Mitigating Design Flaws in Family First
(Originally published July 2019)
Last month’s Sounding Board summarized key elements of Family First legislation and discussed the risk of seeking dramatic reductions in the use of residential care during a period of acute and chronic placement shortages in most states. This month’s commentary identifies design flaws in Family First and makes recommendations for mitigating these flaws.
The Family First Prevention Services Act (FFPSA) of 2018 is a misnomer. This legislation should be titled Family First Prevention of Placement Act. FFPSA takes dead aim at long term residential care for behaviorally troubled youth; and aspires to reduce the use of foster care for other children and youth on the verge of out-of-home placement due to abuse or neglect. FFPSA does not fund prevention or early intervention services for at-risk families prior to a CPS report or during or following a CPS investigation unless or until a child is deemed to be a “candidate for foster care,” i.e., at imminent risk of out-of-home placement. Policymakers and advocates with a superficial acquaintance with FFPSA may assume that the federal government has made a major investment in prevention of child abuse and neglect, which is not true. The federal government has expanded the entitlement to IV-E funded services to a (probably) relatively small percentage of children with open CPS cases, (i.e., “candidates for foster care”), only because of the strong possibility that most of the cost of the expanded entitlement will be recouped by a reduction in the IV-E reimbursable cost of residential care, the most expensive type of foster care by far. Federal policymakers appear to require a near certain return on investment before they will fund the prevention of anything pertaining to child welfare.
At this point, it is uncertain how strictly the federal Administration for Children, Youth and Families (ACYF) will enforce the requirement that children who receive FFPSA services be “candidates for foster care.” During the implementation of Family Preservation Services (FPS) in the 1990’s and early 2000’s, ACYF allowed state agencies to make children and youth eligible for FPS services by asserting that, in the absence of FPS services, a child would be placed out of the home. Some initial evaluations of FPS services found that 80% or more of children receiving
FPS services remained in the home; a finding which appeared to indicate that FPS was highly effective in preventing foster placement. However, several randomized controlled trials subsequently found that placement rates of children receiving FPS were only modestly lower than in the ‘business as usual’ control group. Child welfare staff around the country who made children eligible for FPS by asserting that these children would be placed out of home if not provided FPS services did not, in fact, place these children in foster care in the great majority of cases that did not receive FPS. These findings from randomized trials of FPS led to rancorous debates among child welfare scholars regarding reasons for the alleged ineffectiveness of FPS services and the goals of these services. In retrospect, what appears to have happened is that child welfare caseworkers faced with acute and chronic shortage of services played fast and loose with the concept of “imminent risk of placement” in order to obtain much needed resources for families. A few controlled trials of FPS programs obtained much more impressive results when agencies and researchers were able to narrowly define “imminent risk of placement” in a way that could not be easily manipulated by caseworkers and supervisors.
Who Is A Candidate for Foster Care?
“Candidate for foster care” is not a widely used concept in child welfare agencies or in research studies. It is a legal abstraction that ultimately means whatever ACYF says it means. “Imminent risk of placement” is a familiar concept from FPS programs, but can be easily “gamed”, as discussed above. At first glance, “risk of imminent harm,” i.e., an unsafe child, and “imminent risk of placement” may appear to have similar definitions, but this supposed similarity is an illusion. Contrary to the assumptions of policymakers, children and youth placed out-of-home are not always in danger, i.e., deemed unsafe, prior to placement. Some foster care placements are voluntary, and some children are still placed due to behavior problems or other disabilities, though to a lesser extent than in past decades. Furthermore, categories of risk are highly fluid. Caseworkers who choose not to place a child they recently asserted was at imminent risk of placement may not always be fudging eligibility criteria for FPS or other services. They may have changed their mind regarding the likely benefit and harms of placing a behaviorally troubled school age child or adolescent whose placement may quickly disrupt. Experienced caseworkers are thoroughly familiar with the emotional harm placement instability and cocktails of psychotropic medications can inflict on school age children and youth.
The developers of FFPSA and some of the legislation’s advocates appear to have a strong belief in evidence based assessment tools; so it will not be surprising if ACYF requires state child welfare systems to use a validated assessment tool for determining if a child is a “candidate for foster care.” However, assessment tools are easy to “game” with a goal in mind. In my experience (both recent and in past years), it is common to find completed risk and safety assessment tools in case records that have no discernible relationship to case narratives. The widespread belief that use of validated assessment tools in child welfare make assessments of safety, risk and well-being more objective is a fiction that many policymakers and scholars are unwilling to give up, regardless of decades of experience that suggests otherwise. The idea that completion of assessment tools equals objectivity just isn’t so; the reality is that a completed assessment tool often reflects the caseworker’s goals, i.e., establish eligibility for services or close the case, rather than actual safety threats and risk factors.
If ACYF holds to a narrow definition of “candidate for foster care”, no more than 10% of children in open CPS cases will be eligible for FFPSA services. On the other hand, if ACYF allows state agencies to utilize a broad definition of “candidate for foster care” where most families who currently receive in-home services become eligible for FFPSA funded services, a much larger percentage of children and families (possibly 20-30%) in open CPS cases will be affected by this legislation. The implementation of FFPSA and its costs and outcomes will be greatly influenced by ACYF’s operational definition of “candidate for foster care” in coming years. A broad definition of “candidate for foster care” will empower state child welfare agencies to fund a wide range of in-home services but make it more difficult to demonstrate that these services prevent foster care. A narrow definition of “candidate for foster care” will limit the scope of the legislation to a small percentage of children being served by child welfare systems but increase the likelihood of demonstrating that FFPSA has reduced foster placements.
