Developmental screening using the translated PEDS tools in low-income South African settings: Evaluation and adaptation of referral criteria

Abstract

Globally, early childhood development is increasingly recognised as the foundation for a good academic performance later in life. In South Africa the National Curriculum Framework (NCF) together with the side-by-side campaign, aims to establish a strong foundation for early childhood development to combat developmental challenges later in life. Developmental delays are more prevalent in low- and middle-income countries (LMICs) such as South Africa. This is due to children being exposed to a variety of risk factors with limited protective factors. Despite the need, access to developmental screening and early intervention services for children with developmental delays are limited. In South Africa, the Road-to-Health Booklet (RTHB) has been used with limited success for developmental screening. The RTHB is often completed incorrectly or not completed at all. Other developmental screening tools such as the Ages and Stages Questionnaire (ASQ), the Parents Evaluation Developmental Status (PEDS) as well as the Parents Evaluation Developmental Status: Developmental Milestones (PEDS:DM) collectively referred to as the PEDS tools, has been explored for use in the South African context. The PEDS tools has been used in combination due to is affordability and quick administration time. In South Africa, screening children’s development is imperative to improve early identification of developmental delays and have early intervention. A developmental screening tool such as the PEDS tools need to be contextually and linguistically relevant for the South African environment. A contextually relevant tool will enable adequate management of referral rates, while linguistically relevant developmental screening tools will increase accessibility to developmental screening services. Therefore, this research project focused on the use of translated developmental screening tools in low-income settings and to further evaluate the referral rates of the PEDS tools and to adapt it for the South African context. In Study 1 of this research project, a retrospective analysis of the PEDS tools data for 406 participants between the age of five and 36 months who are at risk of developmental delays in a primary healthcare (PHC) setting, was conducted. In Study 2, the translation accuracy of the Northern Sotho and Zulu PEDS:DM in comparison with the English version of the tool, was evaluated. The study was conducted at a regional healthcare facility in South Africa. Stratified convenience sampling was utilised and a total of 546 caregivers representing children from the age of 0 month to 7 years, 11 months were included. In Study 3, the content validity of the translated PEDS tests when compared to the reference English PEDS test, was determined. Similar to Study 2, a stratified convenience sample of 546 research participants from a government healthcare facility in South Africa were recruited. In Study 1, the PEDS test had a high referral rate of 41% (n=168) and the PEDS:DM had a slightly higher referral rate of 52% (n=213). Referral Criteria (RC) of the PEDS:DM and the PEDS tools was computed. The RC prioritises moderately to severely affected children and it is used in conjunction with a surveillance programme to monitor mildly affected children. In Study 2, the findings generally reflected a strong correspondence between the English and Northern Sotho as well as the English and Zulu PEDS:DM tool. There were a few test items in the expressive language domain that did not correspond significantly between the languages which could be due to cultural and social differences and should be investigated. Furthermore, in Study 3 a higher referral rate 42% (n=99) was found for Zulu participants when compared to Northern Sotho participants 17% (n=52). There were significant associations between the Zulu and Northern Sotho PEDS test when compared to the reference English PEDS test. Kappa scores for the Zulu PEDS test ranged from 0.798 (substantial agreement) to 1.000 (perfect association) and for Northern Sotho from 0.395 (fair agreement) to 0.843 (almost perfect agreement). There were significant associations between the Zulu and Northern Sotho PEDS test when compared to the reference English PEDS test. RC of developmental screening tools must be sensitive to the demands on under-resourced healthcare systems. Translated developmental screening tools increase accessibility and in turn, enables timely identification and intervention which allows children to reach their potential. The translations as well as the RC of the PEDS tools are viable for use in prioritised surveillance at baby wellness clinics in low-income settings.