ECD Measurement Hub.

The assessment of children’s developmental and psychological well-being, both in the short and long term is expensive and time consuming. Population-based data on child development is essential to improve the lives of children globally and to substantively advance global efforts to improve child development. Measuring global progress towards these goals will require a suite of new research tools designed to assess key developmental processes, While technology-facilitated innovations have attained some traction, penetration is limited. This is even more so when it comes to novel measurement innovation (for instance, heart rate monitors). This is particularly the case in African contexts, where logistical, structural, and social barriers to measurement are common, and resources for research, limited.

In these early years, the family has the most profound influence on the child and early intervention should be provided by no later than 6 months of age. The long term economic cost to countries is also high. The World Health Organisation (WHO) estimates that children who do not receive early intervention can cost the state up to $1 million in their lifetime.

Despite the need and outcomes achieved, HH, a free, high quality early intervention programme reaching 6 provinces, has reached only 3% of SA children born with hearing loss in the last 12 years. HH has struggled to transition to scale. Adaptations to the model are required in order to expand reach and to leverage HH’s value proposition to facilitate sustainability.

The aim of this venture is to promote the use of emerging technologies to solve problems of child measurement at the population level across Africa. The two innovative tools explored to date:

• Words Count uses the latest available technology – in the form of the Language Environment Analysis System (LENA) – to analyse and segment audio data within children’s everyday environments. The system includes a sensitive recording device for capturing the number and type of adult words spoken to and near the child under assessment over the course of a defined period. This data will help researchers determine which interventions are most effective in enriching home language environments for young children.
• Google Street View Image Analysis can be used to derive indicators of neighbourhood built environments. Software programmes using convolutional neural networks can create indicators of the child-friendliness of the built environment in a given area, and these can be mapped onto individual-level or area-level human health data.

Data-driven decision-making is an essential component of driving meaningful change for children. Exploring these tools can generate information about what interventions work, for who and under what conditions, to inform good practice and the scale-up of these. There is the opportunity to leverage the use of big data emanating from emerging technologies to inform practice and policy for young children in Africa.

Another key element of interest is the adaptation of technologies and euro-centric data models for use in Africa. Tools need to be culturally sensitive and contextually appropriate in order to be sustainable and useful.

Institute of Life Course Health Research at Stellenbosch University