Assessing school readiness outcomes in young children admitted to the pediatric intensive care unit using machine learning and population-based registry data in Queensland, Australia (2023–2025)
- Abstract:
- Each year, over a quarter of a million children are admitted to pediatric intensive care units (PICUs) in the USA, the UK and Australia and New Zealand.1-3 These children are the sickest, most vulnerable children in our communities and although the PICU mortality rate in developed countries has reduced to 2-4%;4, 5 one third of PICU survivors experience physical, cognitive, and mental health impairments beyond PICU discharge,6-9 and as such, supporting children post-PICU is at the forefront of research in our field. Due to a lack of evidence in screening and follow-up programs for the broader PICU population, standardized follow-up for most PICU survivors in Australia is sorely lacking. Evaluation of long-term outcomes for PICU survivors is needed so that earlier referral and access to services can be provided before these children start school, to mitigate the long-term impacts of critical illness. This project will use a unique population-based dataset of all PICU admissions from 1997 to 2019 in the state of Queensland, Australia (24% of all national PICU admissions), linked with data from the Australian Early Developmental Census (AEDC), which assesses school readiness outcomes of Australian children in their first year of school. We will use machine learning models to identify children with developmental vulnerabilities across multiple domains and identify risk factors measured at PICU admission that are associated with these vulnerabilities. These models will be used to develop a decision support system, codesigned by clinicians and consumers, that will aid in the planning of targeted interventions and referrals.
- Grant type:
- ZOLL Foundation Grants
- Researchers:
-
-
- Funded by:
- The ZOLL Foundation
