Empowering Patients With Chronic Disease Using Profiling and Targeted Feedbacks Delivered Through Wearable Device (EMPOWER)

Empowering Patients With Chronic Disease Using Profiling and Targeted Feedbacks Delivered Through Wearable Device

Empowering Patients With Chronic Disease Using Profiling and Targeted Feedbacks Delivered Through Wearable Device (EMPOWER)

Chronic diseases are the leading cause of deaths in Singapore. The rising prevalence in chronic diseases with age and Singapore's rapidly aging population calls for new models of care to effectively prevent the onset and delay the progression of these diseases. Advancement in medical technology has offered new innovations that aid healthcare systems in coping with the rapid rising in healthcare needs. These include mobile applications, wearable technologies and machine learning-derived personalized behaviorial interventions. The overall goal of the project is to improve health outcomes in chronic disease patients through delivering targeted nudges via mobile application and wearable to sustain behavioral change. The objective is to design, develop and evaluate an adaptive interventional platform that is capable of delivering personalized behavioral nudges to promote and sustain healthy behavioral changes in senior patients with diabetes. The aim is to assess the clinical effectiveness of real-time personalized educational and behavioral interventions delivered through wearable (FitBit) and an in-integrative mobile application in improving patient activation scores measured using the patient activation measure (PAM). Secondary outcome measures include cost-effectiveness, quality of life, medication adherence, healthcare cost, utilization and lab results. Together with the experts from the SingHealth Regional Health System and National University of Singapore, the investigators will conduct a randomized controlled trial of 1,000 eligible patients. This proposal aims to achieve sustainable and cost-effective behavioral change in diabetes patients through patient-empowerment and targeted chronic disease care.

Traditional healthcare facility-based consultation model of episodic contact in managing chronic disease patients have limited exposure to monitor and intervene patients' lifestyle factors. These factors have been found to be more effective in managing 3H than medication. The proposed adaptive platform will utilize wearable and mobile application technologies which has the ability to continuous track several physiological and lifestyle factors data (e.g. moderate to vigorous active minutes, resting heart rate, sleep hours and quality and dietary habits) Similarly, due to the limited exposure that healthcare workers have with patients under the current consultation model, current health education and intervention tends to be "one size fits all", passive and "top down" knowledge-loading. Patients are expected to change their behavior or to remember health education knowledge after a consultation session. The proposed adaptive platform will be built using educational and behavioral cues obtained from multiple stakeholders (including patients) and multiple data sources with the aim to gather more comprehensive and targeted feedback that is relevant to patients' needs in their management of their 3H condition. As changes in lifestyle factors and habits takes time, the proposed platform can also provide timely and appropriate feedbacks and reminders to patients at a more constant interval as compared to current model of care when advice was only given during consultation follow-up To be able to add healthy years to the life of the current and future seniors,behavioral interventions that are closely studied and carefully implemented without disruption to the daily activity of the seniors is needed to achieve a revolutionary improvement in current primary care management. The investigators will conduct a qualitative study to have a deep and enriched understanding of the types of nudges that are suited for patients with chronic diseases. Through modelling approach using the electronic medical records, the proposed adaptive platform will profile patients into groups and pre-set the nudges that are suitable for them. This allows the investigators to identify patients that have a higher risk of complications of 3H and quickly match the desired nudges to change behavior. The proposed adaptive platform also aims to empower patients by providing patients with automated bite-sized knowledge of their health conditions. Coupled with real-time personalized feedback to their health behaviors, patients will be equipped with the knowledge to take charge of their health using far lesser healthcare manpower and resources. The proposed adaptive platform will be integrated into common mobile wearable which are readily available devices that are widely used by many Singaporeans now. As such it can also be scaled up relatively easily with minimal resources and education. Therefore, the proposed adaptive intervention will improve health outcomes and reduce healthcare utilization. An empowered patient will result in lesser complications and improve health outcomes, resulting in lower patient and caregiver burden, improving quality of life.

A pragmatic 2-arm (1:1) randomized controlled trial (RCT) on 1,000 eligible diabetes patients using the Pragmatic Explanatory Continuum Indicator Summary Framework-2 (PRECIS-2) criteria for pragmatic trials. Patients with diabetes will be randomly allocated in a 1:1 ratio to either the intervention or control group. The intervention group will receive the personalized feedback intervention through the personalized and adaptive intervention platform app on a FitBit wearable on top of their usual clinical care for their diabetes condition. The control group will receive the FitBit wearable on top of their usual clinical care for their diabetes condition but will not receive the personalized and adaptive intervention platform.

Patients will be screened and recruited for the RCT by research coordinators positioned in the SingHealth polyclinics. They will identify eligible patients according to the inclusion and exclusion criteria. Informed consent will be taken and they will be referred to the research coordinators who will randomly assign the patients to the intervention or control arm using a site-specific pre-generated randomization list. A research coordinator will keep custody of the 3 randomization lists (1 for each recruitment site), and assign treatment accordingly to the intervention listed and not be involved in the recruitment or assessment of patients.

Patients in control arm will have FitBit. However, there are no personalised nudges given to the patients in the control arm. Occasional reminders to encourage adherence to wearing of the FitBit will be sent.

Patients in the intervention arm will be given a FitBit device and will be encouraged to wear it as often as possible. Using FitBit built-in tracking technologies such as PurePulse and SmartTrack54, patient's daily activities such as number of steps taken, sedentary time, heart rate, sleep time and exercise will be captured and synced to the adaptive intervention platform as developed in Phase 2 for real-time tracking.

Behavioral nudges will be delivered to patients' FitBit device through adaptive intervention platform via notification syncing. To ensure the delivered nudges are timely and personalized, predictive nudges will be developed based on patterns in patients' sociodemographic, clinical and baseline activity tracking. These nudges will be sent automatically to patients upon specific triggers. The nudges will also be assessed for its effectiveness in behavior change. For example, a predictive nudge to encourage patients to take a short walk after detecting long periods of sedentary time will be assessed for its effects by step counts data after delivery of nudge. An iterative approach will be used to generate an effective set of nudges and its most appropriate delivery times for specific activity patterns.

Inclusion Criteria: Aged 40 and above at time of recruitment Have been diagnosed with diabetes at time of recruitment Most recent HbA1c more than or equal to 7.0% mmol/l Physically able to exercise Literate in English Agreeable to be monitored by FitBit and adaptive intervention platform Able to conform to the FitBit monitoring schedule Exclusion Criteria: On insulin treatment Require assistance with basic activities of daily living (BADL) Have planned major operation or surgical procedure in the coming year at the time of recruitment Cognitively impaired (scored more than or equal to 6 on the Abbreviated Mental Test)