Where are the vulnerable children? Identification and comparison of clusters of young children with health and developmental vulnerabilities across Queensland

Abstract

Background: Developmental vulnerabilities within children in Queensland have a variety of domains; these domains measure the development of children in their first five years. It is crucial to understand how these domains are grouped, or clustered, with respect to population risk factor profiles. These groups inform policy implementation, which can help to provide assistance to the most vulnerable children across Queensland. Methods: K-means analysis was conducted on data from the Australian Early Development Census and the Australian Bureau of Statistics. The clusters were then compared with respect to their geographic locations and risk factor profiles. The results are presented in this paper and are publicly available via an interactive dashboard application in R Shiny. Results: This study presents a comprehensive clustering analysis for child development vulnerability domains in Queensland. In addition, all of the clustering analyses reveal a strong relationship between developmental vulnerability and socio-economic and remoteness factors. In addition, we found that children who attend preschool and whose primary language is English are, in most cases, in the lowest developmental vulnerability cluster. Conclusion: In this study, the performance of the K-means clustering algorithm has been developed to study the clusters inside child development vulnerabilities when analysing the data at the small area level. Further, R shiny application was created, and the feature of the risk factors in each region was studied.