“For me context is the key – from that comes the understanding of everything.” Kenneth Noland, American painter, April 10, 1924–January 5, 2010

In May last year, I was preparing materials for social workers on a Post-Qualifying course in Child Care at Queen's University Belfast. The module I was helping with was examining outcomes for children in care and I wanted to set my own research findings within an international context. That is when I first came across Ainsworth and Hansen's (Reference Ainsworth and Hansen2014) article in Children Australia questioning the use of family foster care for vulnerable children. To be honest, I nearly fell off my chair when I was reading it. It was clear that the article was written with the intention of stimulating debate, but it did argue quite strongly that the apparent weaknesses in the family foster care system suggested that it was not fit-for-purpose, and was either not improving outcomes for vulnerable children, or making things worse. The suggestion was that children might be better off remaining at home, with supports, in the context of ‘less-than-optimal parental care’ (Ainsworth & Hansen, Reference Ainsworth and Hansen2014, p. 1). This really concerned me for a number of reasons.

First, in the almost twenty years that I have been researching the lives of care experienced childrenFootnote ¹, speaking to the children themselves, their birth parents, carers and adoptive parents and social workers, and extensively reviewing social work case file material, I had never once come across an incidence of a child entering the care system due to less-than-optimal parenting. For me, this conjures up notions of children perhaps not being given enough reading material at home, or watching too much television. However, the young children whose early lives I was familiar with had not entered care due to this type of minor parental failing, but as a result of experiencing significant harm, or being at a risk of experiencing significant harm, more often than not as a direct or indirect result of their parents’ actions or inactions. Thus, I felt that the use of the term less-than-optimal parenting was quite inappropriate in the context of children's entry to care.

Second, despite our doubts as academics, as to whether or not anyone ever takes anything that we write seriously, social care and legal practitioners actually do read our work. Once material is published it becomes, to some degree, legitimised, and that is when things can get a bit concerning. This is because arguments such as those developed by Ainsworth and Hansen in their article could be presented in Court as justification for the non-removal of at risk children from their birth parents. And it is further concerning when the evidence presented for a particular perspective may be flawed.

Third, the presentation of evidence in the article was particularly lacking in context. I was extremely fortunate as a young academic, not long after the completion of my PhD in 1999, to join a relatively new research unit (Centre for Child Care Research) in Queen's University Belfast that was developing a number of longitudinal studies aimed at addressing some of the key questions in contemporary childcare. One of these was a longitudinal study of children in care, namely the Care Pathways and Outcomes Study, which I was fortunate to lead from 2003. My good fortune was amplified by the fact that the work of the Centre was being overseen at that time by some of the leading scholars in the field, such as Professor Dorota Iwaniec, Professor John Pinkerton, Professor Sir Michael Rutter, Professor Ian Sinclair, and Dr Greg Kelly, from whom I learned so much. Talk about standing on the shoulders of giants!

Although I greatly admired all the members of this esteemed group, as a psychologist, I was particularly influenced by the work of Michael Rutter. At that time, he had just published his now seminal paper on children in substitute care (Rutter, Reference Rutter2000). If you have never read this paper, I would strongly recommend that you do. This focuses on a range of conceptual considerations and research implications. A key argument developed in his paper and reiterated by the aforementioned colleagues at the Centre was that in order to fully understand outcomes for children in care, you need to fully understand the context of their lives, their individual journeys, why they entered care in the first instance, what they experienced prior to entry to care and what happened to them whilst in care. The power of this argument has stayed with me ever since and has acted as a guiding principle in the ongoing development of the Care Pathways and Outcomes study. Unfortunately, it was this type of contextualised perspective that was missing from the Ainsworth and Hansen article.

