Can we improve attachment or attachment-related outcomes in young children?

Last updated 21 February 2020

The concept of attachment, which was first developed by Bowlby in the 1960s, refers to the capacity of a child to be comforted by their primary caregiver and also to be able to use them  as a ‘secure base’ from which to explore the world. Since the concept was first developed a number of ways of measuring attachment have been developed perhaps the most significant being the Strange Situation Procedure. This classifies children as secure; insecure avoidant or insecure anxious/ambivalent. More recently, a new category known as disorganised attachment was developed to capture the contradictory behaviours of young children who are frightened of their caregiver.

Since then, a number of longitudinal studies have been conducted to track the outcomes for children who fall into different attachment classifications.  This research shows quite convincingly, that attachment is a significant source of risk/resilience for young children. Thus, secure attachment has been shown to be significantly associated with a range of improved outcomes for children across a number of key aspects of their functioning including, emotional, social and behavioural adjustment, school achievement and peer-rated social status. While both insecure and disorganised attachment are associated with a range of later problems including externalising disorders, dissociation, PTSD, and personality disorder.  For example, one longitudinal study of children with disorganised attachment at 1-year of age, found that by 6 years of age the children were showing signs of controlling behaviours towards their parents, avoidance of their parents, dissociative symptoms, behavioural/oppositional problems, emotional disconnection, aggression towards peers and low social competence in preschool.

In addition to this longitudinal research, other studies have begun to chart the prevalence of the different attachment classifications. This research suggests that irrespective of geographical or cultural location, around two-thirds of children are securely attached, and that disorganised attachment has a prevalence of 15–19% in population samples, up to 40% in disadvantaged populations and 80% in maltreated populations.

We also now know quite a lot about the factors that are associated with different attachment classifications.  One of the earliest pieces of research (a systematic review of 12 studies) found that parental sensitivity was a significant predictor of attachment security. However, this study also found that such sensitivity only explained around one third of the total variance. Studies conducted since then have been successful in identifying a number of other factors that are important.  So, for example, research has identified the importance of the specific nature or quality of the attunement or contingency between parent and infant, the parent’s capacity to understand the infant’s behaviour in terms of internal feeling states (termed ‘parental reflective function’ or ‘mind-mindedness), and a range of anomalous forms of parent–infant interaction.  Research also clearly shows the parents capacity to provide this sort of care is influenced by a range of factors such as poverty, parental mental health problems, and domestic abuse.

This research points to the importance of promoting resilience in early childhood by supporting parents in providing the type of parenting that is associated with a secure attachment, and also in working to reduce the type of parenting that is associated with a disorganised classification. But what do we know about whether this is possible or works?

In order to address this question, we conducted a systematic search of key electronic databases to identify reviews and any RCTs that have been published since the reviews (i.e. between 2008 and 2014). We found 6 systematic reviews and 11 randomised controlled trials that had evaluated the effectiveness of universal, selective or indicated interventions aimed at improving attachment and attachment-related outcomes in children aged 0–5 years.

This review identified a number of methods of working with parents as being promising approaches to improving attachment in a range of high-risk infants, including those with maltreating parents, including parent–infant psychotherapy, video feedback and mentalisation-based programs. These and other interventions, such as home visiting and parenting programs, appear to be effective in improving a range of attachment-related outcomes, such as aspects of parent–infant/toddler interaction related to maternal sensitivity and reflective functioning.  Perhaps most importantly, the findings of this review were consistent with the findings of earlier systematic reviews.

The theories of change underpinning the different programs are diverse and range from psychoanalytic models (e.g. parent–infant psychotherapy) that focus primarily on changing the parents’ internal working models, through programs that focus explicitly on improving parents’ capacity for reflective functioning (e.g. Minding the Baby, the Mother and Toddler Program) to those that focus more explicitly on the interaction between the parent and infant/toddler, and on sensitive parenting based on attachment theory (video feedback programs). There is, however, an increasing eclecticism, with programs focusing explicitly on parent–child interactions drawing on different theoretical traditions, and many (apart from the home visiting program) building on the use of video feedback.

There is also considerable divergence in the frequency and duration of interventions, with home visiting programs such as MTB involving intensive visits over a prolonged period of time, and most other types of program, involving intensive work over brief periods of time, typically a few months (e.g. Video-feedback and parent–infant psychotherapy). The limited evidence available regarding the comparative effectiveness of these interventions shows that there is little difference between them, and increasing evidence supports the use of brief, sensitivity-focused interventions.

Although some of these interventions need to be delivered by specialist practitioners (e.g. psychologists and parent–child psychotherapists), many of the remaining interventions are manualised (e.g. ABC, VIPP), and some can be delivered effectively by health visitors as part of the Healthy Child Program, following appropriate training (e.g. video-feedback). There is a high prevalence of disorganised attachment, particularly in disadvantaged populations, and the strong association between such attachment patterns and later problems suggests the need for specialist CAMHS practitioners to also have the necessary skills to deliver some of these modes of working.