Home First: Stability and Opportunity in Out-of-Home Care

Abstract

In this report, the concept of “Home First” is introduced for those children who require long-term, non-kin placements. The term “Home First” connotes a placement engendering stability and continuity; this concept is introduced in conjunction with an evaluation of the historical, theoretical, and empirical evidence surrounding different forms of out-of-home placement, including group-care placements and foster family care. In light of these observations and studies, this report will argue that stability is a major factor, perhaps a necessary if not a sufficient condition, in successful child development. It will argue for the initiation of a new focus on the creation of long-term positive and stable residential placements within the out-of-home care system and show that such placements can and have contributed to the development of healthy, happy, and successful adulthoods. This report offers a bio-psycho-social perspective on child development in out-of-home care. It provides a brief overview of the multiple bio-psycho-social theoretical perspectives that inform us on the necessary role of stability in growth and development and the contribution of instability to dysfunction. This report considers stability in out-of-home care in relation to its associated outcomes and those factors believed to enhance or detract from these outcomes. It reviews the history of substitutive care provision for children and youth and the role of the “stability objective” in that history. Finally, it looks at how child welfare system priorities have influenced stability, and it offers some suggestions for ensuring more stable growth and development in child placement provision.

1. Introduction

Children who for various reasons cannot live with their birth family have historically been housed in a myriad of out-of-home placements. These placements include foster family homes, alms-houses, work-houses, the traditional orphanage, children’s homes, homes for dependent and neglected children, and more specialized group-care such as homes for unwed mothers, psychiatric hospitals, residential treatment centers, halfway houses, and even nursing homes and homes for the elderly. Other places where populations of children have resided include boarding houses and schools, as well as detention homes for juvenile delinquents, training schools, and other correctional institutions. In the United States, today’s out-of-home placement of choice for the majority of children who are unable to live with biological parents is foster care, generally including all types of foster-family arrangements and group settings, excepting those associated with the criminal justice system and/or privately funded boarding schools and hospitals. In any given year, about 600,000 children are served in foster care in family and non-family settings, with a daily census of about 400,000 [1].

Out-of-home care placement has generally reaped considerable criticism. Despite the high level of need of children in foster care for services and supports, a significant percentage of these children do not receive the services they need to secure their well-being: 34% received no immunizations; 32% continued to have at least one unmet health need after placement; 12% received no routine health care; and less than 10% had received special education services despite research indicating that more than 50% of children in foster care have developmental delays [2]. Appallingly, between twenty to thirty percent, one-third of children in foster homes, are living well below the poverty threshold [3,4,5].

McDonald et al. [6] examined 29 studies published between 1960 and 1992 on the impact of childhood out-of-home care on adults’ self-sufficiency, adjustment, family, social support, and personal well-being. Study results indicated that, in comparison to children not placed in out-of-home care, adults who experienced out-of-home care had poorer school performance, lower marriage rates, poorer mental and physical health, and higher rates of school dropout, public assistance, homelessness, arrest, and chemical dependency. More recent reviews consistently show similar negative results in comparisons of those who experienced out-of-home care with those who have not in national samples [7,8,9,10,11,12]. However, Buehler et al. [7] found that when compared with a sample matched on demographic indicators (age, race, gender, parents’ education, and presence of a stepparent), the unfavorable differences attributable to out-of-home care experience disappeared. The results of his study suggest that much of the negatives attributed to the out-of-home care experience may be attributable to the fact that many children who are placed in out-of-home care are members of lower socio-economic groups. Yet, in Missouri, it is reported that over 50% of foster children do not graduate high school, upon leaving foster care over half become homeless, and 80% of the young women become pregnant before reaching the age of 21 [13]. This description of the achievement of foster care youth is consistent with the more recent reports on their employment status. Groups of 24-year-olds who did not experience out-of-home placement were found to be more likely to be employed, have higher earnings when employed, and were more likely to be earning a living wage than their aged-out foster youth counterparts [14]. Additional study findings continue to confirm these disparate results showing low rates of employment and earnings persist for age-out youth compared to the low-income and national samples through age 24 and in some areas through age 30 [10]. Therefore, it appears that the central issue emerging from these studies is whether growing up under State protection can yield no better outcomes than growing up in poverty or even match poverty youth outcomes.

Collectively, these findings raise much concern about the nature of out-of-home placement and its underlying assumptions. They have and will continue to spur efforts to find vehicles for improving the system and for addressing the challenges posed by system procedures. In this report, the concept of a “Home First” is introduced for those children who require long-term, non-kin placements. The term “Home First” connotes a placement engendering stability and continuity; this concept is introduced in conjunction with an evaluation of the historical, theoretical, and empirical evidence surrounding different forms of out-of-home placement, including group-care placements and foster family care. In light of these observations and studies, this report will argue that stability is a major factor, perhaps a necessary if not a sufficient condition, in successful child development. It will argue for the initiation of a new focus on the creation of long-term positive and stable residential placements within the out-of-home care system and show that such placements can and have contributed to the development of healthy, happy, and successful adulthoods. This report offers a bio-psycho-social perspective on child development in out-of-home care. It provides a brief overview of the multiple bio-psycho-social theoretical perspectives that inform us on the necessary role of stability in growth and development and the contribution of instability to dysfunction. This report considers stability in out-of-home care in relation to its associated outcomes and those factors believed to enhance or detract from it, and reviews the history of substitutive care provision for children and youth and the role of the “stability objective” in that history. Finally, it looks at how child welfare system priorities have influenced stability, and it offers some suggestions for ensuring more stable growth and development in child placement provision.

2. Challenges to the Status Quo

2.1. Current Approaches: Child Welfare Policy and Out-of-Home Care

The Federal Adoption Assistance and Welfare Act of 1980 (Public Law 96-272) eliminated the foster care funding portions of the AFDC program (Title IV-A of the Social Security Act) that provided open-ended funding for long-term foster care. It instituted a new program—Title IV-E of the Act—that established the goals of preserving families and securing permanence for children. The Act also conditioned federal financial participation in state foster care programs on a number of factors intended to achieve these goals. Among these goals were the provision of pre-placement, preventive services prior to entry into care, the development of written case plans, the provision of reunification services, and regular court hearings—including dispositional hearings within 18 months of the child’s entering into care to address the child’s permanent plan. The Federal goals were later revised with the passage of the Adoption and Safe Families Act (Public Law 105-89) to include safety, permanence, and child well-being [15]. The changes to child welfare policy, practice, and financing in this legislation were driven by the assumption that if children spent less time in open-ended foster care—if they achieved earlier “permanency”—the quality of their lives and their consequent outcomes would be improved. The applied psychological theory forming a foundation of this work was derived from Goldstein, A. Freud, and Solnit’s [16] work Beyond the Best Interests of the Child that suggested that by the time the child entered foster care, meeting their best interests was no longer possible. Instead, this framework asserted that what remained possible was finding the least detrimental available alternative.

Consequently, child welfare policy and practice focused on achieving “permanence” of placement for children removed from their family home—i.e., a placement that will last indefinitely [17]. Reunification was deemed most desirable, yet other “permanent placements” (in order of preference) included adoption, guardianship, long-term foster care with kin, and long-term foster care with non-kin. Notably missing from this list was group care and the disappeared children’s home or orphanage.

One of the most important principles of child placement law is placement stability [18]. In 2015, California, the state with the largest out-of-home care population, passed its Continuum of Care Reform Act (CCR), reforming the state’s policy regarding placement and treatment for youth in out-of-home care—a law consistent with the federal Family First Prevention Services Act of 2018 [19] which seeks to curtail using congregate care for children and youth. Family First limits the use of Title IV-E funds for children and youth placed in nonfamily settings and creates Qualified Residential Treatment Programs (QRTP) as a new federal classification of congregate care facilities [20]. Metcalf et al. [21] conducted an initial assessment of the short-term impact of the CCR in a large urban county, focusing on the extent to which the legislation (a) increased placement stability, (b) decreased reliance on congregate care, and (c) reduced the time it takes for youth to be placed in family-based care (i.e., kin or foster care) during the first two years of its implementation. They compared two years of placement data among 359 maltreated youth (aged 6–18 years) whose placement experiences occurred before, during, or after the CCR was enacted. They found that youth had fewer placement changes during and after the CCR compared to before; however, youth did not differ in the proportion of time they spent in congregate care before, during, and after the CCR was implemented. In fact, the most common pathway for youth post-CCR was three congregate care placements—likely due to limits on the amount of time youth were allowed to spend in a particular congregate-facility. Finally, a greater proportion of youth received at least one family-based placement prior to the CCR’s implementation than afterward, though the time until youth reached these placements did not differ across groups. Unfortunately, at the same time that the CCR stressed greater reliance on family-based care settings and limited use of congregate care facilities, there was a significant deficit in the number of available foster homes. In 2018, there were 53,411 youth in out-of-home care in California but only 33,985 licensed foster homes available [22]. It would seem that limiting time in a given congregate care facility in the absence of adequate supply of family-based facilities, and longer waits for such placements (a fact that would have reduced the family-placement failure window during the study’s two-year follow-up) will have the potential for increased instability going forward.

Placement instability is characterized by disruption and, from a psychological perspective, loss of important formative relationships. It is characterized by the disruption and loss of attachment figures, home, school, and community, and often becomes chronic instability as children are transitioned to different placements over their placement “career”.

“Permanent placements” are those anticipated to lead to stability in the child’s life, but they unfortunately do not guarantee stability. Of each cohort of children entering out-of-home care placement, approximately 20% or more fail to get a “permanent placement”, 10% to as high as 53% because they fail in reunification attempts and 10% because they remain “in care” cycling between various placements—some of which may have been considered permanent. This point is illustrated by the experiences of 29,000 California children over five years who entered care for the first time in 1998 (10% were still in care five years after coming into care; 35% in care for 12 months experiencing ≥3 placements [23]), the experience over three-years of an Ohio cohort (N = 2616) (28.4% remaining in care >24 months after first entry [24]), the experience in a Midwest cohort (N = 4177) (53% experiencing a placement change within three years [25]), and in a California sample called the Alameda Project [26,27] where the 20% figure held despite the fact that the study was an intervention designed to promote permanency placement. In this later study, approximately 60% of the control group was expected to be in long term, out-of-home care placement, without the permanency placement intervention. Finally, USDHHS [28] reported that 13% of children in the child welfare system will never achieve permanency. In these studies, there is no way to know how many of these children spent the rest of their childhood in out-of-home care. However, the message of these studies is clear: since these are cohort studies that have yielded similar findings over several decades, one might expect the experience of each cohort to be repeated each year. Given this repetition, a very large cumulative population is left either without ‘permanent’ arrangements cycling through placements and/or moving in-and-out of the system in a recycling pattern. It is on the needs of this 20% of each cohort group that this report on stability in out-of-home placement is focused—those children who may require long-term, non-kin, out-of-home care placements. The laws are set up to prioritize the needs of those who can be reunified, are adoptable, and/or have potential kin or kin guardians. This report considers the needs of the cumulating “20%” cohorts requiring long-term out-of-home care.

2.2. Sustained Instability in Out-of-Home Care

Stability or instability in out-of-home care may be defined in terms of a child’s experiences of residential environment change beginning with and following their removal from their home of origin. When part of a process that involves an extended stay out of one’s home of origin, all residential changes involving a new primary care-giver are of concern. Instability may be measured by the child’s residence change, including their removal from their home of origin, attempted reunifications, recycling back into the out-of-home care system, and moves within the system. The extent of instability associated with each change may be compounded by changes in primary caregiver, changes in school, loss of social support, loss of community/church/extracurricular activities, and loss of adult and peer relationships. Given this definition, it would appear that instability in the child placement population is increasing in spite of efforts to the contrary. Surveys of children in foster care [29,30,31,32] have shown that in the 1950s through the mid-1970s, the majority of foster children experienced only one or two placements before returning home or being placed for adoption. These same surveys also demonstrated that some children experience repeated changes in placements. Jeter [29] reported that 28% of the children in foster care in 1961 had experienced three or more placements and that 2% had experienced eight or more placements. Shyne and Schroeder [30] estimated that 17% of the children in foster care in 1977 had experienced 3 or more placements and that approximately 2% had experienced 6 or more placements. Similarly, the Children’s Defense Fund [31] estimated that 4% of the children in foster care nationally had experienced six or more placements. The Casey Family Programs [33] reported on a sample of 1609 alumni who were placed between 1966 and 1998 with a Casey foster family for 12 months or more and had been discharged from foster care for at least 12 months. They found that placement changes continued to be a controversial issue even as agencies tried to minimize this form of disruption in a child’s life. While 18% of their alumni sample had three or fewer placements during their entire history of foster care, over half of the sample (56%) had seven or more placements while in public, Casey, and/or other agency foster care. About 3% of the alumni had 20 or more placements.

Between 1998 and 2007, the proportion of the population of children in foster care in California with more than two placements (exclusive of transitional shelter placements) and in the system for more than 24 months (from their last entry date) had steadily increased: from 39.3% in 1998, to 46.6% in the Sept. 1999/Oct. 2000 reporting period, to 65.1% in the Oct. 2006/Sept. 2007 reporting period. This increase in instability in placement seems unrelated to the change in the population figures between 1998 and 2007—i.e., since the trend in increasing instability was occurring during an initial rapid increase in population in the first three years (1998–2000) and continued through the subsequent population decline in the population in the later six years (2001–2007) [34]. Thus, it was unlikely that the increasing instability in foster care resulted from the presence of a residual group of more disturbed, increasingly unstable, children in long-term foster care and is more likely that it was the result of systemic issues. From these and previously noted data, it would appear that instability is increasing despite efforts to reduce it. More recent data continue the trend despite the new FFPS Act. Between 2011 and 2020 across the US, 35% of foster children in the year had accumulated more than two placement experiences [35]. In a Texas study, Font et al. [36] found 16% of reunified children reentered care within 5 years and that among those that remained in care beyond 18 months, the hazard of reentry was 52%, 68% for those with substance abuse and mental health issues (the latter potentially a result of instability).

Moreover, despite the best efforts of child welfare workers and policy changes to quickly secure permanent placements for children, over the past three decades, many youths have continued to spend multiple years in increasingly unstable out-of-home care. For example, in 2001, 51% of the youth who were placed in out-of-home care in America spent one year or more in placement; over 27,000 spent five years or more in care in 2001. Alumni in the Casey study spent a median average of 6.2 years in foster care. The considerable lengths of stay of the long-term youth is not typical of foster youth today, although about 10% of youth leaving foster care in the United States continue to have such extended placement histories. For example, the average age of entry into Casey was 13.2 years (SD = 3.0); Median = 13.8 years. The average length of time in foster care, adjusting for periods of time spent at home was 7.2 years (SD: 4.3), Median: 6.2 years. The average length of time in foster care without adjusting for periods of time spent at home or in non-placement living situations was 10.5 years (SD: 5.1), Median: 9.8 years [33]. There is generally no follow-up in child welfare government statistics. These statistics focus on single episodes of out-of-home care experience—i.e., experiences from entry to exit. Thus, there is no link from the first to subsequent episodes of care, re-entries into the system, and total time from first entry to final system exist.

