Author
GOVINDSHENOY M.; SPENCER N.;
Abuse of the disabled child: a systematic review of population-based studies.
Journal citation/publication details
Child Care Health and Development, 33(5), September 2007, pp.552-558Summary
Four studies, including two from the UK, are included in this review. There is no evidence that autism or sensory impairment predispose to abuse, and limited evidence for cerebral palsy. There is rather more evidence of an association between abuse and psychological, emotional and learning disabilities but this may arise not because of the conditions but because of dysfunctional parenting and other family factors. The overall evidence base is weak and practitioners should avoid the assumption that disability necessarily heightens the risk of abuse or neglect.
Context
Child abuse is a significant cause of death in childhood but research on its causes has primarily focused on the perpetrators, with less attention given to child-related variables. One such variable is childhood disability, but the relationship with abuse has not to date been systematically reviewed.
Methods
What sources were used?
The following databases were searched: Medline, January 1966-January 2006; EMBASE (Excerpta Medica), 1986-2006; CINAHL (Cumulative Index to Nursing and Allied Health Literature), 1990-2006; Social Sciences Database [this may be Social Sciences Index; no dates given]; PsycINFO [no dates given]; and EBSCO Online [no dates given; this is an electronic journals service]. The Cochrane Library and the National Research Register [of research funded by, or of interest to, the NHS] were also checked. Three named journals were hand-searched, and references in relevant papers identified by the searches were checked.
What search terms/strategies were used?
The search terms listed are: disability, handicap, abuse, maltreatment, special needs, exceptional child, vulnerable child.
What criteria were used to decide on which studies to include?
Disability and abuse were defined in relation to named US statutes. Eligible studies were population-based and reported primarily empirical data on the association of child abuse with disability. Only studies in which it was reasonable to assume that the disability preceded the abuse were included. Exclusion criteria are also noted.
Who decided on their relevance and quality?
The searches yielded 107 papers, of which 47 were acquired in full text as potentially relevant. These were independently assessed by the two reviewers following published methods recommended by the Centre for Reviews and Dissemination at York University. Disagreements were resolved by discussion. Each study was scored for quality on the basis of sample size, attrition rate, accounting for confounding variables, definition of disability, and definition of abuse.
How many studies were included and where were they from?
Four studies were reviewed and are summarised in Table 1, which gives details of study type, population, sample size, quality score, type of disability and odds ratios. They were conducted in the UK (2), Chile (1) and the USA (1). The lead author on the largest study appears to be the second author of the review.
How were the study findings combined?
Odds ratios with 95% confidence intervals were used, either obtained direct from the studies or calculated from available data. Meta-analysis was not possible because of the variability of the studies.
Findings of the review
Children with some, but not all, disabling conditions seem to be increased risk of abuse and/or neglect. On the basis of the findings from the largest study (of nearly 120,000 English children), autism and sensory disorders (hearing and visual impairment) do not seem to predispose to abuse or neglect. The same study also shows that children with cerebral palsy may be at increased risk of physical abuse and neglect ‘although the association with all forms of abuse does not reach conventional levels of statistical significance.’ In neither case are the findings of a single study sufficient to come to a definitive conclusion about the relationship between these disorders and the risk of abuse.
Findings from three of the studies suggest that psychological and emotional problems, especially conduct disorders and learning difficulties, pose an enhanced risk. However, ‘this association may arise not because these problems predispose to abuse but because the dysfunctional parenting and other family factors that contribute to psychological and emotional problems may also culminate in abuse.’
Although all four studies ‘reported significant associations of disability with child abuse’, there were significant differences in the types of disability studied, the definition and category of abuse, and the magnitude of the association. The studies were also of variable methodological quality. Although avoiding the bias associated with studying samples from highly selected populations, there were wide variations in the definition of disability and none of the studies adjusted for all potential confounding variables.
Authors' conclusions
The evidence base in this area is ‘relatively poor’ and interpretation of the findings should accordingly be ‘cautious’. There is a need for further high quality, population-based research, preferably based on a prospective birth cohort large enough to examine the association between abuse and different forms of disability. A possibly cheaper alternative would be population-based case control studies involving samples of abused or disabled children and control groups randomly selected from the same populations.
Implications for policy or practice
‘The main implication for practice from this review is that practitioners should avoid the assumption that children with any form of disability are more vulnerable to abuse and/or neglect.’ Although they need to remain alert to the possibility, there is currently insufficient high quality evidence to justify the assumption of higher risk, ‘with the possible exception of children with psychological, emotional and learning disabilities.’