Consequently, I and my colleague Montse Fargas Malet submitted a commentary to Children Australia challenging the conclusions of the Ainsworth and Hansen article, and this was published last year (McSherry & Fargas Malet, Reference McSherry and Fargas Malet2017). This then encouraged further discussion between myself and the journal editors, Jennifer Lehmann and Rachael Sanders, about the possibility of building upon the commentary and preparing a Special Issue to develop our understanding of outcomes for care experienced children – and the rest, as they say, is history. This special issue, then, presents six papers that aim to further our understanding of outcomes for care experienced children. They range from understanding the impact of early adversity (Hambrick, Brawner, & Perry) to post-care outcomes (Van Breda).

The first paper is a commentary by John Simmonds which discusses the foster care system within its historical context, and establishes the tone for the issue through thoughtfully unpicking a complex array of issues that need to be considered when attempting to fully understand outcomes for care experienced children. The paper acknowledges that children in care are not a homogenous group and that outcomes will vary, often as a function of age at entry, reason for entry and duration of care. The paper emphasises that “foster care is an opportunity to re-establish a framework of resources that enable recovery for the child”. However, it cautions against the common practice of removing social work support for these young people as they leave formal care and enter adulthood because this risks undermining any positives achieved up to that point.

The second paper by Erin Hambrick, Thomas Brawner and Bruce Perry examines developmental adversity and connectedness affecting child welfare-involved children. In addition to also highlighting the heterogeneity of welfare-involved children, it provides a fascinating account of the usefulness of utilising a neurodevelopmentally informed approach to intervention, namely the Neurosequential Model of Therapeutics (NMT), to inform policy and practice regarding welfare-involved children based on an analysis of risk, connectedness and neurodevelopmental functioning. Their findings highlighted that although early life developmental risk has a persistent effect on future functioning, relationally supportive contexts may mitigate these risks. They conclude that the quality of children's relationships is central to positive longer-term outcomes, and that the focus for policy and practice should be upon improving the quality of these relationships, regardless of placement type. This mirrors findings from the Care Pathways and Outcomes study that the quality and longevity of relationships for young children in care are more important for positive outcomes, in terms of attachment and self-esteem, than the social or legal definition assigned to the placement, i.e., foster care, kinship care or adoption (McSherry, Fargas Malet, & Weatherall, Reference McSherry, Fargas Malet and Weatherall2016).

In keeping with the commentary of the previous two papers, the third paper by Anouk Goemans, Mitch Van Geel and Paul Vedder builds on the theme of variability in developmental outcomes for foster children, mostly resulting from the heterogeneity of the care population. They reflect on the findings from a series of meta-analyses, which indicate that once in care, children's functioning in terms of their cognitive, adaptive and behavioural development does not appear to change. How might these findings be interpreted? The indication is that child functioning does not improve significantly when in care, nor does it deteriorate, but remains steady. It is also worth bearing in mind that children's entry to care tends not to be driven by concerns regarding their functioning, but about risk of significant harm. So, it could be argued that removing children from significant harm or the risk of significant harm, without impacting their overall functioning, is a positive outcome.

Goemans, Van Geel and Vedder conclude that, due to the heterogeneity of the care population and the lack of an accurate model for predicting foster children's development, there is a need for greater screening and monitoring of their development from entry to care. If possible, this should begin prior to entry when the child first comes into contact with the social care system and initial child protection processes commence. Such systems would enable timely identification of those foster children at greatest risk of negative developmental trajectories. This echoes recent calls for greater use of screening for children entering the care system in Northern Ireland using Goodman's (Reference Goodman1997) Strengths and Difficulties Questionnaire (SDQ) (McSherry et al., Reference McSherry, Fargas Malet, McLaughlin, Adams, O'Neill, Cole and Walsh2015), as is currently the case in England and Wales, as well as for those systems currently in operation in England and Wales to be further developed (Bazalgette, Rahilly, & Trevelyan, Reference Bazalgette, Rahilly and Trevelyan2015). The authors propose the Brief Assessment Checklist (BAC) (Tarren-Sweeney, Reference Tarren-Sweeney2013) as an alternative screening measure to the SDQ on the basis of their own experience of successfully using the measure with a Dutch sample of foster children (Goemans, Tarren-Sweeney, Van Geel, & Vedder, Reference Goemans, Tarren-Sweeney, Van Geel and Vedder2017).