2.3. Theoretical Considerations in Out-of-Home Care

2.3.1. Building Social Margin: A Refocusing of Theoretical Perspectives from Disorder to Opportunity

Life’s outcomes are contingent on the development of social margin. “Social margin refers to the set of resources and relationships an individual can draw on to either advance or survive in society. It consists of family relations, friendships, possessions, skills and personal attributes that can be mortgaged, used, sold or bartered in return for necessary assistance. Social margin aids advancement and protects, or softens the fall of the downwardly mobile” [37]. The accumulation of social margin begins at the earliest stages of life in one’s home setting and continues throughout life at each successive place we can call “home”. There is a growing theoretical literature on the importance of a “home” and homelessness. As approximately a fifth to one-quarter of those who are homeless in America have grown up in out-of-home care, this is a definitive negative outcome associated with out-of-home placement [38,39]. This outcome continues according to Dworsky et al. [40], who, using data from three mid-western states, found that during the transition from foster care to adulthood, between 31% and 46% of study participants had been homeless at least once by age 26 years.

As Athol Fugard, the South African playwright, describes it, home is “where you finally belong and where you have to go back to. It’s where you are owed. Home is a very deep transaction [41]”. At home there is a debt obligation to you, an entitlement. The principle of State custody is “parens patriae”, where the State is the stand-in for the parent. Therefore, the State must provide the “home” and the opportunities and social margin attached to it for those taken into out-of-home care. Much of the theory of out-of-home care is focused on disorder; a theoretical refocusing is required to concentrate on the tenet of building opportunity through the development of social margin.

2.3.2. Developmental Theories and Finding Opportunity

Developmental theory posits that children and youth develop their identity and social competence by completing sequential stages of maturation, each building on the previous stage. Erikson [42] suggests that psychosocial development, the changing ways we perceive ourselves individually and in relation to society, occurs in eight stages, four of which deal with childhood, a fifth “identity vs. role confusion” more relevant to the teen years and the transition to adulthood. The first of Erikson’s stages, “trust versus mistrust”, usually occurs from birth to 1 year and forms the basis for all future relationships. This ability to trust is largely dependent on whether the child’s immediate physical (food, sleep, and comfort) and emotional needs are met. Emotional needs refer primarily to attachment needs [43]. In the second stage, “autonomy versus shame and doubt”, usually occurring between 1 and 3 years of age, the child learns to be independent and autonomous on the condition that the child is adequately encouraged to explore their world and given the freedom to do so. The third stage, “initiative versus guilt”, usually occurs between ages 3 and 6 and is marked by the child’s further exploring of their world by seeking out and initiating new experiences. The child’s guilt is triggered when unexpected consequences occur as a result of their initiations. Erikson’s fourth stage of development in childhood is called “industry versus inferiority”, and lasts from age 6 to 12. In this stage, children seek to become industrious in all areas of life, from school to interpersonal relations. Mastery of these skills, with adequate support at home and in school, brings about a sense of overall competence, whereas failure brings about a sense of inferiority. Erikson’s fifth stage, “identity vs role confusion” experienced by teens and early adults, focuses on establishing an identity. Teens struggle with the question “Who am I?” This includes questions regarding their appearance, vocational choices and career aspirations, education, relationships, sexuality, political and social views, personality, and interests. Erikson saw this as a period of confusion and experimentation. The culmination of this exploration is a more coherent view of oneself. Those who are unsuccessful at resolving this stage may experience “identity diffusion”, a status characterizing those who have neither explored the options, nor made a commitment to an identity. Those who persist in this identity status may drift aimlessly with little connection to those around them or have little sense of purpose in life.

To foster a new path toward healthy child development, it is necessary to ensure that new placement environments support children in two ways: (1) by addressing the issues they face as a result of their removal from home, and (2) by supporting them through developmental issues that all children face. In seeking to understand the impact of out-of-home care placement on children, it may be hypothesized that a child’s success in each of Erickson’s stages of development must be reaffirmed if not initially accomplished in each new substitutive care environment. From the child’s perspective, each new residential placement comes with all the requirements of learning how to function in a totally new environment, establishing new while securing what remains of old relationships, learning a new set of social expectations and roles. They must conduct their private lives while dealing with public system rules, procedures, and functionaries. New placements often invalidate previous learning and challenge one’s sense of trust and security. To say that residential moves are “stressful” life events is an understatement. If the objective of healthy child development is the prime directive for the design of the child welfare system, then the focus of redesign efforts should seek to remedy what appear to be the negative developmental consequences of system involvement—i.e., the social, behavioral, psychological, and educational effects of repeated care giving failures and disruptions consequent to foster care placement. These efforts should be focused on enabling the child to best deal with the challenges they will face growing up “in care”.

2.3.3. Attachment Disorder and Out-of-Home Care

Secure attachment is evidenced by relationship bonds filled with safety, authenticity, reciprocity, and loving presence. Insecure attachment implies that relationship bonds are entangled with fear and survival states. Attachment disorders are believed to develop from inadequate care-giving. The physical, emotional, and social problems associated with attachment disorders may persist as the child grows older.

Children entering foster care are already exposed to a disruption in their early care-giving relationships. Much of the contemporary literature on ego psychology, object relations theory, developmental psychology, and related disciplines emphasizes the importance of early care-giving relationships for cognitive and affective development, which shapes later interpersonal experiences and emotional well-being [44,45,46,47]. It is generally accepted that loss of or prolonged separation from a parent during childhood or adolescence has immediate and long-term consequences that are dramatic, intense, and complex [48,49]. Numerous studies have investigated the association between early losses and the development of different types of emotional disturbances and psychopathologies [50,51,52,53]. Disruption, stressful life events, and trauma in childhood can also affect biological outcomes such as brain development [54], and increase the risk of medical outcomes such as ischemic heart disease, liver disease, and risk of early death, as well as bio-psycho-social outcomes such as illicit drug use, smoking, unwanted pregnancies, and suicide attempts [55].

The notion that early loss or disruption of relationships has serious long-term consequences for social functioning and adjustment is rooted in psychodynamic theory and observation [56] and attachment theory [57,58]. In the 1940s, psychologists observed negative effects in children who had experienced impoverished maternal contact that they attributed to institutionalization, war-related separations, or other traumatic losses. Their studies showed that early disruptions in maternal care impaired personality development as well as social functioning and adjustment [59,60,61,62,63,64].

Some theorists question whether it is possible to reestablish a healthy trust bond lost in early childhood. As such, there are advocates [65] for parent-involved foster care, noting that if children’s early attachments have not generated trust, children may not have achieved a level of psychological identity that allows them to be comforted by the image of their parent during a time of separation. Some workers may therefore assume these children would be free to form a secure attachment to foster care providers, but she notes this is unlikely. Young children whose parents have abused, neglected, or rejected them have frequently been found to develop dysfunctional attachments to their abusers [66]. They may feel some relief at being placed with better caregivers, which will tend to mask their underlying feelings [67]. However, for example, an anxious attachment is not likely to disappear without help: “physical and emotional distancing promotes, rather than weakens, psychological dependency” [68] (p. 191). Consequently, children whose relationships with their parents have been tenuous or conflicted may be expected by subscribers to this theoretical orientation to have difficulty relating to new caregivers, and be vulnerable to placement breakdown. The move into out-of-home-care has to be seen as an experience that makes the child’s relationship with their parents a conflicted one. From a developmental perspective, when children are in the egocentric stage, (and probably after as well) they would naturally conclude that if they love or care for their new caregiver, then they must not love or care for their original caregiver—it is a betrayal to them. Each child then is met with a host of challenges in relating to and perhaps trusting their new caregivers.

Upon placement, the child (depending on their age) will seek reaffirmation in efforts to develop a sense of trust and establish their autonomy, initiative, and industry. To the extent that the out-of-home care environment accommodates and reinforces such efforts, it will be supportive of quality outcomes. In brief interim placements, the home of origin remains the base for the child’s evaluation of these tasks. As time in placement proceeds however, it is the new substitutive care environment that takes on more importance in the accomplishment of these staged tasks. If the child is continually moved from one care environment to another, then finding trust, validating autonomy, and establishing initiative and industry becomes increasingly difficult. Residential and caregiver stability in out-of-home care supports these endeavors as the new placement becomes a secure base in which the child can establish their autonomy, explore, and master their environment.

It seems reasonable to suggest that the long-term impact of early losses and disruptions may have repercussions on various life domains and that repeated instability within the foster care system may only exacerbate an already damaged and vulnerable child. In fact, Segal, Hines, and Florian [69] found that growing up in out-of-home care in early life was related to lack of placement stability in the adult mentally ill population requiring supervised living arrangements in the community. This finding remained significant even after considering contemporary facility characteristics, demographics, environment, and type of psychopathology.

Investigations that purportedly documented the immediate and long-term consequences of institutional child rearing found deficits in intellectual and social development. These authors highlighted problems in the children’s development of relationships with caregivers and linked these disruptions with later disturbances in their ability to form relationships with peers. Today, this might be viewed as a consequence of a “reactive attachment disorder (RAD)”—a condition found in children who may have had grossly negligent care (including multiple disruptions and forced changes in caregivers) and as a consequence do not form healthy emotional attachments with their primary caregivers. Studies in the 1950s and 1960s also suggested an association between early loss and impairment in sexual identity, development of autonomy, and capacity for intimacy of adults [70,71,72]. However, such studies—in their attribution of the negative effects of neglect and abuse experienced by children in such settings to the institution itself—failed to distinguish between the under-resourced and neglectful group-care settings they observed and the lack of such negative consequences in better resourced and operated institutions. The results of neglect were correctly documented, yet the attribution of cause to group-care upbringing was inaccurate.

In an effort to advance their theories, Wolins [73] notes that many writers when discussing the consequences of institutional care for children “exploited the ‘evidence’” to support the successes of families and the failure of group care. They promoted the adulation of motherhood and families, an optimistic view of the higher classes who employed nannies or had their children reared in boarding schools while the poor had access only to the “orphanage”. Wolins notes the children’s institutions—like the stable family—can also claim responsibility for the moral learning of their members, for the acquisition and knowledge of role performance, and for teaching a member “to be a somebody rather than remain a something [73] (p. 8)”.

Contrary to early reports, the IQ of institutionalized children, their capacity to attach to another person, and their sense of identity need not be adversely affected [74,75,76] in an institution. Temperament, the quality of institutional care, the post-institution environment, and experiences play a role in determining their eventual psychosocial adaptation [74,77,78,79,80].

In fact, the close association of the group-care experience with attachment disorders in out-of-home care policy has masked the more proximal causes of abuse in care. Failure to secure attachment to person place and space, as demonstrated below, is most likely to be precipitated by continuous disruption of a stable life and State-sponsored neglect. The result of the association between group-care and attachment disorder in child welfare policy has been to make “group-care” a dirty word and provide the theoretical justification—to cost cutters—for the demise of many benign, helpful, and stable settings. Perhaps in the necessary search for stability in care, it is time to revisit group-care settings as successful care options.

2.3.4. Stress and Coping: Considering the Effects of Transitions

The stress and coping literature offers some useful theoretical expectations surrounding children and their transitions from one substitutive care setting to another. First, separation from a child’s early caregiver is stressful and possibly traumatic to a young child. This stressful life event can also be preceded by stressor(s) and trauma(s) that challenge children’s basic sense of safety with their primary caregivers. One may therefore expect that the period following placement in foster care would be associated with declines in short-term emotional distress and hyperactivity as well as improvements in the child’s behavior at home and in school. Fanshel and Shinn [81] document this in foster care noting that for those who remain in care, IQ scores (both verbal and nonverbal) at first rose significantly. For younger children who stayed in care for 5 years or more, IQ increases were higher than those who returned home. Barber and Delfabbro [82] add additional support documenting this trend, noting that as children get the opportunity to settle into a new and (hopefully) child-friendly environment, they show functional gains. Children who remain in care for the medium- and long-term sustain short-term gains [82]. This finding challenges some developmental theorists who are skeptical about the development of new attachments [65].

Research findings indicate that children can tolerate some instability as long as it is within certain limits. Given an established U-shaped character of stress and coping response, we should expect that up to a certain point, the child’s coping strategy is challenged and then enhanced as he/she effectively copes with the stress. To illustrate this point, research has found that a greater number of routine moves are associated with a 46% decrease in the rate of behavior-related residential placement change [83]. James et al. [83] interpret this to mean that for children who experience policy-related/routine placement changes, it does not increase their likelihood of experiencing a behavior-related change. The authors consider a stable placement one that lasts for at least 9 months. From about 12 months onwards, however, placement instability is associated with psychosocial deterioration [82]. Children whose placement disruption was attributable to their own behavior were at greatest risk of decline in psychosocial functioning. One might speculate that for these children, each placement disruption may be considered, given the stressful nature of the circumstance, an “Adverse Childhood Experience”, the accumulation of which is believed to have long-term negative health consequences [54,55,84]. These findings are related because if placement instability (the failure to find a stable 9 month placement within the first 12 months) continues to the 12 month point, the likelihood of placement change increases and after two placement breakdowns due to behavior, the future course of placement disruption and psychological deterioration becomes so predictable that the child should be removed from conventional family-based foster care and offered something different [82].

Another concept relevant to stress and out-of-care placement-related transition is the “hardy survivor theory”. Segal, Tracy, and Silverman [85] report that of individuals experiencing trauma significant enough to qualify for a PTSD diagnosis, for a large proportion, the adverse event seemed to contribute to enhanced future social adaptation. Related to this finding is the fact that a significant group of all out-of-home care children make successful adaptations to adulthood. This is true regardless of whether in foster family care or group-care [86]. Searching for the elements distinguishing resilient survivors of out-of-home care, Gonzalez [87] considered adulthood success stories of 18–25-year-olds who aged out of care. The inclusion criteria included older youth (ages 18–25) who had aged out of foster care and were deemed successful. Success was defined as living independently, not being incarcerated or homeless, and participating in an education/training program or active employment. She discovered that coping with the difficulties associated with placement may add to resilience; the difficulty rests with finding out how to foster resilience and the growth of hardy survivors. In considering how to foster growth and resilience in children placed in out-of-home care, it becomes important to consider through a stress diathesis lens at what point the level of stress, combined with the tools of the child, becomes too stressful for children and stressors such as loss are no longer met with effective coping strategies.

2.3.5. Deviant Identities, the Looking Glass Self, and Degradation Ceremonies: Sociological Perspectives on Out-of-Home Care Experiences

According to sociological theory, failure in life is synonymous with the development of a deviant identity. First, self-consciousness involves continually monitoring the “self” from the point of view of others. As Cooley [88] put it, we “live in the minds of others without knowing it”. Second, living in the minds of others, imaginatively, gives rise to real and intensely powerful emotions such as pride or shame. Given this theoretical framework, does the change in primary care-giver and repeated changes convey a message of failure and place the child in a deviant identity? One of the primary predictors of placement change is acting out behavior, yet we understand little of this as a cause or consequence of such changes. The change is made based on the behavior and the message to the child is: you are different, or even “you are deviant”. The more this message is conveyed, the more reasonable self-identification with it becomes.

