Childhood disability population-based surveillance: Assessment of the Ages and Stages Questionnaire Third Edition and Washington Group on Disability Statistics/UNICEF module on child functioning in a rural setting in South Africa

BACKGROUND Epidemiological information on childhood disability provides the basis for a country to plan, implement and manage the provision of health, educational and social services for these vulnerable children. There is, however, currently no population-based surveillance instrument that is compatible with the International Classification of Functioning, Disability and Health (ICF), internationally comparable, methodologically sound and comprehensively researched, to identify children under 5 years of age who are living with disability in South Africa and internationally. We conducted a descriptive pilot study to investigate the sensitivity and specificity of translated versions of the Ages and Stages Questionnaire Third Edition (ASQ-III) and the Washington Group on Disability Statistics/UNICEF module on child functioning (WG/UNICEF module) as parent-reported measures. The aim of our study was to identify early childhood disabilities in children aged 24-48 months in a rural area of South Africa, to determine the appropriateness of these instruments for population-based surveillance in similar contexts internationally. METHODS This study was conducted in the Xhariep District of the Free State Province in central South Africa, with 50 carers whose children were registered on the South African Social Security Agency (SASSA) database as recipients of a grant for one of the following: Care Dependency, Child Support or Foster Care. The researchers, assisted by community healthcare workers and SASSA staff members, conducted structured interviews using forward-backward translated versions of the ASQ-III and the WG/UNICEF module. RESULTS Both measurement instruments had a clinically meaningful sensitivity of 60.0%, high specificity of 95.6% for the ASQ-III and 84.4% for the WG/UNICEF module, and the two instruments agreed moderately (Kappa = 0.6). CONCLUSION Since the WG/UNICEF module is quicker to administer, easier to understand and based on the ICF, it can be considered as an appropriate parent-reported measure for large-scale, population-based as well as smaller, community-specific contexts. It is, however, recommended that future research and development continues with the WG/UNICEF module to enhance its conceptual equivalence for larger-scale, population-based studies in South Africa and internationally.