95 Child and Family Social Work 2006, 11, pp 95–106 © 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd Blackwell Science, LtdOxford, UKCFSChild and Family Social Work1356-7500© 2006 The Authors, Blackwell Publishing Ltd11 295106Original ArticleDisabled children and attachmentD Howe Correspondence: David Howe, School of Social Work and Psychosocial Sciences, University of East Anglia, Norwich NR4 7TJ, UK E-mail: d.howe@uea.ac.uk Keywords: attachment, children with disabilities, parent–child interaction Accepted for publication: October 2005 Disabled children, parent–child interaction and attachment David Howe Professor, School of Social Work and Psychosocial Sciences, University of East Anglia, Norwich, UK ABSTRACT Although caregiver factors are generally considered the more potent in determining children’s attachment organization, a number of child factors have also been considered. Among these have been temper- ament and disabilities. The present paper examines the effect of various types of children’s disability on parent–child interactions, including how disabilities affect parental sensitivity and communica- tions. A brief outline of attachment theory and patterns of organiza- tion is followed by a review of the research evidence that has looked at children with disabilities and insecure attachments. A complex picture emerges in which it is not a child’s disability per se that is associated with insecure attachments but rather an interaction between children with disabilities and the caregiver’s state of mind with respect to attachment. Transactions between both child and caregiver vulnerability factors affect sensitivity, communications and security of attachment. Practice implications for prevention, advice and support are considered. INTRODUCTION An early premise of attachment theory was that chil- dren’s patterns of attachment were determined by the characteristics of their caregiving, and that child fac- tors did not play a significant part in their classifica- tion. In the case of children’s temperament this has been found to be broadly true, although over the last two of decades the picture has been modified (see Vaughn & Bost 1999 for a review). For example, a study by Belsky & Rovine (1987) found that although caregiver factors determined whether an infant devel- oped a secure or insecure attachment, the child’s tem- perament did appear to affect the manner in which the security or insecurity was expressed. There is a further suggestion that parental sensitiv- ity might be affected by the child’s innate reactivity, temperamental difficultness, arousability and ability to self-regulate. In this analysis, children who are irri- table, sensitive and easily aroused might place higher demands and increased levels of stress on their par- ents. For example, studies by Susman-Stillman et al . (1996) and van den Boom (1994) found that irritable infants in disadvantaged environments were more likely to be classified insecurely attached. This sug- gests that challenging children cared for by parents under stress act as an additional stressor and so are particularly at risk of developing insecure attach- ments. Summarizing this research, Vaughn & Bost (1999) observe that It may be that when a parent’s economic, social, and/or psy- chological resources are strained, an irritable or otherwise difficult infant elicits less than optimal caregiving, which in turn potentiates the assembly of an insecure attachment. (p. 220) Substituting disability for temperament as a ‘child factor’ allows us to explore attachments and the rela- tionship between parent–child interactions in the case of children with disabilities. The intuitive suspicion is that the behavioural, interactional and communica- tion characteristics of some children with particular disabilities are likely to affect levels of parental stress, quality of caregiving and therefore security of attachment. After a brief outline of attachment theory, some of the communication and interaction characteristics between children with disabilities and their parents are considered. This is followed by a review of research that has looked at the association between