New GP tool to help avoid unnecessary hospitalisations

The Risk Stratification Tool (RST) extracts data from GP clinic systems, including age, smoking status, alcohol consumption and body mass index (BMI), and uses the information to power an algorithm that helps predict a patient’s risk of hospitalisation.
 
Originally developed for the Federal Government’s Health Care Home trial, it was ‘trained’ using information from nearly two million de-identified patient records and has been installed in more than 160 practices.
 
Using data specifically chosen for its common availability in primary care settings, the algorithm has so far identified more than 7000 patients who had a greater than 9% probability of hospitalisation within 12 months.
 
‘This is the first publicly available, scientifically validated Australian risk algorithm that has been used to identify patients in GP practices who are at risk of hospitalisation,’ CSIRO Australian e-Health Research Centre CEO Dr David Hansen said.
 
‘Providing this information to the GP has the potential to improve patient outcomes while reducing healthcare costs.’
 
Dr Jane Cooper, a GP in Davenport, Tasmania, has been using the tool at Don Medical Clinic since 2017. She told newsGP that the tool can be effective, though it is reliant on the practice itself – especially concerning the specific type of note-taking and data entry required – as well as the desire to help identify at-risk patients.
 
‘These tools are driven by certain personalities within your practice. There needs to be a drive to identify those patients with multiple comorbidities, chronic disease, and the risk of sequential hospital presentations,’ Dr Cooper said.
 
‘Having worked in a few different practices with varying levels of attention paid to note-taking, and having opened my own practice … I would imagine that some of the bigger, older practices would struggle because the software just collects so much information.’
 
According to the Australian Institute of Health and Welfare (AIHW), costs associated with long-term illness account for approximately 60% of total Australian healthcare expenditure per year, or roughly $120 billion in 2016–17.
 
Dr Cooper believes that while the work required to effectively use the tool can be time-consuming and require major changes, it is worth pursuing in order to help keep people out of hospital and reduce unnecessary healthcare expenditure.
 
‘If you want to do quality medicine and make a change for your patient community, then these are useful tools,’ she said.
 
‘It’s worth pursuing because I think we do need to change our health system.’
 
The RST links to coordinated care platform cdmNet, while a smart connector also allows it to work with most GP clinical systems, including Best Practice, Medical Director, MedTech, ZedMed, and Communicare.

algorithm chronic illness eHealth