Paper four comes from myself and my colleague Montse Fargas Malet, in which we attempt to disentangle to some degree the concepts of placement stability and relational permanence. The findings are from the initial stages of the fourth Wave of the longitudinal Care Pathways and Outcomes study, which is being funded by the Economic and Social Research Council (ESRC) in the United Kingdom. The findings reinforce the contemporary literature regarding the capacity of adoption to provide very high levels of stability through early adulthood for children who enter the care system at a young age, and that levels of stability are lower for those who enter long-term foster-care and kinship-care placements. However, although lower than adoption, the levels of stability in long-term care are still considered to be high, particularly, if one tracks back over a 9-year, rather than a 14-year period. However, it is acknowledged that it can be difficult to maintain long-term placements in foster care due to systemic pressures on these placements, such as, leaving care planning processes that can commence as early as when the young person is 14 years old.

A key finding of the study at this early stage has come from interviews with young people (aged 18–22 years) and their parents/carers. For those placements that had broken down (6 of 30), in all bar one the relationships with the carers had continued after the breakdown and persisted currently, with each of the young people indicating that they remained part of the family and considered their carers to be their parents. These findings suggest that the focus on placement stability overlooks the nature of relationships within these placements, and that physical endings do not always directly result in relational endings. As was highlighted by Hambrick, Brawner and Perry, it is the quality of the relationship that appears to matter most.

In paper five, Nikki Luke and Aoife O'Higgins provide compelling evidence from a systematic review and National Database that, despite the multiple pieces of evidence of a marked attainment gap between children in care and their non-care peers, this can be mostly accounted by factors other than being in care. In their analysis of National Database data, the authors disentangle children's educational performance and care status by comparing groups of children who have been in care (for varying lengths of time), children in need who were not in care and children who were both not in need and not in care. They found that although children in care performed more poorly than those who were both not in need and not in care, they performed better than children who were in need but not in care and living with their birth parents, with this difference increasing the longer the period spend in care. In keeping with the conclusions of other contributors to this issue, they argue that their findings on the impact of care duration reflect the heterogeneity of the care population and the importance of considering the needs of different groups.

The issue concludes with a paper from Adrian Van Breda, which develops a highly reflective perspective on the relationship between care factors and post-care outcomes. He presents findings from a residential care programme in South Africa, which indicate that demographic, pre-care and in-care variables all contribute to one-year outcomes. However, he explores these findings in a way that enables him to foreground the complexities in interpreting longitudinal outcome data on leaving care. This discussion very helpfully flags key considerations and challenges for researchers working in this complex area in other countries globally.

Although this collection of papers has come from academics working across a range of countries, with different legislative and policy frameworks, and reflecting a diverse range of research methodologies and questions, two consistent themes have emerged: the importance of relationships; and the need to reflect the heterogeneity of the care population when considering outcomes. For me, both these themes reinforce the importance of context. In terms of relationships, examples of contextual considerations would be the following: what were these like before the child entered care? Were these sustained or allowed to diminish after entry? Were new relationships developed and nurtured whilst in care? To what extent did the quality of these relationships impact upon the child or young person's decision-making over time? In terms of heterogeneity, examples of contextual considerations would be the following: when did the child enter care? What were the reasons for this entry? How long did he/she remain in care? Was he/she male or female? What was his/her ethnic origin? Did he/she have a disability? What services were available to him/her whilst in care or to his/her carers? What was his/her experience of school? Was he/she prepared for leaving care? Was he/she supported after leaving care?

The collection of papers presented in this issue has further demonstrated that a multitude of contextual factors need to be considered when attempting to draw conclusions about outcomes for care experienced children. We need to link these factors up in ways that allow us to feedback to the care system, so that we can learn when and where challenges and opportunities emerge, and use this information to improve provision for this vulnerable group of children. This is not a task for the faint-hearted, but these children are worth the effort.