Retrospective and concurrent predictors of the mental health of children in care

Abstract

The paper reports retrospective and concurrent predictors of baseline mental health estimates for 347 children recruited to an epidemiological study of children in foster and kinship care in New South Wales, Australia. The study design addresses limitations of previous risk-related research with this population. While a large number of potential risk and protective factors were associated with children's mental health, most were confounded. Predictors identified in two-stage hierarchical regression models accounted for 0.31 of the variance of continuous mental health scores. The key predictors were: “age at entry into care”; developmental difficulties (intellectual disability and reading problems); three specific types of maltreatment; recent adverse events; and factors referring to placement insecurity or lack of permanence. Mechanisms accounting for these findings are proposed, with particular reference to cumulative exposure to adversity, attachment theory and foster care practices. Implications for social care policy include the need to identify children who are in need of care at younger ages, and the harmful effects of perceived impermanence for children in long-term care.

Introduction

Children residing in alternate (or “out-of-home”) care manifest exceptional frequency and severity of emotional and behavioral difficulties (EUROARC, 1998, Heflinger et al., 2000). Indeed, the scale of these difficulties more closely resembles that of clinic-referred children, than of children at large (Tarren-Sweeney & Hazell, 2006). The mechanisms that determine developmental change and resilience among children in care are not well understood. Information about other high-risk groups suggests they include complex, time-sensitive interactions between genotype, pre-natal conditions, ‘pre-care’ and ‘in-care’ psychosocial conditions and events, and early neurological development (Rutter, 2000). But, developmental psychopathology models pertaining to maltreated children (Cicchetti, Toth, & Maughan, 2000) and profoundly deprived inter-country adoptees (O'Connor et al., 1999, Zeanah et al., 1997) are only partially valid for children in care, since there is convergence and divergence of their respective developmental pathways. Certainly however, the attachment experiences of children in care are likely to have profound effects on their development, and especially their mental health (Cicchetti et al., 2000, Fonagy, 2003, O'Connor and Rutter, 2000). Children in care are at very high risk of attachment insecurity (Marcus, 1991) and of attachment disturbances (Minnis, Everett, Pelosi, Dunn, & Knapp, 2006), given their common experience of emotional deprivation, loss, and inconsistent caregiving. Attachment theory also predicts that the therapeutic potential of alternate care should vary according to: 1. the characteristics of children's attachment systems at entry into care; and 2. carer sensitivity and ability to provide a ‘secure base’ (Bowlby, 1988, Schofield, 2002).

The few studies that have identified correlates and predictors of mental health problems among children in care have largely been compromised by methodological limitations (Taussig, 2002), specifically: lack of retrospective and/or prospective design (most are cross-sectional surveys); small sample size; problems with sample retention; limited scope of mental health outcomes (failure to measure problems that are specific to high-risk populations); inadequate measurement of pre-care risk exposure; and failure to control for confounding. Despite the significance of these children's attachment experiences, risk studies have typically aggregated mental health outcomes without reference to their developmental histories. It is premature to identify the effects of foster care from such aggregated data, even where changes are measured prospectively.

Among other purposes, the Children in Care Study (CICS) was designed to address some of these limitations, with a view to identifying prospective, concurrent and retrospective predictors of mental health. The CICS is a prospective, epidemiological study of the mental health of children in court-ordered foster and kinship care in New South Wales (NSW), Australia, and of associated study factors. The present paper identifies retrospective and concurrent predictors of children's baseline mental health problems, from a large number of associated study factors, and proposes some likely developmental mechanisms. Prospective predictors will be identified in a forthcoming follow-up study.

In the decade 1996–2006, the number of children residing in alternate care in NSW rose from about 5500 to almost 10,000 (82% increase) (Australian Institute of Health and Welfare, 2007). In that period, the rate of NSW children residing in care rose from 3.4 to 6.2 per 1000 children. The equivalent rate for NSW indigenous children in 2006 was 44.7 per 1000 children. Almost all pre-adolescent children who enter court-ordered care do so following maltreatment (abuse and/or neglect) by their parents, or because of parental inability to protect them from maltreatment. In the early 1990s the state's large residential care facilities were closed as part of a reform of alternate care services. At the time of the present study (2000–2003), less than 5% of placements were in non-family type settings (Australian Institute of Health and Welfare, 2004). Conversely, kinship care has expanded in NSW to the point where it is now the predominant form of alternate care (56% of all placements) (Australian Institute of Health and Welfare, 2007). There are two distinct kinship care systems in NSW, demarcated by custody (or ‘parental responsibility’) rights. The Children's Court may assign such rights to kin, with the child welfare department (Department of Community Services — DOCS) providing backup support. Alternatively, custody may be assigned to DOCS, with kinship carers having similar legal status to foster carers. Until recently, permanency planning in NSW lacked legislative support. Adoption by foster and kinship carers is uncommon for children under 12, given difficulties of gaining parental consent and running contested cases in the state's Supreme Court. However, NSW made some progress in this direction, with introduction of permanency provisions to the Children's Act in 2001.