The placement process in out-of-home care is akin to what Garfinkel [89] might call a degradation ceremony, whereby “…the public identity of an actor is transformed into something looked on as lower in the local scheme of social types”. Children are routinely processed through shelters and temporary foster placements before being taken to more permanent facilities, an experience that is objectifying and can be likened to the degradation ceremony. Children are told little about where and when they are moving and what they will need in the transition [90]. They are not treated as participants in the choices made about their lives. There also may be no choice in the change and parents have little positive involvement. A British–Canadian study showed low levels of parental inclusiveness—involvement of birth parents in the placement process and helping children with their feelings about living apart from their families [91]. Another Canadian study, reporting children’s responses, had similar findings: only 14% were accompanied by parents to the initial placement; only 16% had the separation explained to them; and only 34% had definite visiting arrangements [92]. Failure in placement reinitiates the processing. The integrity of the identity of any child going through such a process is necessarily challenged.

2.3.6. Organizational Theory and Out-of-Home Care

Out-of-home care offers the child many organizational contexts arrayed from multiple structural variations of the “family” (including that provided by kin-care) to varieties of group-care settings. Each in idealized form is perfect and each poses great challenges to growth and development when it is under-resourced and dysfunctional. In a bureaucracy, the way to destroy a program is to repetitively move it from one department to another and focus it on internal reorganization as the consequence of each move. By comparison, the way to destroy a productive individual is to place the person in a position of powerlessness from which they cannot exit and to subject them to perceived nonsensical tasks that will have no positive outcomes even if performed to perfection [93]. The analogy to the situation of the child in out-of-home care is almost perfect. A child is moved from one situation to another, never given enough time or resources to gain traction in any setting. The child is also constantly focused on his internal psychology when the world around him demands the investment of energies in dealing with external realities—school, peers, socialization, etc. The child is occupied with therapeutic interactions that offer conceptual if not practical solutions to his/her external demands even if he/she masters these emotional and conceptual challenges with perfection. The child is viewed as irreparably damaged at the outset of entry into out-of-home care [16]. Yet, despite all of these challenges the child adapts, like the bureaucrat, and if given the resources and opportunity in many situations, rebounds.

The administrative structure of an orphanage can and did affect outcomes of such care. A hierarchical, authoritarian administrative structure in the orphanage can result in poor psychological functioning in children, whereas an egalitarian structure can enhance functioning [80]. Recent research has revealed some other potential mitigating factors. Academic or athletic competence, or the presence of an interested, caring, involved adult during adolescence, and the presence of a stable mate in early adulthood can contribute to positive adaptation in adulthood [78,94,95,96]. Maturity of defenses may also play a role [97,98].

2.3.7. Social Structure and Opportunity

Social structure defines and sets limits on opportunities. Those with natural abilities blocked by structural barriers will make rational if sometimes deviant adaptations to cope with their situations [99]. The five children followed through middle school on the HBO television series The Wire present one of the best dynamic views of the urban underclass challenge to children growing up in abusive and neglectful situations —it is the height of urban sociology. All five children were candidates for out-of-home care and the outcomes of their efforts that result in each of them finding a “home” and vocational niche are illustrative of the overwhelming influence of social structure in the developmental efforts of children to find opportunity [100].

Ultimately, children are the products of their environmental opportunity (c.f., p. 30 below outlining the situation of the Quebec orphans). The psychological focus of out-of-home care on “severe emotional disturbance” (SED) and attachment disorder has minimized the importance of social structure obstacles to achievement and the legitimate anger and rational resistance to being abused by the system that results from such anger and are significant determinants of problem behavior in children. The attribution of behavioral issues to psychological disorder frequently facilitates blaming the victim. In coping with structural issues out of their control, children either turn their anger at their powerlessness inward or externalize [99,101]. While no such study exists in the child welfare literature, Segal, Watson, and Goldfinger [102] observed that half of the behavioral aggression used to determine that a person should be considered a “danger to others” due to a mental disorder for purposes of involuntary commitment to a psychiatric hospital was precipitated by staff action. During their study of almost 700 patient evaluations at ten psychiatric emergency rooms in California, general hospitals’ staff were observed to be equally as likely as patients to be the first to “lay hands on”. Such staff action violated the personal space of the patient as well as leaving the person in a state of powerlessness over their own fate. The patient then struck out at the staff making the patient a “danger to others”.

Much of the microscopic focus of today’s out-of-home care psychology is nothing more than blaming the victim for the failure of the system. The responsibility of “parens patriae”, the State in the role of the parent, is to open the structural opportunities necessary for adult achievement by their charges. Today’s out-of-home care population suffers from under parenting and over medication [103]. This situation is to some extent brought on by over attribution of behavioral problems to psychological disorder and the assumption that these problems are amenable to being fixed the easy way, with medication. Reliance on the “pill” as the solution in most situations does not address the child’s structural reality nor does it bode well for his/her long-term achievement.

3. From “Family-First” to “Home-First”

Children are placed in out-of-home care because they have lost their family, have been given up by their family, and/or been abused or neglected by their family. Why then should family be first?

Alfred Kadushin [104], a prominent scholar in the child welfare field for at least fifty years, charges that professionals in the child-care field “… are victims of their own propaganda …. We are family chauvinists …. The tendency is to compare institutional care with an idealized version of the foster family. It needs to be remembered that foster families are not highly selected…that there is a high rate of turnover of foster families making for discontinuity of care. [Research indicates 30% to 50% of foster families left the program within their first year [20]] It needs to be remembered, also, that a structurally intact, but emotionally broken family may be more pathogenic than institutional placement [104] (p. 170)”. Why not, in the best interests of the child, choose “Home-First”?

3.1. Stability in Out-of-Home Care and Empirical Outcomes

Given the above noted theoretical perspectives, it would appear that stability should be a highly valued objective in child placement and a basis for ensuring access to opportunity through the development of social margin. This section evaluates this thesis by reviewing the relationship between stability and child outcomes in the out-of-home care research. Emphasis is placed on indicators of the achievement of positive personal values and direction, a sense of self-worth, education, skill, and vocational development—the things that seem associated with success. In examining the extensive literature on child out-of-home care, stability, and child outcomes, three questions are of most concern: (1) Is there an empirical association between stability and outcome? (2) If so, how can stability in out-of-home care be promoted; what are those factors that facilitate and hinder it? (3) Is group-care a viable Home First placement that might serve to insure stability? The following section explores these important relationships.

3.2. The Association of Stability with Outcome

The Casey Family Programs [33], as noted above, investigated foster care alumni and found a significant relationship between placement stability and high school graduation—a key determinant of lifetime outcomes. In a multivariate analysis, having an average of one fewer placements per year than others in the sample made high school graduation 1.8 times more likely, two fewer placements made high school graduation 3.1 times more likely whereas one more placement per year made high school graduation 43% less likely and two more placements per year 66% less likely. This was one of the largest consequences reported on in a sample of 1609 alumni served by 23 Casey Field offices between 1966 and 1998.

In one of the largest studies conducted (n = 34,600), two to five years following emancipation, it was found that “multiple placements” is the characteristic most associated with negative outcomes for youths in out-of-home care. Of 14 variables associated with outcomes, multiple placements were negatively associated with 12 and neutral for 2. Lack of stability was associated with increased unemployment, school dropout, relationship troubles, and teen parenthood [105,106].

Other studies confirm these associations though often noting the significance of various timeframes, gender, age (at time of placement), circumstance, and behavior problems as possible mediators or moderators in the relationship between instability and outcome. Overall, it appears these studies replicate the negative impacts of instability showing how it manifests itself among different subgroups in out-of-home care. Considering the timeframe of placement, it was found that children with more than one placement move during their 1st year in care are more likely to experience placement instability in long-term care than if they did not move or moved only once during their 1st year (n = 5557) [107]. Female life satisfaction is negatively related to the number of moves, and positively related to length of stay in current residence, though no such relationship was observed for males [108]. Placement instability is associated with increased risk of delinquency for male foster children, though not for female foster children (n = 415) [109]. Children first placed between ages 12–15, with multiple placements and multiple spells in care, and children with multiple placements supervised by probation, had a higher risk for incarceration for a serious or violent offence during adolescence (n = 79,139) [110]. Multiple placements and emancipation from group-care, rather than family situations, are associated with incarceration in young adulthood [111]. Frequent moves are associated with weaker foster family relationships and poorer school achievement, though differentially for boys and girls. For girls, there is an indirect effect related through poorer school achievement and investment, and for boys, placement movement is related to difficulty forming strong relationships with foster parents (n = 199) [112].

More recent studies [113,114,115], as noted above, continue to show more frequent and numerous changes in placements are associated with increases in negative outcomes including delinquency, drug use, low self-esteem, and poor socio-emotional competency and mental health functioning.

While these remain an impressive array of negative associations with instability, they need to be considered with caution since, with the exception of those results predicting adult behavior, the association may be from the negative behavior to the instability rather than the opposite direction. Thus, instability may be the most common result of getting into trouble rather than the precipitant of such trouble, or may be interactive whereby trouble creates instability, which in turn creates more trouble and instability. Rubin et al. [116] considered these issues. Their study sought to disentangle this cascading relationship in order to identify the independent impact of placement stability on behavioral outcomes downstream. They considered placement stability over the first 18 months in out-of-home care for 729 children from the National Survey of Child and Adolescent Well-being. They categorized as “early stability” (stable placement within 45 days), late stability (stable placement beyond 45 days), or unstable (never achieving stability). Propensity scores predicting placement instability based on baseline attributes were divided into risk categories and added to a logistic regression model to examine the independent association between placement stability and behavioral well-being using the Child Behavior Checklist and temperament scores from the National Longitudinal Survey of Youth. Half (52%) of the children achieved early stability, 19% achieved later stability, and 28% remained unstable. Early stabilizers were more likely to be young, have normal baseline behavior, have no prior history with child welfare, and have birth parents without mental health problems. Unstable children were more likely to have behavior problems than children who achieved early stability across every level of risk for instability. Among low-risk children, the probability of behavioral problems among early stabilizers was 22%, compared to 36% among unstable children, showing a 63% increase in behavior problems due to instability alone. After accounting for baseline attributes, stability remained an important predictor of well-being at 18 months. This finding establishes the independent contribution of instability to the negative outcome. It would thus appear that children in foster care experience placement instability unrelated to their baseline problems, and this instability has a significant impact on their behavioral well-being. This finding would support the development of interventions that promote placement stability as a means to improve outcomes among youth entering care.

3.3. Factors Promoting and Detracting from Stability

Placement change and the resulting instability can be attributed to child welfare system-initiated factors between 50% and 70% of the time [21,83,117]. While some of this instability may be due to the child’s behavior, instability precipitates behavioral reactions. It would appear, at least according to the cited reports, however, that lack of stability is primarily a structural problem and only to a lesser degree can it be attributed to the problems of individual children and their families. A majority of studies attempting to specify factors that are associated with placement stability, therefore, must be considered with care as these studies often do not include in their analyses context and/or system factors nor do they consider interactions between child behavior and service setting characteristics. In looking at factors associated with facilitating or hindering stability of placement, we must first consider moves in and out of the system then moves within the system.

3.3.1. Stability and Factors Associated with Reunification and Re-Entry into the System

Reunification. Consistent with the experience of other studies noted above, Wells and Guo [24] studied the outcomes of first foster care placements (n = 2616) and found that by the end of their 4 year-3-month study period, 39% (1016 out of 2616) were reunified, 32% were placed with guardians, and 9% had other outcomes such as adoption or reaching age 18. Twenty-one percent were still in foster care. Herein considered are those factors likely to predict slower reunification and thus longer tenure in the system as well as returning and/or recycling in a revolving door pattern—the “slow re-unifiers” perhaps indicating those who might best achieve stability within the system and the “returning” and/or “recycling” describing those with defined unstable childhoods.

A few consistent relationships have been found with reunification, movements out of the system. Being in kinship-care [118,119], the presence of a child health problem or disability [120,121,122], and the number of placements or moves while in care [118,119] have all been associated with slower rates of reunification [24]. These variables along with others will be considered in greater detail below.

Courtney [120] found that African American children were reunified 39.8% more slowly than Caucasian children. Children with health problems were reunified 39.8% more slowly than children without such problems. Children who lived with their mother only were reunified 32.9% more slowly than children who lived with both parents. Children who lived with those classified as “others” were reunified 48.1% more slowly than those who lived with both parents. Children in custody because of neglect or dependency were reunified more slowly than children in custody because of physical abuse (26.2% slower for neglect; 23.3% slower for dependency). Children placed in hospitals were reunified 32.6% more slowly than children placed in kinship homes.

Thus, it would seem that resources and threat to child safety via health and violence matter most in reunification aside from the aforementioned structural/system factors.

Re-entry. Courtney [120,121], studying reunified foster children in California, found several groups of children had faster rates of reentry than their comparison groups: children who had health problems, children who were African American, children who were from families receiving Aid to Dependent Children, children who spent three months or less in care, children who were placed in non-relative care, and children who had a higher number of placements during their first spell in care. He also found children 7–12 years old had slower rates of reentry than the other younger and older age groups studied.

Wells and Guo [24] found that among those who were reunified within 24 months of first entry into foster care, 24% (225/933) reentered foster care by the end of the study period. The following six variables were related to the hazard rate for reentry in the following way: With respect to child’s age, a l year increase in age at exit increased by 9.7% the rate at which a child reentered care. With respect to child’s ethnicity, an African American child reentered at a rate that is 97.9% faster than a Caucasian child. With respect to the reason for a child’s placement, a child in custody because of physical abuse reentered at a rate that is 70.9% slower than a child placed because of dependency. With respect to the number of moves made in the first spell of foster care, an increase in the number of moves increased the rate of reentry by 30.5%. In contrast, an increase in the number of months a child spends in care decreases the rate by 5.1%. With respect to type of last placement, a child in non-relative foster care re-entered at a rate 226% faster than a child whose last placement was kinship foster care; and a child whose last placement was a group home reentered at a rate 232% faster than a child whose last placement was kinship foster care [24]. The high hazard rates for reentry attributable to the latter two situations may reflect selection issues and/or unspecified system characteristics such as the readier availability of such placements for needed reentry. Such structural issues are not given enough attention in these studies. Despite this problem, it would still appear that instability in one’s initial placement history is among a myriad of influences on reunification failure and starts the revolving door. It is consistently related to reentry and recycling.

Though studies of reentry have involved different geographic locations and sampling parameters, most estimated reentry rates are at 20% to 40% within 1–5 years [123,124,125]. A review of the reentry literature [126] identified several risk factors at the child level (age; race; mental, physical, or behavioral problems), family level (poverty; parental substance abuse; lack of support; maltreatment type), and service level (number of placements; prior child protective services involvement).

More recent studies have found parents’ substance abuse and mental health problems (SAMH) have steadily increased as a reason for removal: in 2016, over a third of child removals (foster care entries) involved parental substance abuse [127]. Font et al. [36] found that among children who reunified within 12 months, the hazard of reentry was twice as high among those removed for SAMH compared to those removed for no neglect, and 43% higher compared to those removed for neglect without SAMH. They note that reunification after 12 months was associated with increased reentry risk overall, though permanency guidelines that restrict the length of time to achieve reunification may have the unintended consequence of pushing reunification before maltreatment risks have been resolved thus starting the revolving door moving.

