WA-based Telethon Kids Institute is using big data and analytics to provide the empirical evidence to inform policy and planning changes.
The not for profit, which aims to protect and improve the lives of children, analyses large government and administrative to provide insights to agencies to enable more efficient and effective delivery of services.
Scott Sims, the Institute’s Data Analyst, told Which-50 it is difficult for the agencies responsible for delivering these services to make changes without rigorous empirical evidence.
This requires Telethon to prepare complex data sets uses de-identified, longitudinal, population-based data for research and analysis.
For example, in 2016 Sims assisted with a study aimed at understanding the risk of child maltreatment across different disabilities.
While Telethon already knew that disabled children experienced elevated rates of abuse and neglect, the study showed maltreatment risk is not consistent across all disability types.
“Children with intellectual disability, mental problems, and conduct disorders had increased risk, but not autism, Down syndrome, or birth defects,” Sims said.
The study highlighted how different risk levels relate to different disability types, meaning the services or information provided could be better targeted. The findings were shared with the Australian government to provide further evidence that a prevention and early intervention framework is needed.
The research provides “an evidence base they can use to help with their policy and planning. Without that evidence base it’s very hard to make changes, real change,” Sims said.
Telethon predominantly uses SAS analytics to prepare and process the data used in these studies. The system is able to handle the loading and analysing of files comprising more than two million records and 100 variables.
Time saving is a key benefit of the software, Sims said.
“With SAS what’s very useful with that is how powerful it is and how it allows us to process and manage the data, manipulate the data quickly, efficiently and accurately whereas some other software we have used in the past just can’t handle data of that size.”