Evidence Based for What?
FFPSA requires state child welfare agencies to utilize services deemed promising, supported or well supported by a federally funded evidence- based practice Clearinghouse, presumably modeled on the California Evidence Based Clearinghouse. However, very few of the programs typically described as evidence-based have been tested for placement prevention. For school age children, possibly Multisystemic Therapy (MST) and Functional Family Therapy (FFT) might plausibly be described as promising, supported or well supported for placement prevention. As with other evidence-based programs, it is uncertain whether these programs will be effective when taken to scale in state systems. The idea that programs demonstrated to reduce behavior problems, e.g., oppositional behavior and conduct disorder, anxiety, depression or PTSD in children or adults will consistently prevent foster placement is a leap of faith; it is not an evidence-based gamble. Some EBP’s have been shown to reduce CPS reports or CPS re-reports, but reducing reports or re-reports is not guaranteed to reduce foster placements, given that only a small percentage of children reported to CPS are placed out of the home. At least one promising program, Triple P, has been mired in controversy regarding the quality of comparison group research conducted in the U.S.
Programs for young children that would almost surely reduce out-of-home care include Pregnant and Parenting Women’s (PPW) residential care for substance abusing mothers and their babies, and early childhood education programs modeled on Chicago’s Parent-Child Centers. However, these are expensive programs. Even with the federal government paying 50% of the costs, I doubt that states will make the large new investment in costly programs such as PPW required to decrease foster care placements by even a small percentage. Hopefully, I will be proven wrong.
FFPSA offers states incentives for using EBP’s approved by a federally funded Clearinghouse rather than creating incentives for innovative programs and services developed in states or communities. Programs developed by state child welfare systems must be shown to be effective in rigorous research studies (a costly process), while programs with the Clearinghouse’s stamp of approval can be utilized without in-state research studies. This is, to put it bluntly, ass backwards. The evidence for placement prevention is thin and inadequate; what is needed is a large investment in innovative ways of protecting children and youth, for example, through use of resource families which offer birth parents respite care, coaching of parenting skills and ongoing emotional support.
The developers of FFPSA apparently believe that a powerful array of placement prevention services/programs already exist; and that public policy should offer incentives for the widespread use of these programs. However, this is a misreading of the research on EBP’s, and is an exaggeration of the limited knowledge regarding effective and safe placement prevention. This is a serious design flaw which can and should be fixed in law before most states fully implement FFPSA in 2021.
Strategies for Achieving Placement Prevention
FFPSA uses both good and bad ways of achieving placement prevention. Hopefully, FFPSA will make it possible for state and county-based child welfare systems to fund a wide array of programs and services, while rigorously evaluating their effectiveness and possible unintended consequences. Even with greatly expanded in-home services, child welfare practitioners will need to be frequently reminded that “treatment is not protection,” and that treatment of chronically relapsing conditions such as substance abuse and mood disorders is a rocky road that takes time and includes “two steps forward, one step back.” Children in maltreating families may not be immediately safe (or safer) because parents have entered a treatment program. There will also need to be experimentation with a range of safety-oriented services such as crisis intervention for children and youth with mental health problems, extensive use of child-care and respite care, crisis nurseries, etc. Nevertheless, expanding and strengthening a continuum of family support services is the best way, and only sound way, to achieve placement prevention.
Measuring a state’s rate of placement prevention and comparing this rate to a national standard (per the CFSR) is a risky approach to placement prevention. For example, using “the percentage of children receiving FFPSA funded services who do not enter foster care within x months” as a performance indicator may act as an incentive for child welfare caseworkers and supervisors to refuse to acknowledge that placement prevention services are not working, and that a child is unsafe despite these services. It is not wrong for ACYF to measure the effectiveness of FFPSA services, but it is dangerous to use such measures to evaluate and grade state child welfare systems before Family First has been carefully evaluated, including some randomized controlled trials and quasi- experimental studies.
Data experts may argue that measures of the effectiveness of placement prevention services can be balanced by use of child safety measures, and that performance indicators can refer to “safe reduction” or “safe use of prevention services.” However, currently public child welfare systems in this country lack adequate and reliable measures of child safety, which makes any reference to “safe reductions” or “safe and effective prevention” empty rhetoric, absent development of better measures of child safety.
The most dangerous way to achieve placement prevention and foster care reductions is by continually denigrating foster care while giving lip service (only) to child safety; and making foster care reduction a policy goal on par with child safety and permanency. Child welfare staff are skilled at “reading” the words and actions of agency mangers to discern their actual intentions. When caseworkers and supervisors understand that agency leaders want large reductions in foster care and are willing to take big risks to achieve this goal, they will act accordingly. When given incentives to do so, caseworkers will sometimes use denial of a child’s need for foster care and/or use rationalization to defend their actions in ignoring safety threats. These are not merely hypothetical considerations, but a reflection on the past two decades of child welfare history in many state and county systems that have achieved large reductions in foster care, until these reductions were checked by high profile child deaths or other perceived failures of child protection.
It’s true that foster care is not always safer than in-home services. It’s also true that other things being equal, fewer children in foster care is better than more children in foster care - much better. Nevertheless, policymakers and managers need to be extremely careful in how they go about achieving placement prevention and in how Family First is
implemented. The less ideological rhetoric the better. Caseworkers and supervisors should be considering the need for out-of-home care in specific circumstances, not in the light of a placement prevention rate or an agency’s foster care reduction goal.