3.3.2. Stability Factors and Movement within the Out-of-Home Care System

As previously described above, approximately 20% of children entering in each cohort will remain in the system, either never leaving care until they age-out or experiencing a failed reunification. While several factors have been associated with stability in foster care, six characteristics seem most salient in that they are likely to be both the cause and consequence of instability and interact with it to exacerbate the situation.

First, as previously mentioned, are conduct and psychological dysfunction issues. Being a teenager with a conduct disorder is an important variable related to likelihood of placement breakdown [128,129,130,131]. A study investigating placement disruptions in treatment foster care reported that the likelihood of disruption for a sample of emotionally and behaviorally disordered youth was two times higher during their first 6 months in care compared to the second 6 months [132].

James et al. [83] point to the significance of externalizing behavior problems as the main distinguishing predictor in understanding patterns of stability and instability in care. The odds of experiencing delayed entries into stable placements, late disruptions, and multiple short stays in care increased among their sample with progressively higher levels of externalizing behaviors. This finding is consistent with previous studies that have reported a link between disruptive behaviors and placement instability [65,133,134,135,136,137]. While this study was unable to shed light on the causal direction of behavior problems and unstable patterns of movement, as noted above, Newton and colleagues [109] found that behavior problems were not only a predictor, but also an outcome of multiple placement changes—that placement disruptions are not only precipitated by behavioral problems but cause them, further propelling the foster child toward increasingly unstable patterns of placement movement.

The second characteristic associated with within-system movement might be termed an issue of damaged identity. Webster et al. [107]—in a longitudinal study specifically focusing on placement stability—found that placement changes occurring during the first year in care seem to put children on a trajectory toward further instability [107]. If the issue is labeling and recycling in the system becomes a self-fulfilling-prophesy, then history should prevail in placement breakdown. Having experienced breakdown in a former placement does increase the failure risk in future placements [6,138,139,140], though in a large-scale study, Fratter et al. [141] found that history of placement disruption was not associated with placement breakdown once other variables were controlled. This latter finding indicates a need for more investigation of the labeling/damaged-identity hypothesis.

The third characteristic hypothesized to affect within system movement relates to the mission/function of the setting. The function of various facilities within the out-of-home care system by definition is confounded with stability. Some facilities are the recipients of children who cannot adjust to other settings while others are transitional in nature and still others are considered terminal placements.

A Swedish study looked at premature placement breakdown for children in out-of-home care. Risk factors in relation to breakdown were analyzed in the four main forms of Swedish out-of-home care separately (foster homes, privately/publicly run residential care, secure units). Analyses point out different risk factors within each placement context; the only consistent risk factor is antisocial behavior at time of placement. The study found that the lowest rates of breakdown were in kinship care and secure units—the former a terminal placement and the latter to some extent a placement of last resort. The highest rates were in non-relative foster homes (n = 776) [142]. The investigators offer an interesting table naming several different risk/protective factors (gender, immigrant, run-away/throw-away, abuse, mental health, antisocial behavior, breakdown of earlier placement, relationship problems, assessed by residential home specialized in assessment, voluntary/court order, distance from home to care setting) for placement breakdown for each type of placement (foster home, private residential care, public residential care, secure units). Risks/protective factors change across type of placement (see [142] p. 149), leading one to conclude that selection of children for given types of placement, as well as other system procedures having an unspecified influence on results.

The fourth characteristic hypothesized to affect within-system movement relates to issues surrounding “settling in” or dependencies. Stability may be a double-edged sword—it provides opportunity to maximize the use of one’s resources and build social margin to advance one’s development and position in the context in which one is involved. It provides the ability to develop an understanding of how to navigate the social norms of the social system that one is imbedded in. Most importantly, it enables the establishment of long-term relationships with peers that may be supportive throughout a life time—and perhaps may be more significant than those established with adult care givers who frequently come and go in group-care settings while peers remain. Settling in to such settings, however, can lead to dependencies. Such dependencies may occur either due to a lack of experience with alternatives—a situation that can more readily be overcome—or in poorly designed environments, a learned helplessness. In either situation, transitions to an adult lifestyle out of quality or poor placements are difficult. This is illustrated in an Israeli study that examined the retrospective reports of alumni of Israeli group homes on their experiences of leaving care. A sample of 94 alumni (38 men and 56 women) revealed major difficulties associated with the transition from their group homes. About half perceived their transition from care as “quite hard” or “very hard”. A fifth reported having no one to talk to during the first period of leaving care. Generally, the transition was more difficult for girls than boys. The longer the alumni (especially boys) stayed in care, the more difficult they found the transition to independent living. Better relationships between the child and the group-home parents while in care were associated with a more difficult transition to independent living. This study highlights the need for longer and more extensive preparation for leaving out-of-home care [143] as well as the need to understand the nature of the “settling in” experience that influences the transition. Segal and Moyles [144] in a study of 215 residential care facilities have determined that the “settling in” effect is moderated and its negativity vitiated by the management style of the facility. The effect appears to be one that promotes learned helpfulness in management-centered facilities and independent evaluative skill if not transition experience in client-centered settings. Alternatively, learning how to use the relationships established over the years and building in supports for the transition into the group setting can lead to successful long-term relationships and life success. This is evident in the experience of the Pride of Judea Children’s home, a facility open from 1923–1958, “where of 2006, alumni of Pride of Judea still met regularly to reminisce about their times at the home” [145].

The fifth characteristic affecting within system movement is the residential format—notably the structural mission and limitations as defined by the system of group-care vs. non-relative foster family care or kinship care. It would seem that stability in a particular type of placement is most dependent on the system within which the placement in found. Knapp, Baines, and Fenyo [146] found that once in care, the average “expected” length of time in care was longer for those initially placed in a group-care than those in a foster home. Wulczyn et al. [147] (n = 16,170) analyzed movements of children in the child welfare system. They provide a chart showing the “number of children first placed in group-care by age and number of movements.” The chart shows that the vast majority of children first placed in group-care have zero movements (3519 of 4100) as compared to foster care (11,125 of 16,170). The chart is informative in that it is only the first placement that divides the groups and only about 100 of the children placed in group-care were under the age of 8. This could indicate that children over 8 may successfully settle into group-care once initially placed there.

Wells and Whittington [148], however, found children in group-home or residential care settings are reported to have experienced more volatile placement histories. Perhaps this was because within the system they studied, in the United Kingdom, failure in foster care is likely to be followed by group-home placement. In the United Kingdom, high levels of instability of placements are reported, as well as differences between children placed in group-care and foster care. Children who enter foster care before reaching teenage years have a degree of stability, but older children have many more placement changes and are much more likely to be in group-care accommodation (n = 848) [149].

Other investigations have shown that other residential formats impact the child’s out-of-home care stability experience. It has, for instance, been found that children in kinship care are more likely to experience stability when compared to children in other types of out-of-home care arrangements [150,151,152]. With each day spent in kinship care, a child’s hazard of experiencing a behavior-related placement change is reportedly reduced by 1% (n = 580) [153].

The new wave of priority placement being sold In today’s market, based on these types of findings, is kinship foster care. If placement disruption, however, is the criterion against which this placement type is evaluated, the initial impression of kinship-foster-care placement as noted above appears consistent with the strong version of kin altruism—i.e., it endures with a greater degree of stability than non-kin foster care. However, looking longitudinally at kin vs. non-kin care, Testa [154] found that the initial stability advantage of the former seems to weaken the longer the placement lasts. The advantage vanishes in between 2.7 and 3.1 years of care. This study’s results are based on a multi-year, multi-cohort assessment with almost thirty thousand records, and suggest that the stability of non-kin placements approach the stability of kin placements after the third year of care. There are many possible explanations other than altruism for the initial advantage of kin-care as there are for its disappearance, the fact that the stability disadvantage of non-kin placements disappears in a period that is less than a third of the way into the average child’s out-of-home care experience is of concern in offering preference to this pattern of care especially for those likely to need long-term out-of-home care placement.

The sixth characteristic affecting movement has been described as the “placement career”. Looking at the total placement experience within the out-of-home care system the research literature most frequently defines a child’s stability by the “number of moves” experienced and “the duration of residence”—how long the placement lasts. When considering movement in and out of the system or when considering a revolving door pattern of care, researchers measure entries and exits, “recycling into the system” and the “duration of home-of-origin reunification”. Investigators have also tried to find consistent patterns of care, patient careers [83,155].

The child welfare bureaucracy moves children through transitional facilities that in the past were not (by the system) considered “real placements”. More recently, they are counted as congregate placements [21]. A child’s history with these placements and with those considered to be “permanent” varies considerably in duration and pattern across time. This history is the child’s “career” in the system. James et al. [83] considered a sample of 430 children during their first 18 months of care and defining stability as a placement lasting at least nine months tried to rationalize child placement into four care patterns:

(1)

Early stability, those who achieved placement stability within 45 days of entering care,

(2)

Later stability, those who achieved placement stability between 46 days and 9 months of entering care,

(3)

Variable pattern, those who had both stability and instability; stability being that one placement lasted at least 9 months, and

(4)

Unstable pattern, those who experienced multiple placements, none lasting 9 months.

The authors found that 36% of children experienced early stability, 29% of children experienced later stability, 16% of children experienced a variable pattern, and 20% of children experienced instability across the 18 months they were followed. The latter finding validates the need for a stable alternative placement, a Home First for at least a fifth of the children if they are to avoid a year and a half of disrupted life on top of the trauma precipitating placement and the stress of placement itself.

In looking for a placement that might provide a Home First and ensure stability, it would be important to know if any of the current forms of care are particularly vulnerable to disruption and/or offer a greater probability of stability. The James et al. [83] study breaks their patterns of care into those placements (foster, kin, or residential/group-care) where the children achieved stability. Of the 19.8% of children in the unstable pattern, 11.2% had episodes in group-care while 8.6% never did. The authors explain that instability is not only associated with stays in group-care as commonly perceived. They note that while children with the most unstable placement patterns in this cohort clearly experienced the highest number of episodes in group-care, these children generally started their placement history in family-based settings, and, following an episode in group-care, were often ‘stepped-down’ again to family settings. They point out that 37 out of the 85 children (43.5%) meeting the criteria for an “Unstable Pattern” spent no time in group-care. They simply moved back and forth between different family-based settings, never achieving stability. These children tended to be younger, and thus reluctance by the child welfare system to place them into group-care settings might be understandable. However, the question needs to be raised as to whether the volatility of their placement history could have been averted if they had been placed into group-care early on. This is particularly so, since even when considered by most a placement for the more troubled children and thought to be the repository of such children, group-care did not seem to contribute to their instability. The results suggest that it may simply not be possible for some children to be sustained in family-based settings [83].

States seem to have trouble recruiting sufficient numbers of family-based settings [21]. Craft [156] indicates that sometimes “…placements just don’t work and disruption, or asking that a foster child [be] moved [is]… not a matter of if it happens, but when, because disruptions do occur, even for the most experienced and tolerant of foster homes. Foster parents choose disruption of a foster placement for a number of reasons and not all have to do with the child’s behavior.” Craft [156] offers four such reasons having nothing to do with the child’s behavior.

• Sometimes the foster parents need time to refocus and have the child moved to another foster home.
• Sometimes the foster family needs to relocate—e.g., If they can remain close to the foster care agency, they are usually allowed to maintain the placement of the foster child, but if the family is moving out of state, then the placement is disrupted and the child is moved to a new foster home.
• The foster parent or someone within the family is sick or there is a death within the foster home.
• The birth family, foster family, or other team members are at odds and it is better for the case if the child is moved to a new foster home.

In attempting to find “career patterns” in child placement of 474 foster youth, Havlicek [155] cluster analyzed their 4323 placement events occurring between 1984 to 2003. Five clusters emerged:

“The Early Entry” cluster (13.5%) had the longest time spent in foster care. They entered foster care at age 3.4 and spent 13 years in care, with a relatively low rate of movement (0.8 times per year). This rate reflects just under one change per year for the 13 years in care. These youth spent the longest time in Regular Foster Care settings (66.9 months), although they also spent considerable time in Treatment Foster Care (61.7 months) and Congregate Care (17.4 months).

“The Settled with Kin” cluster (25%), distinguished by placement with relatives, had the lowest rate of movement (0.6 moves per year), though not significantly different from the Early Entry cluster.

“The Late Movers” (28%) first entered foster care at age 11.9 and spent 3.4 years in care. They had the shortest tenure in care and the highest number of moves, 2.2 per year (significantly higher than other clusters). These youth were estimated to have spent almost equal lengths of time, on average, in Relative Care (10.4 months), Congregate Care (10.3 months), and Regular Foster Care (9.9 months). This finding suggests that foster youth in this pattern did not settle into any one placement setting.

“The Institutionalized” cluster (16%) first entered care at age 6.7, spent 9.5 years in care, and had 1.2 changes per year. They spent the majority of their time in “Congregate Care” settings (63.2 months, or over 5 years)—more than three times as long as youth in any of the other cluster.

“The Community Care” cluster (17%) first entered foster care at age 7.5, spent 8.5 years in care, the majority of time moving in and out of Regular Foster Care settings. They had the second-highest rate of movement; though not statistically different from the Institutionalized cluster. They experienced 11 different placements and 1.4 moves per year. They experienced movement across several types of placement settings, spending time in Regular Foster Care (44.1 months), Relative Care (26.5 months), Treatment Foster Care settings (16.6 months), and Congregate Care settings (8.1 months).

Untangling the web of understanding placement moves and rationalizing system decision-making seems to have produced a quagmire of misunderstanding based upon the objectification of the child’s experience, abstract research reality, a misunderstanding of the child’s reality, and the misperceptions of foster family advocates/group-care critics. The latter being that the use of group-care should be limited to only the most troubled cases and if possible should not be used at all. Simply reading James et al.’s [83] post-hoc structuring of a child’s experience in moving into and through the out-of-home care system should give one a sense that entering the system must be terrifying. The system’s objectification of the child’s experience seems appalling—the very idea of accepting a classification of “early stability” involving a 45-day delay for getting into a stable residence should be intolerable. The notion that stability is defined as 9 months in a placement is absurd. One might argue that a school year is nine months and since the child’s reality often revolves around school this would be a reasonable stability indicator. Yet, school relationships are built over years and peer friendships span school years. Learning and skill acquisition are dependent on continuity of curriculum and cumulative skill acquisition. Furthermore, there is no indication that the nine-month period defining stability is coterminous with the child’s school year. If it is not, the child has had two disrupted school years. Meanwhile, there is an ambivalence within the foster care system to acknowledge that many family homes are not prepared to deal with the level of attention that children need, thus setting children up for placement breakdown and solidifying the message that they are “too much”. There is most importantly a reluctance to acknowledge, as is evident in Havlicek’s [155] Community Care cluster, that many family homes cannot guarantee long-term stability.

