Govindshenoy, M., & Spencer, N. (2006). Abuse of the disabled child: a systematic review of population-based studies. Child: care, health and development, 33(5), 552-558.
Childhood disability is often assumed to be a risk factor for child maltreatment, however prior to this recently published review, the evidence for this association had not been systematically reviewed. Children with disabilities were included in this review if: a) they were suspected to have mental limitations, hearing, speech, visual, and/or physical impairments, and/or be emotionally disturbed; and b) as a result of these impairments they were limited in mobility, self-care, language, learning, and independent living. Population-based studies between 1966 and January 2006 were identified and those studies reporting empirical data on the association between disability and child maltreatment, specifically when the disability preceded child maltreatment, were retained for review. Of these, four studies met the inclusion criteria for analysis: two longitudinal studies (n=14,893; n=644), one retrospective birth cohort study (n=119,729), and one cross-sectional survey (n=422). Results showed little evidence that physical disabilities predispose children to maltreatment. However, children with psychological and emotional difficulties were at greater risk of experiencing maltreatment than the general population. Of these children, those with conduct-related psychological difficulties or speech and language impairments were more likely to be maltreated. Without the presence of learning disabilities or behavioral difficulties, disabled children are not at increased risk of child maltreatment.
The studies were examined in accordance with recommendations from the U.K. National Health Service Centre for Reviews and Dissemination (2001). Reviewers scored the studies based on sample size, attrition rate, whether confounding variables were accounted for, and whether disability and child maltreatment were well defined. The search strategy resulted in 107 abstracts, of which 47 papers were reviewed. Four met the criteria for inclusion. Odds ratios with 95% confidence intervals were used but meta-analysis was not considered, given the variations between study populations and child disabilities. Although the selection of studies controlled for sampling biases, several limitations restrict the findings of this review, including:
- the limited number (n=4) of studies included in the analysis, softening the conclusions of the review;
- a wide variety of child disabilities, making it difficult to consolidate findings based on a specific type of disability;
- none of the studies controlled for confounding factors (i.e. parental substance abuse, environmental risk factors, etc) which may reduce the estimated association between child disability and child maltreatment.
Further population-based prospective birth cohort studies are suggested in this area.