The principal objective of the project is:

• to bring together academic and business research organisations in a joint partnership integrating technological and social innovation,
• in order to develop proof-of-concept of a scalable and cost effective personalised digital health program for (pre)obesity patients,
• by integrating, in an ethical manner, adaptive AI components aimed at personalisation of the program as well as its cost management,
• with the ultimate objective to support sustained behavioural change leading to adoption and internalisation of healthy lifestyle choices.

This research project is aimed at leveraging digital tools to personalise preventive healthcare interventions for at-risk populations in ways that address the above-mentioned challenges. At risk populations, in the context of this project, generally means: Younger to middle aged (aged 30-50) sedentary individuals with (pre)obesity expressing behavioural, clinical, biochemical and anamnestic indicators of increased risk for accelerated decline of cardio-metabolic and cognitive health.

Primary outcomes will be the rate of wearable adoption/use, individualised behavioural change persistence and elicited health benefit

Secondary outcomes will include step count, activity of medium & vigorous intensity, sleep duration/efficiency, heart rate, quality of life, body weight, body composition, blood pressure, and level of fitness as determined by the wearable or validated questionaires available on the landing page.

Participants will be randomly assigned into one of the three intervention groups and a control group. The three intervention groups:

1. Self-health coaching will be conducted using wearable and technology provided nudges allowing personalised technology-assisted real-time feedback. With this approach, on-line intervention with minimal human resource costs and high technology involvement with a potential to deliver sustainable and cost-effective behavioural change to a large at-risk population will be tested.

   This intervention approx. 80 participants will be examined for the duration of 14 weeks. Long term follow up examination is envisaged, but it is beyond this study duration and will be supported by subsequent funding.

2. Health coaching (online health-coach community based) intervention will be based on applying wearable providing the most effective impact on health-related outcomes as evidenced in the exploratory study in combination with the online supervised intervention by the professional health-coach in an online group (community setting). Within this intervention the efficacy of an online supervised health coaching applied in community settings will be tested. Professional health coaching strategy using wearable-related outcomes will be provided as a golden standard of the behavioural intervention with proven efficiency. This intervention will be tested on approx. 80 participants for the duration of 14 weeks.

3. Peer coaching: Peers will be identified from the study population based on their response to specific community challenges. Natural leader characteristics will be deciphered from knowledge provided by the past supervised community-based behavioural interventions at BMC SAS and the other open sources subjected to AI driven prediction modelling. Within this intervention the efficacy of an online health coaching led by peers will be tested in an online community based setting. Peer-based health coach will lead participants through health training based on the "protocol" defining minimal coaching criteria but imposing no limits to any communication activities including the real-time access to the wearable-provided outcomes. The intervention will be tested on approx. 80 participants for the duration of 14 weeks. Long term follow up examination is envisaged, but it is beyond this study duration and will be supported by subsequent funding.

4. Control group participants will receive health coaching - related information enabling self-health coaching in a printed form during initial phenotyping visit and a wearable. The control group will consist of approx. 50 participants and observation period will be 14 weeks. Long term follow up examination is envisaged, but it is beyond this study duration and will be supported by subsequent funding.

Innovative strategies to employ AI and machine learning will be tested to support their meaningful (providing real health benefits) and cost effective use.

Inclusion Criteria:

• obesity (BMI 27-42 kg/m²)
• capacity and willingness to use smartphone
• capacity and willingness to use wearables
• readiness to change (validated questionaire)

Exclusion Criteria:

• diseases that prevent participation in the program, or insufficiently controlled diseases, or terminal-stage illnesses
• excessive alcohol consumption or a history of other addictions
• patients currently participating in a weight-loss program (commercial, medically supervised, or a research study)
• patients using anti-obesity medications who are in the dosage adjustment phase within the last four weeks
• patients who are not mentally, physically, or technologically capable of participating
• patients who are unable to provide informed consent
• patients not adhering to the study protocol

Data will be provided in scientiffic publications. Anz data related to this study could be available upon reasonable request after their publication.