3.3.3. Iatrogenic Instability in the Child Welfare System

Both in efforts to move children into and out of substitutive care as well as to move them between facilities while they are in the system, child welfare bureaucracy and priorities—notably permanency planning, conceptual myths—such as the continuum of care and levels of care, and misguided views of the impact of facility types on child outcomes foster placement instability.

Bureaucracy. While age, behavioral difficulties, and absence of mental health support may contribute to instability of child placement, the majority of placement changes are transitions, shifting the child from one temporary placement to another. The change rationale may be because some placements are believed to be transitional—e.g., a “step-down” or “step up” move—or possess a greater degree of “permanence”—e.g., a change from group to non-kin foster care. Such change satisfies system conceptualizations of setting functions, priorities, requirements, etc. Ward and Skuse (n = 249; 2001) found that planned change accounted for 54% of transitions, often from temporary or emergency shelter to more secure care. Similarly, James et al. [83,157] (n = 580) examined reasons for placement changes in foster care in San Diego and found that while 20% of all changes are behavior related, 70% of changes were system or policy-related. The most common moves were a change of placement to short-term facilities (29.3%, from a shelter to short-term foster home) next to long-term facilities (24%, to long-term foster home) and to a relative (15.3%). The actual placement numbers for this cohort within an 18-month period ranged from 1 to 15. Only 14 (3.3%) children in this cohort experienced one placement. Even children in the defined category labeled “Early Stability” experienced on average 2.5 placements. San Diego County procedures required at the time that children were first placed into shelter care and then into emergency foster care settings before finally moving them to their intended placement site. If a placement fails, this event precipitates a move back to a shelter before another placement setting can be identified. Given these studies, it would seem that the child welfare system promotes significant placement instability to facilitate bureaucratic efficiency via central entry points into the system, thus facilitating more efficient assessment of children’s immediate needs. Whether such assessments actually have a positive impact or are simply people-processing is an open question. The fact that James et al. [83] thought it necessary to comment that “many child welfare professionals would not regard the early moves as placements” is perhaps tragic evidence of the objectifying nature of the system.

Permanency planning. The mandate for permanency planning is confounded with the objective of achieving stability in a child’s placement—a confounding of means and ends. More than sixty years ago, Maas and Engler [32] found that children in “temporary” foster care often remained there for long periods, neither returning to their families nor into a stable alternative—these children found themselves in “foster care drift” [158]. The current assumption is that children will be harmed if they are subjected to temporary or unstable living arrangements other than for very short periods of time during which a crisis is either resolved or averted. This reasoning, accompanied with the finding of “foster care drift”, has led to an emphasis on permanency planning—a remedy emphasizing the development as soon as possible of plans for speedy return to the family of origin or termination of parental rights via adoption, i.e., if the former proves impossible within a reasonable timeframe. To some extent, permanency planning and achieving stability, at least in the eyes of some [82], are viewed as one and the same phenomenon. This confounding of means and ends however assumes that permanency planning, as previously defined, will ensure stability in the child’s life. In fact, such planning requirements seem to place pressure on the family to take a child back prematurely or put pressure on social workers to accept less than desirable “permanent” placements inducing instability. While there is no direct evidence of this phenomena in the child placement process, Ahart et al. [159] in their report on reunification programs add credence to this observation noting: “Few reunification services [are] available generally. The first failure [in the process] involves the largest group of children, those who exit foster care quickly after placement and are returned home to their biological families. While this group represents the preferred foster care outcome, it also points to a major service gap. During the time that children are out of the home, their families often receive inadequate support services to remedy the problems that may have precipitated the initial placement. In addition, families rarely receive assistance with the often-traumatic reunion when their children are returned home. It is widely believed that this lack of effective support services is partly responsible for a consistently large proportion of children re-entering the foster care system when family reunification attempts fail”.

Further, the new legal time-limit requirements introduced under FFPS [21] changes the practice decision from one based on practice principle—“in the best interest of the child” in the foster-care placement process—to one based on prescriptive mandate—carry out the placement by a given date or face the consequences (usually a negative statistic that may have negative funding consequences). This legislation leads to a displacement of goals. Instead of doing the best for the child, the social worker is meeting a deadline for organizational protection. The objective of providing stability through permanency planning is often lost in bureaucratic and procedural issues leaving the child cycling between short-term placements.

The equation of permanency planning with achieving stability distracts from the focus on the importance of stability in the life of the child and from understanding the circumstances under which stability may be achieved in out-of-home care. It might be argued that stability was achieved in the past where foster care or institutional care drift occurred and by default children grew up in a single care setting. Given this review, better procedures for ensuring stability in the life of children in out-of-home care are needed as well as a better understanding of the type of environments that can best meet this objective for those needing out-of-home care.

Continuum of care. There are two continua of care in the United States out-of-home care system. The first is a system-processing continuum based on expected duration of stay—shelters, temporary foster care, non-relative foster care, and kinship care. The second is one theoretically based on functional requirements of the child. Placements are aligned on a continuum of living situations that provide at baseline a more normative environment mimicking the family situation and step up into more specialized care—usually group-care settings. These latter facilities are designed to focus on and meet the needs of children deemed uncontrollable in the baseline setting. The family and usually the nuclear family setting are believed to foster the healthiest context for growth and development. In fact, the National Conference of State Legislators [20] has noted that the 2018 federal Family First Prevention Services Act was based on “research”—that being a survey of state legislators asking for their opinions as to whether family or congregate care were better placements. It thus becomes the mandate of the child welfare worker to move the child toward a family placement even if such moves disrupt a stable placement.

Movements through the system seem justified by the continuum mandates rather than child need. James et al. [83] described children they classified into their Variable Pattern group (16.0%) as having at least one placement that lasted 9 months—half the time of their data collection period. Some seemed to have achieved stability in a non-relative foster care home but were then moved to the home of a relative. Children with the most unstable placement patterns generally started their placement history in family-based settings, and, following an episode in group-care, were often ‘stepped-down’ again to such settings. All such placement changes are disruptive of stability--those involving system processing are objectifying, the step-ups and step-downs are unclearly specified as being in the best interests of the child.

Misguided views of the impact of facility types on child outcomes. James et al.’s [83] study points to the importance of stability in placement and the role of developing social margin for future success. Stability is a necessary but may not be a sufficient condition to justify a choice of a particular type of placement. The role of different placement types and strategies needs to be better understood. Herein are addressed some of the misguided views of current placement options.

Normative group-care vs. foster family care. The Director of San Francisco’s Human Services Agency in the 1970s endorsed foster family care or adoption exclusively rather than allowing for an “institutionalized care option” in group homes such as Edgewood Center. Edgewood was a facility that had been in existence since 1851 with a record of success [160]. While in the 1970s, it became a placement for more psychologically troubled youth and as such became quite expensive, in his proposal for budget cuts, the Director pointed out that foster care and adoptive homes are better options for kids than group homes because they are less expensive and provide more stability [161]. While perhaps true for those obtaining rapid adoption, and true with respect to enriched treatment facilities, the illusion that foster family care provides greater stability and is more effective seems largely unsupported in the literature (see for example, Havlicek’ [155] Community Care vs. Institutional clusters that did not differ in placement moves and misrepresentations of foster family vs. group care experience in Barth [162] and Festinger [163].

Ironically, one model of a Home First group-care facility at this time was located just fifty miles north of Edgewood Center. Children’s Village was the winner of the 2007 Jacqueline Kennedy Onassis Award, the National Award for Public Service. In this facility, three “grandparents” who lived and rented separate homes on the property cared for twenty-one children. A “Community Village Center” housed administrative offices, a recreation center, and staff–child meeting places. Habitat for Humanity was building three more units on site. Viewed as a unique alternative model for kids in out-of-home care, Children’s Village served sibling groups, keeping them together and providing “long-term care for children for as long as they need it [164] (p. B3).” It was inexpensive care in that the “grandparents” are paid rent for the housing and are voluntarily provided care for the children. It was group care with a normative focus, not specialized mental health care.

Robert Frost, in The Death of a Hired Hand said “Home”, is “the place where, when you have to go there, they have to take you in” [165]. Though many believe this place to be the family, in fact the only such place that had to take you in for 150 years prior to its demise was the state mental hospital. Today’s array of foster care settings is not the current alternative. When actually focused on a specific type of setting, it is most likely that the only place able to and so required to “take you in” is a group-care facility. Thus, it is no accident that such facilities have become, in the United States, the residual repository of the failures of the foster care system. Yet, one needs to inquire why they are not a placement of first resort especially when instability in the child welfare system seems to be one of the few stable characteristics of out-of-home care. The foster family setting is itself less stable or enduring than the group care setting, suffering from the exigencies of family life that today create general instability in families and the ability to avoid the difficulties of parenthood by returning the child. In other words, foster family placements structurally are more likely to be a source of iatrogenic instability. Notably, in one of the few comprehensive studies available, 43.5% of a cohort meeting the criteria for an “Unstable Pattern” of care spent no time in group-care. They simply moved back and forth between different family-based settings, never achieving stability [83].

The assumption that this is a child problem remediable by “interventions” and not a system problem needs to be addressed and there needs to be some consideration of whether the volatility of children’s placement history could be averted if they are placed early on into a normatively designed group-care setting based on a boarding school type model. As noted previously, The Casey Family Programs Study of alumni found, all other things being equal, that total time in group-care (in years) was predictive of high school graduation. Why not use normatively designed group-care as a first placement of choice? The answer is perhaps in the history of placement change in the United States and the experience with institutional care abroad.

In the 1860s, large numbers of children were transferred from almshouses to orphanages so as to remove the “deserving poor” children from the contaminating influences of the poorhouse and increase the likelihood of such children being indentured or placed-out. The Children’s Law of 1875 in New York State, and similar laws in other states, prohibited future placement of children in almshouses. The ranks of orphanages were swelled by the availability of per-capita state funding in New York, a system that gave aid according to the numbers of inmates kept. Many superintendents packed their institutions, shed educational and religious programming, and became fiscally efficient warehouses for under-serving children waiting to be indentured. Other small orphanages retained their mission to “save” the children [166].

Introduced at the 1909 Conference on Dependent Children in Washington, D.C., where President Roosevelt urged that children not be removed from their homes for the reason of poverty alone, the new policy of moving children from institutions to foster family care came to be promoted with arguments of fiscal efficiency and the notion that social control would be easier to maintain in the family than in the institution. It was also believed that children could be protected from the abuses of institutional settings in these family environments. The move away from the children’s home (as early as 1910) was attributed to such things as “hospitalism”—or failure to thrive—and child labor exploitation. Other abuses notable in the first half of the twentieth century included placement in ever-larger facilities with decreasing amounts of resources to keep government costs down and the use of institutionalized children as experimental guinea pigs at all points in the effort to eradicate diphtheria. Ironically while institutional care was derided for its ability, albeit inappropriately, to deliver experimental medications, foster care, as previously noted, has allowed a third or more of its charges to go without immunizations [2,167].

Yet, as children often contributed to their own support in institutions with their labor in the earlier part of the twentieth century and state subsidies continued to flow for their support in children’s institutions, the ranks of these organizations continued to grow until in 1933, their population reached 140,352. Foster care was also growing at a rapid pace during the 1930s. One may speculate that it was spurred on by the fiscal incentives associated with providing foster care, much needed by families during the great depression and World War II.

Deinstitutionalization for children from the traditional children or orphan’s home was over by 1960. At this time, a major change occurred in the character of child institutional care in the United States. Facilities went from being organizations focused on the normalized growth and development of their children to being residual repositories of the failures of foster family care. Children’s homes became residential treatment centers, run-a-way homes, detention facilities, and substance abuse treatment centers, among others. Of particular note is the increasing use of residential treatment centers for emotionally disturbed children and juvenile justice facilities for juvenile offenders (such as detention centers, reception and diagnostic centers, training schools, and halfway houses). The specialized focus of these institutions—emphasizing the structuring of the child’s life in out-of-home-care around the particular problem-focus of the setting rather than around “healthy child in development”—has been paralleled by a significant expansion in the numbers of mental health professionals, the use of medication for behavior control, and the types of behaviors classified as medical/psychiatric disorders requiring specialized care. Since 2018, the Family First Act limits the use of Title IV-E funds for children and youth placed in nonfamily settings and creates Qualified Residential Treatment Programs (QRTP) as a new federal classification of congregate care facilities [19].

While the move from the children’s institution has been justified to be in the child’s best interest for purposes of child protection, the literature on the outcomes of group-care fails to support such conclusions (c.f., [73,80]). The abuses children suffer in institutions, given experiences in the United States and abroad, are less likely to result from the nature of the setting than from state sponsored neglect. For each negative experience, there are positive ones. Group-care offers opportunities to provide more enriched educational, vocational, and social environments, easier oversight of quality control, long-term peer if not always staff support, and most importantly, a greater chance for a stable childhood (cf., [168,169], below). The residualization of group-care is perhaps more responsive to its increasing costs and partially due to financing incentives in the professions. Cost issues in such provision are not considered in terms of the observed cost consequences of failing to provide an opportunity-focused upbringing.

Group-care models need reconsideration especially in view of their potential to reduce instability. Unlike foster care settings, group-care is more tolerant of child difference (e.g., children are not rejected because a foster parent finds out they may not have a high enough IQ) and, contrary to the initial impressions of the 1909 White House Conference that behavior is perhaps more controllable given group interventions, “Hospitalism” can be dealt with given appropriate staffing, and taking account of future costs, an early enriched group setting may in fact have more fiscal efficiency. The greatest danger in the use of any setting is state-sponsored neglect. This is evident in the lack of necessary services so basic as immunization in foster care settings [2,167].

McKenzie’s [168] alumni were housed in 9 orphanages, in rural small towns in the south and mid-west. The institutions housed 100 to 500 youth. The residents indicted the following positives and negatives of their group-care experience:

• Positives of Orphanage Experience:
  • Provided personal values and direction 60%
  • Contributed to sense of self-worth 59%
  • Education skill development and guidance 49%
  • Friendship and close sibling ties 38%
  • Religious and spiritual values 29%
  • Sense of stability and permanence 13%
• Negatives of Orphanage Experience
  • Separation from family 34%
  • Lack of love from institutional staff 31%
  • Lack of education, skill development, and guidance 27%
  • Lack of freedom 15%
  • Excessive punishment 12%
  • Excessive work demands 6%
  • Lack of amenities 6%
  • Poorly trained, underpaid, and unmotivated staff 6%

In one of the few other follow-ups of group-care, a sample of approximately 80 children from Quebec followed for fifty years and drawn from complainants seeking compensation for their incarceration, Sigal et al. [170] illustrate the impact of long-term neglect. A reading of the conditions that the children were subjected to illustrates state sponsored neglect, in this case through the absence of state involvement. With a few notable exceptions [171], the institutions from which these children came were underfinanced and understaffed. Funding came from charitable donations and, to a limited degree, from the government.

