Attachment theory has entered the realm of pop psychology, seemingly providing answers to many common problems. The theory, which is one of the most empirically validated within developmental psychology, revolves around the crucial importance of strong bonds between children and those who care for them.
It shows that the quality of these attachments – often classified as “secure”, “insecure-avoidant”, “insecure-resistant/ambivalent” and “disorganised” – is influenced by parents’ capacity to provide comfort when children are distressed. It also explains how childrens’ attachment quality can influence their social and emotional development. Ultimately, secure attachment is optimal.
In an era of increasing demands on evidence-based child protection decisions, such a theory may seem to be sent from heaven. If attachment classifications are related both to parenting and child development, shouldn’t they inform child protection investigations? In our new paper, examining the use of attachment classifications among child protections workers, we argue that the answer is “no”.
Early attachment theory portrayed the links between caregiving, attachment quality and child development in a somewhat simplified manner. But decades of careful research have since brought important nuances to the picture.
We now know that numerous factors beyond parents’ caregiving capability are also influential for child attachment quality. This includes, for instance, the family’s socioeconomic situation and children’s biological dispositions. For instance, while parents of children with autism are typically equally sensitive to their children’s needs as other parents, their children nevertheless tend to display more behaviour that can be interpreted as “disorganised”.
Insecure attachment increases the risk of developmental problems, such as conduct problems, depressive symptoms and anxiety. But as many as 40%-50% of all children in the total population are classified as insecurely attached. And a substantial portion of these children don’t have any major problems in life.
The “gold-standard” observational measures of child attachment have not been developed for diagnosing individual children. Instead, they are used to examine how groups of children differ, on average, from other groups.
Even in the most validated procedures for assessing child attachment, individual children may therefore be erroneously classified. For instance, some children may, for reasons having to do with the observational setting, experience very high levels of stress during the assessment. Such stress may cause “disorganised behaviour” in the child even in the absence of a disorganised attachment history with their caregiver.
Classifications of attachment quality therefore cannot provide information about individual children’s caregiving history or their future development. At least not with the degree of certainty you’d need for use in child protection investigations and decision-making.
Recent reports from many western countries indicate that such nuances are often overlooked in child protection settings. Specifically, attachment quality seems to be invoked often – and in a simplified and overconfident manner.
Such reports raise questions about the legal certainty for the families subjected to investigations. The development has caused concern among attachment researchers. A recent consensus statement arguing this was signed by 70 scholars from five continents. Yet, there has been a remarkable lack of research into these matters.
Our research is based on a nationally representative sample of almost 200 Swedish child protection workers. It constitutes a first systematic study on how attachment theory is used in child protection. And it shows that the concerns of the researchers are not exaggerated.
Our questionnaire-based study examined how often child protection workers form opinions about attachment quality in their investigations. It also looked at why they do so, what methods they use and what implications they attribute to children’s attachment quality.
We found that it was very common to form opinions about attachment quality. In fact, the vast majority of the respondents – eight to nine out of 10 – formed opinions about attachment in all or most of their investigations. These involved children up to 12 years of age.
We also found that these opinions were rarely underpinned by the use of systematic assessment protocols. None of the child protection workers based their opinions on validated protocols developed for assessing child attachment quality.
Nine out of 10 respondents agreed completely or to a large extent that attachment quality provides information about the quality of care that a child has received. They also thought it sheds light on the individual child’s future development.
Almost as many – roughly eight out of ten – agreed that a child’s attachment quality reveals the parents’ current caregiving capacity. They also agreed that insecure attachment signals “insufficient caregiving”. This suggests child-protection measures may be considered. A clear majority also stated that attachment quality can be used as a basis for decisions to place a child in foster care.
No participant reported formal training in attachment theory. Nor were they trained in using validated attachment measures for children. Yet the vast majority nonetheless deemed that forming opinions about children’s attachment quality made their investigations more reliable. Almost 40% regarded classification of attachment quality as relatively straightforward.
These findings are important since they attest to a wide gap between researchers and practitioners – one that may adversely affect the quality of child protection investigations. This gap probably results from a number of factors. These include insufficient research dissemination by attachment scholars, pressures on child protection workers and increased demands to motivate decisions in evidence-based terms.
To counter this gap, we strongly encourage attachment researchers to spread their research more widely. Ultimately, attachment researchers and scholars in social work should join forces and develop rigorous evidence-based assessment methods. Child protection workers also need more extensive training in attachment theory and how to translate it.
Mårten Hammarlund receives funding from The Swedish Research Council.
Pehr Granqvist receives funding from The Swedish Research Council.
Tommie Forslund receives funding from the Swedish Research Council.