You are viewing this site in staging mode. Click in this bar to return to normal site.

Harnessing the power of artificial intelligence for behavior change

The Internet and uptake of mobile devices has facilitated the democratisation of healthcare. Digital education programmes are generating patient ‘big data’ – in terms of volume, velocity and variety. This can be utilised to apply behaviour change techniques and optimise engagement that far exceeds traditional benchmarks.

The Low Carb Program is a completely automated, structured 10-week digital health intervention personalised for adults with type 2 diabetes, prediabetes and obesity. The intention of the program for people with type 2 diabetes and prediabetes is to place the condition into remission. The effectiveness of the intervention is measured by patient engagement and clinical health outcomes. The Low Carb Program has a 70% engagement at 1 year follow up, with 1 in 2 people who complete the Low Carb Program placing their type 2 diabetes into remission.

Participants are given access to nutrition-only modules, with a new module available each week over the course of 10 weeks. Lessons are consumed by participants according to their learning style, predominantly videos, written content, or podcasts of varying lengths (approximately 3 to 12 minutes long). Participants are encouraged to track their health data during the program. Of note, the program tracks mood, blood glucose and weight. Participants also have access to community-led peer support built into the digital interface of the program. Through the discussion board participants are able to view and ask questions and find support from other people who are currently going through or have completed the program.

The Low Carb Program is underpinned by the Ecological Model of Behaviour Change and utilises a solution-focused approach to motivate individuals to meet their self-selected goals to lose weight, reduce medication requirements, or make healthier choices for their whole family.

Data footprints of participants using the program are extracted, synthesised and interpreted to identify patterns for optimal engagement and success. This is mapped against demographic, behavioural, health beliefs and engagement to create a knowledge base which can be interrogated and learned upon to enhance the usefulness and support of the implementation, and the overall intervention effectiveness. This provides an intelligence to the program that is constant with continuous learning conducted in real-time. Evaluation of the performance and utility of the implementation can be measured through several key metrics, predominantly engagement and impact on clinical health outcomes. This can be used to determine the optimum engagement of the intervention.  

The benefit of a data-driven approach to learning and engagement is that the growth of the knowledge base increases exponentially, uncovering new insights which are subsequently used to optimise engagement and uptake further. This iterative approach to learning is redefining the understanding of chronic illness. Type 2 diabetes was once considered to be chronic and progressive and is now being placed into remission through a digital-only approach. 


Arjun Penesar, co-founder and CEO, 



Charlotte Summers, Chief Operating Officer,