A world-first study by the CSIRO has shown the potential benefits of using AI to improve patients’ engagement and health outcomes from digital health programs.
The study, published in The Journal of Medical Internet Research, used a CSIRO-developed algorithm to accurately predict when a person would drop out of an online weight loss program.
And with 67% of Australian adults currently overweight or obese, 8% of the national burden of disease due to obesity, and about 5% due to dietary risks; the CSIRO said that supporting these type of health interventions and lifestyle changes was more important than ever.
Lead author Dr Aida Brankovic explained there was still little evidence for how people engaged with digital behavioural intervention programs, especially in terms of when and why they quit.
“Recently, there has been an increase in the number and popularity of these programs to help reduce modifiable health risk factors such as obesity and lack of exercise, as engagement in these programs is critical to interventions that achieve successful behaviour change and improvements in health,” she said.
“Yet despite the growing application and adoption of technology in health interventions, one persistent challenge remains – engagement deterioration or non-usage attrition.”
The study used the AI algorithm to analyse data from a large, online, commercial weight loss program, cross-referenced with the information from over 59,000 participants in the CSIRO’s Total Wellbeing Diet initiative, and found that they could accurately predict when an individual was going to disengage from an online program by week three.
“With this information, digital health interventions can become more tailored, supportive, and offer users a greater chance of making long-term lifestyle changes,” Dr Brankovic said.
“And importantly, the machine learning model can be adjusted and applied to large cohorts of data in other online programs requiring engagement.”
Fellow CSIRO researcher and co-author, Dr Gilly Hendrie, said that while effective engagement in a digital health program was vital to a patient’s success, training people to log in and participate was challenged by the fact that it was up to the user’s discretion daily.
“It is hoped that the findings of this study and future work focused on other factors of engagement with digital health programs will lead to improved experiences for users, including tailored content at critical time points,” Dr Hendrie said.
“Past habits really do predict future behaviour, but if we can predict when that behaviour will kick in and adapt so we can re-engage someone – we have a better chance of helping them make long-lasting changes to their health and lifestyle.”