The authors note that:

As soon as possible, these babies were placed in crèches. Every 6 months, until they were 6 years old, they were placed in a succession of rooms, each containing 30 or more babies, toddlers, or young children of the same age. At the age of 6, they were transferred to orphanages, where they remained until their early teens. Subsequently, boys were placed in reform schools or on farms where they often worked as indentured laborers, and girls were placed as maids. Because of their limited education and the limited cognitive and affective stimulation during their incarceration and later placements, they had, with few exceptions, no experience in dealing with the exigencies of everyday life. They were ill prepared to enter into the labor market, they were ignorant of the value of money, and few had the social skills required for everyday life in the outside community

[172]

The results of this study show that members of this group at middle age were significantly more psychosocially dysfunctional and had significantly more chronic illnesses that could be stress related than a community sample. This sample as noted, however, came from those suing the state for their incarceration; it is a sample of the system’s losers and their negative outcomes illustrate this.

The problems experienced by children in Romanian and Russian institutions are well known [173,174]; their misfortunes are attributable to state-sponsored neglect and as current studies are beginning to show, not attributable to the group-care model. Wolins [73] reported on field studies of successful group-care in Israel, the Soviet Union, and Austria. Wolff and Fesseha [80], looking at the experiences of Eritrean war orphans, found group-home orphans had fewer signs and symptoms of emotional distress and greater adaptive skills than either reunified or institutional orphans, and they had fewer symptoms of emotional distress than home-reared children. However, placing orphans in small group homes was far more expensive than reunifying them with extended families.

Art Buchwald, the American humorist best known for his long-running syndicated column in The Washington Post on political satire and commentary, received the Pulitzer Prize for Outstanding Commentary in 1982 and in 1986 and was elected to the American Academy and Institute of Arts and Letters. He grew up in the Hebrew Orphan Asylum and said of this institution: “It was probably one of the most successful organizations in existence because it glues together people who have a unique childhood—not necessarily good and not necessarily bad—but unique” [175].

Foster care vs. group care debates have raged for 100 years with limited utility for the field and for children in need of long-term stable placement. Unfortunately, to my knowledge, there are few long-term follow-ups of foster family care, either kin or non-kin placements, though such foster family care has been a common placement alternative since the 1930s. This is a large gap in the out-of-home care knowledge base. Generic foster care outcome studies are rare as well but the most significant, Pecora et al. [176], gives little cause for optimism. This study evaluated the intermediate and long-term effects of family foster care on adult functioning using a sample of 659 young adults from two public and one private child welfare agencies. Foster care alumni completed high school at a rate comparable to the general population, but a disproportionately high number of them completed high school via a GED. Alumni completion rates for postsecondary education were low. Consequently, many alumni were in fragile economic situations: one-third of the alumni had household incomes at or below the poverty level, one-third had no health insurance, and more than one in five experienced homelessness after leaving foster care. Two foster care experience areas were estimated to significantly reduce the number of undesirable outcomes in the Education outcome domain: positive placement history (e.g., high placement stability, few failed reunifications), and having broad independent living preparation (as exemplified by having concrete resources upon leaving care). For the Employment and Finances outcome domain, receiving broad independent living preparation (as exemplified by having concrete resources upon leaving care) was estimated to significantly reduce the number of undesirable outcomes.

Perhaps the most misguided placement views are those of group-care. Unfortunately, advocates of family foster care over group care, while acknowledging the earlier work of Wolins and Pilavin [177] that demonstrated the potential of such settings to meet the needs of an undersupplied system, offer only lip service to the issue of selection in discussing the potentials of group care. Barth [162], for example, acknowledges the importance of selection of the most problematic children to group care in today’s out-of-home care system but then depends on the results of comparing such current residual group care settings to current foster care in his attempt to discourage the use of group care. Barth’s (2002) often-cited renewal of the foster family care vs. group care debate continues an unproductive tradition that seems to unfortunately perpetuate negative group care myths. Regarding the long-term outcomes experienced in foster family care vs. group care, Barth [162] summarizes the results of two major studies—Festinger [163] and Jones and Moses [178]—on which a previous review of the literature McDonald et al. [6] basse their conclusions: that the “… outcomes for children from family foster homes are better than those from group care.” Barth [162], in summarizing Festinger [163], notes that she

“… studied 277 young adults, aged 18 to 21, who had been discharged from foster care in New York City in 1975 and had spent five continuous years in care. [She] … concluded that subjects who were in family foster care functioned better than children in group care in the following areas: they attained higher levels of education … had a lesser likelihood of arrest or conviction….”

(Barth [162] p. 18)

The negatives go on but suffice it to say that Festinger [163] found no such results. She did no statistical tests for group differences. Such tests conducted on her data for this report show there were no significant differences between these groups on educational achievement or arrest and conviction. There also were no differences in the percentage with fulltime employment—61% vs. 54% among those in the labor market. More disturbing, however, is Barth’s description of Festinger’s comparison groups: “in family foster care” vs. “in group care.” Festinger [163] actually reported comparisons on children “discharged from family foster care” vs. “discharged from group care”. In fact, almost 20% of the males and 60% of the females in the Festinger “discharged from group care” group had been in three or more foster family placements. Being sorted into the latter group could have come about (though is not likely to have occurred with frequency) by having transitioned from a foster home to a shelter just prior to discharge from the system.

Barth [162] notes: “Jones and Moses [178] assessed the current functioning of 328 adults, aged 19 to 28, who spent at least one year in foster care in West Virginia between 1977 and 1984.” Barth reports the study’s conclusions: “that subjects who were in family foster care functioned better than children in group care in the following areas: …they reported fewer substance use problems and stronger informal support…” He notes: “More positive outcomes for adults who had been in group care versus family foster homes included: having closer and less negative contact with biological family; more likelihood of marriage, and, for men, a greater probability of having custody of their own children…”. Jones and Moses [178] also reported (though not mentioned by Barth [162] or McDonald et al. [6]) that no differences were found in employment status, job satisfaction, and school achievement by placement type among those children who aged out of the system. Of more importance than the results are the misrepresentation of placement type comparisons. In reporting placement type effects, Jones and Moses [178] actually compared children who had been in a single stable foster home for an average of nine years to children who had been in “combined placements” and “group or institution only placements.” They do not break out the outcome effects for “group or institution only placements” and note that the “combined placements” group had “more than one period in foster care”. Attributing even the modest positive outcomes to an endorsement of foster family care over group care when stability in a single placement is the primary characteristic of the foster family care group and the comparisons had been in more than one foster family placement for an unspecified amount of time perpetuates group care myths in this literature and is most unfortunate.

Also problematic in Barth’s [162,179] assaults on group care is his propensity to conflate all types of group care, most notably short-term mental health treatment and assessment-focused settings and shelters with the all group care.

Kinship foster care and other alternative placements also have their reality separate from their advocates’ promotional statements. For example, in comparing outcomes of licensed kinship foster care with non-kin foster care, Berrick [180] notes that: “Only one long term study has examined adult outcomes for children who were raised for a significant period of their childhood in kin and non-kin foster homes... [It] focused on licensed caregivers…[and] found that …children reared in the two licensed settings did not differ significantly in high school completion, employment, or mental and emotional health. In other words, when kin caregivers are screened and supported, they can not only protect children from harm but also foster their development [180] (p. 80”). This evaluation of the Benedict, Zuravin, and Stalling [151] study by Berrick [180] is much cited and quite misleading in that the foster care outcomes of this cohort, like foster care outcomes generally, are inadequate. Thus, the finding of “no difference” between kin and non-kin licensed care demonstrates both settings equally produce negative outcomes. Notably, Benedict et al. [151] followed up on 211 children between 1993 and 1994, who were 18 to 31 years old at time of interview (Mean age = 23). These children lived between 1984 and 1988 in licensed family foster homes in Baltimore City. The study found that at follow-up: 56% were not working, 45% reported an annual income of less than USD 10,000 (their Median income was USD 15,000), 31% were receiving public assistance, 38% reported trouble with the law, and 27% had been homeless at least once at some time in the past. Also in the multivariate evaluation presented in Benedict et al.’s [151] study, all other factors viewed as significant considered, time spent in kinship foster care was associated with a reduced probability of a child completing high school (including GED completion). Children placed with relatives have longer periods in care than children placed in other settings [181] and findings indicate that stability of placements declines as time in kinship care grows [155,182,183]. Taken together these findings seem indicative of the failure of kinship care with older children and its inability to provide for the future opportunity of these children.

Of greater concern is that the results of Benedict et al. [151] show a possible pattern of negative intergenerational risk for serious substance abuse. In their sample, children placed in kinship foster care were more likely to be placed there due to maltreatment and their mothers were significantly more likely to be involved in substance abuse. As adults these children were more likely to abuse heroin and to have sold sex for drugs than those placed in non-kin placements. Further, Benedict et al.’s [151] multivariate analysis predicting adult 28-item General Health Questionnaire (GHQ) scores [184] (a measure of non-psychotic psychiatric illness outcomes) indicated that children with a substance abusing mother placed in kinship foster care were more likely to have poorer mental health outcomes in adulthood than children in non-relative foster care. Berrick [180] notes that judges in the past were far more reluctant to place children with kin because they were concerned about the risk of possible continued abuse. She indicates that this attitude has changed to a greater willingness to make such placements. Unfortunately, the research may not support such attitude change. The risk of continued abuse varies from 18 to 30% [185], as abuse and neglect are often intergenerational—dysfunctional families breed dysfunctional families [186,187]. Children in kinship foster care are more likely to have come from abuse situations [151] than children in other types of foster care settings and when placed with kin more likely to maintain birth parent contacts [188] and to have unwelcome/unauthorized contact with their birth parents [189]. These children consequently are being exposed to the same cultural and structural situations that fostered the failure of their parents’ households. In 2016, over a third of child removals (foster care entries) involved parental substance abuse [127], and recycling into the system within 12 months was twice as high among those removed for substance abuse and mental health problems compared to those removed for no neglect, and 43% higher compared to those removed for neglect without such problems [36]. It is likely, 20 to 25 years out, there will be a recycling of many these children back into care. It would appear the attitudes have changed toward kinship foster care due to a lack of alternatives (c.f., discussion of supply of foster placements below), not because kinship foster care offers better solutions to the negatives associated with foster care placement [190].

As noted above, between 1998 and 2000, the California Child Welfare system saw an increase in the number of 24-month “long-stay” children of almost 20% in less than three years, from 60,776 in the Jan.–Dec., 1998 reporting period, to 72,805 in the Oct. 1999/Sept. 2000 reporting period. It would seem that this rapid increase could have played a role in the State’s decision to implement the KIN GAP (the guardian assistance program) beginning in 2000. This latter program offers cash assistance to relatives who take a guardianship role for children and does so without the level of oversight normally accorded to foster care arrangements. On the positive side, a child had to have at least a year in kinship foster placement with the family who became their guardian before the guardianship role was finalized. However, following the implementation of the KIN GAP program the population in the 24-month long-stay group declined by 44% from its high of 72,805 (in Oct. 1999/Sept. 2000 period) to 40,765 in Oct. 2006/Sept. 2007’s reporting period. It is unclear how these children have fared and their disappearance from the State’s responsibility is of concern without such knowledge [34].

Looking at the structural character of kinship foster care, kin-foster parents, in general, are older and less financially stable; they are more likely to be single parents, and they have less education and poorer health [191,192,193,194]. All are structural characteristics associated with poor outcomes for children in adulthood. In particular, single-parenthood, lower levels of education, and poverty in caregivers predict lower educational achievement and employment earnings, and more antisocial behaviors among the children they care for when these children reach adulthood [194,195].

It also seems that the differences between non-kin care and kin care are minimized in studies that report comparisons in ways that seem to mask the problematic nature of the latter environments. This tendency is evidenced in Benedict et al.’s report [151] and in Berrick’s [196] summary of her study. She indicates that “The differences between kin and non-kin are not striking…”. Yet, of the 17 measured differences, 14 reflect negatively on kin-homes and those that do so for non-kin homes are more easily explained and accounted for. All of the study’s significant differences are noted in

Table 1. Characteristics of non-kin vs. kin foster homes summarized from Berrick’s study [196] (N = 34 non-kin and N = 24 kin foster care placements). †

Characteristics Favoring Non-Kin vs. Kin Foster Homes Studied
Services:
Child sees counselor/ therapist, 78% non-kin vs. 44% kin **
Household social characteristics:
Concern about drug or alcohol use by other adults living or visiting home, 6% vs. 32% ** FP has been threatened or attacked, 3% vs. 27% ** Violence connected to drug use or drug dealing perceived as neighborhood problem, 3% vs. 36% *** FP experience in caring for abused child, 29% vs. 54% * FP uses time out for discipline, 75% vs. 46% * FP has received training from social services agency, 79% vs. 18% ***
Household structural characteristics:
General structural conditions rated on a ten-point scale, 10 vs. 9 ** Number of bedrooms, 4 vs. 3 * Child has own bedroom, 79% vs. 46% ** Support for educational and extra-curricular enrichment: Fewer hours of TV watched per day, 1.27 h non-kin vs. 1.97 h kin **
Safety Precautions in Home:
Home has first aid kit, 97% non-kin vs. 71% kin ** FP knows CPR, 93% non-kin vs. 57% kin *** Has fire extinguisher, 97% non-kin vs. 54% kin ***
Characteristics Favoring Kin Foster Home
Safety Precautions in Home:
Own guns: 33% non-kin vs. 4% kin ** (All guns in locked setting in both facility types) Children more likely to know where to go in emergency, 79% vs. 96% * Warmth of relationship between birth mother and child characterized as “non-existent”, 34% vs. 4% kin *** (Yet, no differences between non-kin and Kin were found in frequency of visiting.) ***

† The response rate to this study was 14.9 for non-kin and 9.9% for kin homes. No multivariate analyses were done. * n.s. ** p < 0.01 *** p < 0.001.

. Of most concern are differences related to drug, alcohol, and violence issues that clearly characterize kin care settings. Such differences are very much related to abuse and neglect situations that bring children into care in the first place. It is also of concern that the study did not explore issues related to abuse and its relation to visiting with the abusing parent. The only significant differences perhaps lending to a more positive perspective on the kin care environment was that the gun ownership in the kin environment was less and that there were more frequent reports of “non-existent” relationships between the child and the birth parent in non-kin settings. The former, however, was of less concern as the study found that such weapons were locked way in 100% of the cases. The latter issue raises a validity of report question related to the multiple meanings of the initial question that is scored from “very warm or warm” to “non-existent”—indicating both an affectual and a contact continuum. The fact that no differences were reported in the frequency of visiting with birth parents seems to indicate that the question relates to affections which given past abuse by the parent may be justified as non-existent.

To make matters worse, there are at least three types of kinship foster care: licensed, non-licensed, and kin guardianship. Each type is progressively associated with less oversight by the state. Evidence from California (where kinship caregivers are not required to become licensed or trained) suggests that some kinship homes provide less than satisfactory care [197]. Participants in a statewide survey of child welfare workers reported that approximately one-third of kinship homes fell below the standards they regularly witnessed in average foster family homes. Further, the disappearance of the initial stability advantage enjoyed by kinship foster care over non-kinship foster care has already been discussed [198] and therefore the question of the utility of this placement mechanism for those children needing long-term placements is open to question.

