children still under court jurisdiction. For children who have been released from court jurisdiction, the Toolkit recommends child welfare “hot line” reports on children, new petitions alleging maltreatment, and/or judicial findings of abuse or neglect. While broadening the official information sources makes it more likely to detect further maltreatment of children, these sources all require that the maltreatment has come to some public attention. Further, facts about the family situation must support shifting community perspectives, policies, and politics regarding thresholds for state identification and involvement (Institute of Medicine and National Research Council 2012). Researchers that study child maltreatment have noted that gathering information directly from children and parents, in addition to the public sources, strengthens the accuracy of maltreatment measurement (Everson et al. 2008; McGee et al. 1995; Stockhammer et al. 2001). Further maltreatment to children could more accurately be measured if court researchers used methods and instruments developed in the social science arena. For example, the National Survey on Child and Adolescent Well-being (Casanueva et al. 2012) has three measures that address child maltreatment: Violence Exposure Scale, injury questions from the Child Health and Illness Profile, and an adapted Parent–Child Conflicts Scale that gathers specific maltreatment information. This information is gathered directly from children and families through phone interviews. Direct reports from children and parents would strengthen the measurement of child safety, even though the accuracy of those self-reports may also be limited—by the children and parent’s willingness to tell the truth. Because of the importance of the safety outcome in determining the effectiveness of various court processes and procedures, the use of reliable and valid multimodal (official records and child/parent instruments) measurement methods would considerably advance our understanding of the impact of various court processes and innovations. A first important step in developing a multimodal approach might focus on a single court and set of children. Various approaches of measuring safety could be undertaken and the extent to which the different measures of safety agree could be assessed. If there is not an agreement among different approaches that measure the same phenomenon (e.g., child safety), then more work would be needed to refine the approach. Without this foundational work, we risk misplaced confidence that certain court processes make children safer, when in fact they may not be doing so.

Challenges in Raising the Bar for Science in Dependency Court

Lederman (Chap. 9) discusses scientific knowledge that should inform dependency courts; in particular that infants and toddlers can be psychologically harmed by maltreatment; that relationship focused interventions for them can successfully address the harm; and that many types of parenting training are ineffective although effective models do exist. She could have also noted that substance abuse treatment programs vary in their adherence to evidence-based practice and in their effectiveness (Miller 2007; Garner 2009). Lederman also describes a recent survey that suggests the vast majority of social workers and other stakeholders in the child welfare field do not understand what evidence-based practice means and would presumably not be able to critically evaluate the potential effectiveness of typical child welfare services (Lederman et al. 2009).

The effectiveness of court-ordered child welfare services is critically important. When children are maltreated and the state intervenes, the role of the court and the agency is a rehabilitative one. The agency, with the court’s approval, must determine what changes a particular parent needs to make to be able to nurture her child and keep him safe. Next, the agency needs to determine what services could help the parent to make these changes. Choosing the appropriate services should include a determination of the relevance of the service to the change the parent needs to make, and the effectiveness of the service in helping parents make that change. As Lederman argues, required judicial findings that the agency has made reasonable efforts to reunify a child, must include an assessment that the offered services are effective and relevant and will reasonably assist the parent in ameliorating the problems that resulted in the state’s intervention. If offered services are not evidence based—if they have not been shown to effectively address the type of problem the parent needs help with—then the court may, and should, exercise its authority and make no reasonable efforts findings.

How are judges to know whether particular services are evidence based or whether particular children and parents need the services? Monahan and Walker (2011) have, over the past several decades, developed a scheme for judicial uses of social science. Although the rules of evidence are typically relaxed in the dispositional phases of dependency cases, looking to Monahan and Walker’s (2007) discussion of how social science research can be used as social framework may be instructive. For social framework information, social science provides a context for a particular situation. In dependency court, for example, general research about the vulnerabilities of maltreated infants and toddlers, or the features associated with evidence-based interventions could provide the court the context to make judgments in the individual case before him or her. Monahan and Walker (2007) outline steps that judges can take in using social science that includes:

If the parties or amici do not submit social science studies, request such studies from the parties or amici, or obtain them from the court’s own sua sponte investigation of published sources. (p. 162)

Thus, they would suggest that judges may routinely ask for information about the evidence basis for any service that he or she is asked to order,Footnote 2 about demonstrations that the local intervention has fidelity to the evidence-based model, and about the success rates of programs.

If a judge knows that the majority of services in his or her jurisdiction are not evidence based, what can he or she do? Judge Lederman presides in a relatively service-rich urban area with excellent universities, a medical school, and many well-trained professionals. Most dependency courts do not have access to as wide a range and as high a quality of services as that. Courts cannot build services. However, the judges can provide community leadership to encourage growth of such services. Judges can routinely question whether a proposed service is evidence based. Judges can also make “no reasonable efforts”Footnote 3 findings to accurately communicate that parents have not been provided services that would help them rehabilitate as well as put pressure on the agency to develop appropriate services.

Conclusion

The two previous chapters develop a nice frame for using science to address some of the challenges faced by dependency courts. Gatowski et al. explore research methods that can provide important empirical tests for the consensus best practice recommendations that are currently driving dependency court reforms. This line of research is aimed toward determining scientifically supported effective court processes and practices, that is, clearly defined court practices that are linked to clearly defined desired court outcomes. In contrast, Lederman looks to research to improve the quality of judicial decisions and orders. She challenges courts to ensure that they order effective interventions that have a reasonable chance to ameliorate the problems that brought the family into the system. Both approaches face challenges, as this discussion has detailed. Still, it is laudable and hopeful that dependency courts are looking at science to assist them in improving processes as well as judicial decisions. One hopes for a growth in relevant social and behavioral scientific research to assist the courts in this endeavor.