It might be argued, and certainly needs to be substantiated, that the infusion of financial reward into kinship care has changed it from a situation involving the heroic efforts of those seeking to help a family member to a potential business opportunity almost the equivalent of non-kin foster care. When converted to kinship guardian care, it has the potential to become an outlet for extruding needy children from the system without oversight, allowing the system to claim success in reducing its reported numbers in care and absolving the system of responsibility for these youth, the next form of cheap neglect.

Berrick [180] notes that several sources indicate that kin care families receive less support, fewer services, and less contact with child welfare workers than non-kin foster families [191,192,193,199]. She suggests that offering such assistance may help ameliorate their situation. While these families may benefit from such assistance, such efforts are unlikely, however, to overcome the structural deficits noted above as being most associated with a lack of achievement among children in foster care.

Permanency planning favors reunification and prevention of placement in out-of-home care, yet the literature tends to show that outcomes associated with both options are often less favorable than those achieved by children who remain in out-of-home care. Bilaver et al. [198] compared three groups: a group having aged out of care, a reunified group, and a low-income group in California, Illinois, and South Carolina. No pattern of employment differences emerged in the three groups. Dworsky and Courtney [199], studying 6442 foster youth, found that during the first two years after leaving care, youth who aged-out or were discharged to independent living earned significantly more than youth who had returned home, been adopted, or moved to relatives’ homes. Taussig et al. [200] studying 149 reunified youths reported they had lower grades and more self-destructive behaviors, substance abuse, total risk behavior problems, and arrests than those remaining in foster care. Lau et al. [201] (N = 218) compared reunified children with those adopted and those remaining in care. They found that reunified children were exposed to more adverse life events in the form of elevated family dysfunction, instability, and harm (suffering serious accident and illness) and received fewer mental health services—though were less likely to feel socially isolated than comparisons. Litrownik et al. [202] (N = 254) showed that reunified children and their parents reported more family violence than did children who had not reunified and their caretakers. Runyan and Gould [203,204] (N = 220) studied children in foster care for more than three years compared with a matched group who had not been in foster care. There were no differences in delinquent behavior but school attendance improved for the children placed in foster care, and foster care reduced (though it did not eliminate) the possibility of future abuse.

Despite the fact that family reunification efforts for children in foster care all too often result in repeat incidents of abuse and neglect, rapid family reunification remains the preferred permanency placement goal for children placed in foster care [36,205,206]. Just under half (48% in 2020), down from 57% in 2000, of the children are reunified with their parents [207]. There are pressures to reunify quickly: approximately two-thirds of reunifications occur in less than 12 months. The Adoption and Safe Families Act (ASFA) [208] limited the time parents have to resolve the problems that lead to removal. Family and Child Services Reviews evaluate states on the percent of children exiting to permanency in less than 12 months, and states that underperform or fail to improve on this metric can be penalized by withholding federal funds [209]. The process of family reunification involves returning children, who have been temporarily placed in out-of-home care to their families of origin. A successful family reunification is one that does not result in the recurrence of child abuse, child neglect, or a child’s re-entry into foster care [210]. A successful family reunification is also defined as one that occurs within 12 months or less from the time a child enters out-of-home care [208,210]—the 12-month timeframe deriving from permanency planning mandates in the Adoption and Safe Families Act (P.L. 105-89). National statistics, however, show that states with relatively high percentages of reunifications within the first 12 months of a child’s tenure in foster care also tended to have relatively high percentages of re-entries into foster care within 12 months [210,211]. Research suggests that decisions are being made to reunify children with parents who have yet to overcome the problems that led them to maltreat their children in the first place [210] in order to meet the legal timeline. If true, this indeed would contribute to child instability.

George [119] suggests that the preference of family reunification over other placement options is policy because “states prefer reunification for administrative reasons… foster care is costly and adoptive homes are difficult to find [119] p. 424.” Reunification is preferred because it is the easiest and cheapest permanency plan to enact.

Supportive services. The use of supportive services such as “family first” and “wrap-around services” does not compensate for the lack of stable substitutive care placements. The latter, while shown to be modestly effective [212], and the former, while seeming a good idea, are both interventions that work against ongoing traumas precipitated by instability with limited or no extended responsibility attached to the service provision. Thus, the provider has an immediate successful outcome and in the long-term may not be there for the child when the situation fails. Repeated failures engender a growing mistrust from the child and an increasing unwillingness to participate in future care giving efforts. Evaluation of these approaches is necessary in the long-term, though such evaluations will be hard to implement, as the failures will in all likelihood be underrepresented in a follow-up sample.

3.4. Inadequate Supply of Quality Family-Based Placements and Instability

The objective of the child welfare system should be to maximize the supply/number of quality placements of all types in order to ensure the availability of home-first options to those who require them. Without an adequate supply of well-supported placements, there will be no quality of care [213]. This report, in discussing different placement options, has offered a counterpoint to current advocates who emphasize the need for a given type of placement—notably one approximating the character of a family in the form of non-relative and kinship foster care. This presentation is offered to balance the equation of valuing differential placement structures and to lay the foundation for discussing the most significant issue necessary for the insurance of stability in out-of-home care: the fact that at least 20% of the children in the out-of-home care system need a “Home First”—they will not reunify, nor be adopted, and/or attempts at reunification will fail.

Bogen [175], in his history of one of the largest orphanages, The Hebrew Orphan Asylum of New York, said: “The social work profession’s dependence on only one form of care in the last fifty years was a smug, narrow and short-sighted approach. There will never be enough good foster homes for all the children who need them”. To this observation should be added the child welfare system’s overemphasis on a “family” ideal inappropriately permeates foster care placement priorities. The dearth of foster family placements is a problem across the nation and has been for decades.

Reports of licensed foster care home numbers and availability to meet need from the 1980s through 2020 appear to validate this claim. In the earlier decades, one study found that the number of available foster homes decreased from 147,000 nationwide in 1987 to approximately 100,000 three years later [214]. In Seattle, Washington, the number of foster homes dropped by 11% from 1996 through 2001. In Corpus Christi, Texas, the number of children in foster care increased by 70% from 1998 to 2000, while the number of homes for foster and adoptive children increased by only 7% [215]. Between December 2003 and June 2005, the Department of Economic Security in Arizona increased the number of foster homes available by almost 669 homes to serve 1606 additional children. However, it was not enough to keep up with the need. The number of children in foster care grew by more than 1500 in the second year of the period alone, from 8246 in March 2004 to about 10,000 in June 2005 ([216]. A 2007 report on California counties shows that of the 21 counties surveyed, representing 85.6 percent of the foster care population in the state, 77 percent reported a loss in licensed foster family homes in the last decade. The overall decline reported by the counties translates into a loss of at least 3059 homes statewide and 18,354 potential placements, since each home can be licensed for up to six children. The number of placements with licensed foster families decreased roughly 30% since 1999 [217].

Newly released data compiled and analyzed for “Who Cares: A National Count of Foster Homes and Families” continues to report inadequate numbers of family foster care placements to meet need [22]. The report notes the availability of licensed foster homes took a hit in 2020 according to figures collected directly from state child welfare agencies. This said, the report also shows a bigger increase in the actual number of providers over the numbers in previous decades. It notes that the number of licensed foster homes increased 3.5% between 2018 and 2019, but the numbers for 2020 show a decline. There were 218,927 licensed foster homes in 2019, the total dropping to 214,421 in 2020, a decline of approximately 2%. It would seem that the figures showing an increase in providers in the last decade reflect an accounting change with the inclusion of homes where a relative or family friend had become licensed exclusively to care for a child or children. For example, Florida saw a 40% uptick in licensed homes, but state officials informed us that this increase does not necessarily suggest a major change in the number of caregivers. Rather, it was mostly attributable to a new tier of licensure available to relatives and other kin [218]. Much has been said to blame the loss of providers on the increasing difficulty of children in care. Yet, it also might be speculated that there are fewer people wanting to care for children in Western societies, especially children needing placement. The willingness of current child rearing age adults to have and care for children has diminished as indicated by falling birth rates in all but the poorest minority populations. The reduction in available foster homes is further exacerbated by the growth in the number of single-parent households, the growth in the proportion of women employed outside the home, the increase in divorce rates, and the rising costs of child rearing [218,219]. All these factors make it more difficult for any family to take responsibility for raising additional children.

The emphasis on the family structure of substitutive care, i.e., non-family foster care and the various types of kinship foster care over group care, is dysfunctional to the objective of providing a Home First in a system where the key issue is supply of stable quality long term placements. The problem is finding or creating enough of such placements without having to accept less than adequate ones. With the current supply situation that is exactly what is happening. In California, for example, Berrick et al. [196] indicate that participants in a statewide survey of child welfare workers reported that approximately one-third of kinship homes fell below the standards they regularly witnessed in average foster family homes. In the past judges, were reluctant to place children with kin for valid reasons noted above. It appears attitudes changed toward kinship foster care not necessarily because the quality of such placements was discovered; they changed because there were not enough non-kin foster placements [180].

Given new findings discussed above [21], the emphasis on family type placements has led to a “holding tank” mentality in the system—children are awaiting a family-type placement, often in group care facilities designed for either passing through and/or treatment or disciplinary purposes. Given the lack of supply of quality family settings, it is time to face the fact that many children will not find an ideal family setting and that, even if system personnel believe that they have such a setting for a child, the odds are against it being a successful placement. This is especially true for new foster care settings where research shows as many as 60 percent of new foster parents quit within the first 12 months [217], propelling the child back to a new holding tank placement and exacerbating instability. That group care facilities currently in place are serving as holding tanks rather than solely serving to address the needs of especially troubled children is perhaps validated by Barth’s [162] failure to show differences on health and behavior problem indicators between group care and other foster family care children. From this result, it should be evident that there needs to be better use of group care. Facilities need to be created that emphasize normative development Home First objectives rather than transitional or specialized treatment functions.

The definition of a group vs. a family foster care placement bears some scrutiny. Apparently, foster family homes are licensed for up to six children, a fact defining the limits of foster family care though bringing it close to group care. The “family” number appears to be a result of the limits on numbers of unrelated individuals allowed in a single-family unit residential zone without the necessity of obtaining a conditional use permit. Defining six unrelated individuals in a foster family placement as a “family” is a legal artifact and approaches group care in fact and perhaps childhood perception. Whether it becomes a family is open to serious question. Art Buchwald, who had lived in three foster homes, said in a television interview: “Neither the child or the people who take you in ever make an emotional commitment. A foster child senses this very early and he doesn’t want to get too close because he knows that these people are not for real [169] (p. 222)”. This latter comment may be overstating things, for Festinger [163] found that approximately 40% of her sample was discharged to their foster family placement. This latter finding is perhaps her most important contribution in that it is significant evidence of continuity of relationship and should be given close scrutiny in other studies. Having said this, the 40% figure probably is a best estimate on the limits of the “family” ideal in foster-family care. Jones and Moses [178], in whose sample 78% (N = 315) had been in a single placement for their entire time in out-of-home care, reported only 30% as listing their first residence upon aging-out of the system as their foster home.

SAFE homes [220], though less than effective as short-term assessment facilities, have some good and poor structural qualities. Facilities structured to provide long-term placements, with approximately 12 children (range: 8–20 beds), in a residential area, providing age-appropriate developmental, educational, and recreational activities with access to local schools, the ability to accommodate siblings, and have couple parenting absent of a treatment-driven focus that deprives children of a “normal” lifestyle might provide a fine Home First setting. Eliminating the use of short-term shelters whose sole purpose is assessment and processing by providing reception centers for children who are entering foster care so they can have a dignified, supportive environment for up to 23 h while awaiting their second placement to accomplish some of the functions of shelter care, without the drawbacks as suggested by Barth [162] is a fine idea. Failing to consider a long-term group care placement, however, may be a mistake that deprives “the 20%” of a stable childhood.

3.5. Home First: Group-Care, Stability, and Outcomes

Craft and Friedland [169], authors of “An Orphan Has Many Parents”, were both raised in the Pride of Judea Children’s Home, Brooklyn congregate facility. Craft’s mother died in childbirth. He was sent to an infant home for five years, a time he describes as “filled with indescribable terror and misery”. Craft was then transferred to the Pride of Judea Home in 1932, where he stayed for 14 years. Here, Craft found the adults and peers that became the family he so desperately wanted and needed. Friedland, then nine, and his older brother were sent to the Pride of Judea Home in 1940 after running away from a foster home where they had been abused for three years. Friedland spent seven years in the “Home”—as it was called by his fellow residents. Having attended local schools, these two authors with thankfulness recall their daily lives, including sleeping arrangements, food and clothing provisions, sports activities, religious studies, and music instruction. They remember the caring parent figures who ran “The Home” and the colorful resident-friends they grew up with. Their description of life in this congregate facility is a remarkable memoir of a stable childhood that is both entertaining and moving.

Having a home and the entitlements attached to it are fundamental to the achievement of the accomplishments of Erickson’s first four developmental stages—trust, autonomy, initiative, and industry, and is the foundation of identity development and the transition to adulthood since, at home, there is always something in your social margin “bank account.” The sense of trust that it will be there for you provides a sense of security, if not as deep as that derived from your initial primary caregiver, one that is central to development as a healthy adult. Stability in placement is a necessary, if not a sufficient, condition to finding a home and enables children to build social margin. McKenzie [168] studied a cohort of 1856 alumni who spent upwards of nine years in children’s orphanages. Between half and two-thirds of alumni who evaluated their experience mirrored Eriksonian achievements in their responses. They noted that the positives provided by their experience in out-of-home care included personal values and direction (60%), a sense of self-worth (59%), and education, skill development, and guidance (49%). These youth might be viewed as successes because they were recruited through alumni groups and achieved educationally and vocationally better than the general population. Yet, the long-term stability of their placements may be viewed as a major contributor to their success. Notably, youth (in the previously mentioned study of the Casey alumni) with fewer placements per year while in foster care were more likely to complete high school before leaving foster care: “If youth do not move between homes they are more likely to develop networks of support and coaching that can help them further develop their life skills; and they have more chance to benefit from independent living training,” [33] (p. 41). Friedland, in his book, The Judeo Twins, writes of two successful athletes who were residents of a children’s institution in Brooklyn, New York between 1939 and 1951 [221]. After 60 years these two brothers have maintained many of the friendships they made at their institution. Friedland names these friends individually as a concrete example of building social margin through peer networks. In group-care, staff often turn over on a two-year basis but one’s peers may remain in place—or in the past did so. Ironically, Friedland also names continuing staff relations the brothers maintain and these are not unique outcomes of group-care (see: [169,222]).

Above all, the objective of out-of-home care should be to enable healthy child development to foster opportunity. Children are placed in out-of-home care because their home of origin has been deemed so disruptive to their healthy growth and development as to require the State to provide an alternative home, to become the parent. Parents foster opportunity.

McKenzie’s [168] 1856 alumni of children’s institutions are the “successes” of out-of-home care because of the conformity of their values with Ericksonian achievements supported in common long-term stable placement histories. The alumni were, on average, eight years old when they arrived at their institutions, and they stayed for an average of nine years (with most staying until they graduated from high school). At an average age of 44 when studied, these alumni surpassed the general 40-plus white-population at every rung of the educational ladder, except at the high school level—at which both groups had practically identical graduation rates. The US unemployment rate was over 6 percent in 1994. The alumni who were not retired had an unemployment rate of 1 percent. The alumni 44 to 54 years of age had a median income 16 percent higher than their counterparts in the general white population. The alumni 55 to 64 had a median income 32 percent higher than their counterparts, and the alumni 65 and older had a median household income 75 percent higher than their counterparts. In 1992, the national poverty rate for white Americans in the age group 45 and older was between 5 percent and 6 percent; the poverty rate of the McKenzie alumni was no higher than 3 percent. Less than 3 percent of the alumni have ever been on any form of public assistance (not counting Social Security) vs. 19 percent of the general population. At the time of the study, a white American had a 1.6 percent chance of spending some time in a state prison during his or her life compared to less than 1 percent of the McKenzie alumni who report of ever spending any time in a jail, state prison, or federal prison. About 76 percent of Americans who were 45 years old and older voted in the 1992 election. Nearly 88 percent of the alumni voted.

With respect to their “Attitude toward life” the following question was posed to McKenzie’s alumni: “Taking all things together, how would you say things are going these days?” In 1994, 29 percent of the respondents in the general population in response to this question indicated they were “very happy”; 59 percent indicated they were “somewhat happy”; and 12 percent indicated they were “not too happy”. The McKenzie alumni, on the other hand, indicate a far more positive attitude toward life: 58 percent were “very happy” (exactly twice the percentage for the general population); 37 percent were “somewhat happy”; and 5 percent were “not too happy” (less than half the percentage for the general population).

McKenzie’s [168,223] alumni appear to have an overwhelming preference for their way of growing up over the next best alternative. When asked if they preferred to grow up in their orphanages or foster care, just over 92 percent preferred their orphanages, less than 2 percent preferred foster care, and 6 percent reported not knowing enough to say one way or the other. When asked if they preferred to grow up in their orphanages or with the available members of their own families, 75 percent of the respondents chose their orphanages, whereas less than 16 percent chose their own families, with less than 10 percent not being able to say.

Although the alumni had advantages other children did not have (for example, some reported having access to recreational facilities and the fine arts), the alumni were not reared in the lap of high-priced care. The cost of care (covering housing, recreation, supervision, and basic amenities) per child around 1950 for the alumni from the Presbyterian home was less than USD 3000 per year in 1995 dollars. When the cost of education and administration are added, the per-child cost reached no higher than USD 5000 a year, again, in 1995 dollars [168].

The McKenzie alumni sample was 97.4% non-Hispanic white so comparisons are made to non-Hispanic white adults [168,223]. There are few studies that address the role of race in adult out-of-home care outcomes. Benedict et al. [151], in evaluating adult foster care outcomes using multivariate modeling, did not find that race was significantly associated with adult employment, high school graduation, or mental health status outcomes among adults aged 18–31 who had been in foster care. Eighty-seven percent of the Benedict et al. [151] sample were African Americans. It is possible that structural issues related to deficits generated in the foster care experience are strong enough to overcome racial differences. Thus, those who would dismiss the McKenzie findings on the basis of hypothesized differences between Caucasians and African Americans do so at the risk of blaming the victim for systemic flaws.

Earlier studies, as noted above, reflect a time period when children were likely to evidence fewer placements than currently. These studies report a limited relationship between placement stability and empirical outcomes. Weinstein [224], in a sample of 61 children in foster family homes in the Chicago area, found no difference in the well-being of children who had experienced one or two placements when compared to the well-being of children who had experienced more than two placements. Fanshel and Shinn [81] studied 624 children who entered foster care in New York City for the first time in 1966 and who were 12 years of age or less at the time they entered care. They found no relationship between the number of placements and ratings made by parents, teachers, and caseworkers over a period of time on various indices of behavior. Meier [225] interviewed 82 adults between the ages of 28 and 32, who, as children, had been in foster family care for at least 5 years in Minnesota and who did not return to their own parents’ care while children. She found no relationship between well-being and social effectiveness as adults and the number of moves experienced as foster children. She did find a statistically significant relationship between the number of moves the adults had experienced as foster children and the number of changes in residence they had made as adults; but she found no relationship between number of moves, either as foster children or as adults, and other measures of stability.

Could the findings of the earlier studies be the result of the fact that the sampled children actually all had long-term stable placements and fewer moves and thus had all found a home? Today, children experience many more placement moves than in the past. One wonders whether the abandonment of these children to long-term placement, discovered by Maas and Engler (1959), rather than an emphasis on reunification had an unforeseen positive effect.

3.6. Investing in Children Yields Results

Investing in children yields results; failure to do so is catastrophic. The latter has been demonstrated in the repeated negative outcomes resulting from state sponsored neglect described above. Increased investment does pay off in results as is indicated by the findings of Kessler, R. C., et al. [226] showing that enhanced foster care produced better results than standard system care. The program described as “enhanced foster care” spent 60% more on its charges than the public system. Caseworkers in the model program had higher levels of education and salaries, lower caseloads, and access to a wider range of ancillary services (e.g., mental health counseling, tutoring, and summer camps) than caseworkers in the public programs. In this report, alumni from enhanced foster care…

…had nearly 2 more years in care than public program alumni did in Oregon (8.4 vs. 6.5; F1,151 = 6.6; P = 0.01) and 2.5 more years in Washington (9.8 vs. 7.3; F1,319 = 98.4; P_.001). [They] had significantly more stable placements (mean duration of placement) than public program alumni did in Oregon (32.7 vs. 13.3 months; F1,151 = 20.7; P_.001) and Washington (26.4 vs. 19.2 months; F1,319 = 12.6; P_.001)….[Enhanced program] alumni were substantially less likely to experience adverse events during their time in care….were at 82% (compared with Washington alumni) to 88% (compared with Oregon alumni) lower risk (on the basis of risk in the public program samples) of having experienced a reunification failure during comparable periods of time in foster care (2.9 incidents per 100 person-years in the [enhanced] sample vs. 23.5 in the public program sample in Oregon [F1,151 = 45.1; P_.001], and 1.9 incidents per 100 person years in the [enhanced] sample vs. 10.7 in the public program sample in Washington [F1,319 = 68.9; P_.001]). [Enhanced] alumni were also at a consistently lower risk of documented incidents of foster parent neglect (62.3% lower risk in Oregon and 26.9% lower risk in Washington) and physical (81.0% lower risk in Oregon and 14.3% lower risk in Washington) and sexual abuse (39.3% lower risk in Oregon and 37.9% lower risk in Washington)

(Kessler et al. [226] 625–633)

This is clearly an illustration that you get what you pay for. Yet, it must be remembered that this is a relative comparison. In fact, this is the Casey alumni sample referred to above [176] that still suffered long-term opportunity related deficits. Thus, the importance of this effort is the demonstration that increased investment pays off. The lesson is that more such investment in perhaps more diverse and quality settings that guarantee greater stability is needed. McKenzie [168] shows it is possible to do much better.

4. Conclusions and Recommendations

4.1. Conclusions

Children placed in out-of-home care deserve a Home First. A home in a stable environment—uninterrupted by bureaucratic processes associated with “permanency planning”, “reunification”, and free of child-processing placements. A Home First placement should build on strengths and provide opportunity for normal growth and development. Next to properly resourcing such placements and the child’s transition to adulthood, providing a secure and stable base for child development is one of the few strengths-focused actions the state can offer that seems to show long-term positive results. Fong, Schwab, and Armour [227] point out that there are several different kinds of factors other “… than attachment and relationship continuity that have been found to influence children’s adjustment and well-being [228] including school and peer experiences, leisure-time activities and interests, and the larger social environment [229,230]”. Stability enables these experiences and the literature is replete with repeated evidence of the importance of stability in the lives of children and particularly those growing up in out-of-home care. It is clear that such lives are particularly vulnerable to state sponsored neglect that fosters actively abusive environments. An under-resourced system, particularly one that settles on low cost solutions or organization and financing schemes that encourage potentially abusive situations to satisfy cost cutting and out-of-home care population reduction mandates, is of significant concern as its results will be costly to society in future necessity for expenditures on social control of disordered and deviant system products, the children.

The empirical foster care literature is handicapped by a lack of appreciation of the bio-psycho-social child—it is too narrowly defined to understand the impact of multiple factors in a child’s life. Its theoretical foundation emphasizes the psychology of attachment at the expense of social, environmental, and organization issues. Statistical models used in its research are poorly specified in that they fail to incorporate significant aspects of the child’s environment, notably context, peer relationships, and system characteristics, that will interact with stability and are dependent on it in a manner that fosters positive outcomes. The literature also is lacking in the child’s voice and concerns, offering little recognition of what is a growing literature on the positive or unique contributions of group-care.

Ironically in the past, the emphasis on group-care was associated with failure in child oversight and the achievement of stability by default—children were simply left to live out their childhood in institutions. Prevailing Legislation: The Adoption Assistance and Child Welfare Act of 1980 (Public Law 96-272), the act that has provided the policy framework for public child welfare services, is designed to limit the number of children in foster care and, for those who are placed, to promote their return to their own or to other families. The Adoption and Safe Families Act of 1997 (Public Law 105-89) is designed in part to facilitate these goals. This act expands the Family Preservation and Family Support Program to include family reunification services up to 15 months after a child enters foster care. It reduces the number of months a child may remain in foster care without a permanency hearing from 18 to 12 months. It requires that states file for termination of the rights of parents of children under age 10, who have spent 15 of the most recent 22 months in foster care. Finally, the Family First Prevention Services Act (FFPSA) signed into law in 2018 continues the family focused effort while specifically discouraging congregate care.

While for many, such legislation has led to early adoption and reunification, the consequences for at least one in five as the result of this legislation has been the development of iatrogenic instability—where 50 to 70% of moves are system generated and the child is caught in a turnstile of unstable relationships with all the dysfunctional implications noted above accruing to this situation.

Further, an emphasis on kinship care has led to programmatic encouragement of the kin as paid guardians with less oversight than that of those providing kinship foster care or non-relative foster care and perhaps without the initial stability benefits believed inherent in this situation. This latter system is a recipe for state sponsored neglect—it offers incentives to get children off the roles with little understanding of their potential outcomes or protection against adverse outcomes.

In order to change, the system must be well resourced. It must further consider changes in its policy, practice, and research. The following recommendations offer only a beginning.

4.2. Policy and Practice Recommendations

Recommendation #1: “Home First” The main objective of out of-home placement should be to find a home that will be acceptable over the long-term. It should be the child’s first and hopefully his last placement.

Recommendation #2: Focus the Home First effort on those children most likely to be in the 20%. Begin, perhaps, with those failing reunifications.

Recommendation #3: Convene a national working group to consider alternative design possibilities that maintain a focus on “Home First”. Its tasks would be to mesh the demands of permanency planning with the need to insure a single placement that can be an enduring home if a return to family or adoption is not possible and/or by virtue of clear and convincing evidence unlikely without repeated disruptions.

Recommendation #3a: Panel should draft legislation modifying and/or supplanting the Family First Act of 2018 policies discouraging congregate care and especially congregate care duration time limits.

Recommendation #3b: New legislation should ensure that in addition to “Qualified Residential Treatment Programs (QRTP)” funding is available for Qualified Residential Programs (QRP) focused on normalized and stable socialization as opposed to “treatment”.

Recommendation #4: Consider the redevelopment of group-care settings with a strengths model/boarding school design as Home First Settings.

Recommendation #4a: Home First settings should maintain broad growth and development objectives: Providing a stable environment, access to public educational instruction, moral education, recreation, the arts, and ensuring health and safety provisions.

Recommendation #4b: Home First settings should provide pro-social models to enable children to be kept off the streets and away from crime and to be able to avoid extended tenure in care settings built around deficits, such as diagnosable mental disorders.

Recommendation #5: Separate out the concept of stability from permanency planning. If a placement becomes stable and shows no immediate or long-term potential of dissolution, it should not be sacrificed to a permanency plan unless the latter has, based on hard-evidence, clear and convincing merit.

Recommendation #6: The criterion for retuning the child from a “Home First” setting should be that the family home presents “clear and convincing” evidence that it will provide a better opportunity for stable long-term child development. Evidence of significant probability of a recurrence of the original placement circumstances should be accepted as prima facia justification against the disruption of a stable child placement.

Recommendation #7: Learn from international experiences. For example, group-care facilities for children in India, where they have not gone to the foster care model and still have children’s institutions where the objective remains promoting strengths and getting the child into a viable adult niche as opposed to treating their behavioral issues, need further investigation. Many of the behavioral issues of children are transitory, child diagnoses are highly unreliable [231], and such transitory behaviors, what sociologists might call “primary deviance”, may best be handled outside the medicalized diagnostic framework.

Recommendation #8: Pass national legislation ensuring adequate financing of congregate care facilities including requirements that hold states responsible for ensuring quality oversight of such facilities.

4.3. Research Recommendations

Recommendation #1: Focus research on placement characteristics that will promote strengths vs. those that will address deficits.

Recommendation #2: Focus on specifying the children who make up the 20%. Emphasis should be placed on structural characteristics that predict reunification failure and bode ill for adoption.

Recommendation #3: Promote research into Home First design.

Recommendation #4: Evaluating the effects of stability on child outcomes over the course of their out-of-home care experience has been an objective of much empirical research. Yet, establishing such a relationship with even modest causal certainty is extremely difficult. Confounding, mediating, and moderating factors, subgroup interaction effects, and contextual effects will always create a challenge to causal attribution. Even in a randomized trial, post-randomization differences will challenge such efforts. Stability, given the theoretical perspectives noted above, has its effects across time and in childhood, this will take years—beyond the scope of most studies. A child register, documenting experiences of children ever in care along with a matched-control group, that cumulates information over time, will perhaps provide the best clues to this relationship. All interpretations will still in all likelihood be clouded by the fact that instability is invariably confounded in effect with outcomes such that the more prevalent a negative behavior the more likely it is that instability will occur. Event history analysis, with the ability to consider contributing factors in transitions from one state to another, seems a good likelihood. Yet, even this approach needs to be carefully implemented to account for on-the-ground measurement issues that will alter the observed relationships.

Recommendation #5: Consider looking at factors associated with destabilizing transitions. There is a greater need to correctly specify models, most notably to include structural variables that may contribute to destabilization.

Recommendation #6: There are several complex measurement issues in such research.

a. It is hard to measure a true baseline behavior in that children enter care at the top of the stress ladder.

b. The outcomes measured are likely to produce differential results.

For example, stability is not the sole answer and not all “positive” outcomes are likely to be achieved. Some outcomes occur under one set of circumstances while others are more likely under a totally different set of circumstances and some outcomes are not compatible with each other. Specify competing relationships between predictors and outcomes (see [232]).

c. Attitudinal outcomes such as satisfaction and happiness are likely to have shifting bases. They have no absolute baseline and therefore cannot be consistently measured across time. Satisfaction on a one to five scale may have the same score and different meanings across time and setting. Choose hard outcomes: network size, income, educational achievement, skill development, and health status of children in and out of care. These should be affected by stability.