Aim: To evaluate the effectiveness of an artificial intelligence (AI)-assisted 12-month Health Coach-Led Digital Lifestyle Intervention (HEALDI) versus control on diffuse myocardial fibrosis and ambulatory blood pressure in individuals with hypertensive heart disease (HHD), with secondary outcomes including multi-organ health parameters, health behaviours, social support, psychological health, and health-related quality of life. Background: The global prevalence of hypertensive heart disease (HHD) has increased approximately 1.5-fold, from 7.82 million cases in 1990 to 12.50 million in 2021, and it is now the second leading cause of heart failure worldwide. In HHD, chronic pressure overload drives fibroblast activation and interstitial collagen deposition, leading to diffuse myocardial fibrosis which is associated with cardiac dysfunction, arrhythmias, impaired coronary flow reserve, and an increased risk of sudden cardiac death and heart failure. Although diffuse myocardial fibrosis is potentially reversible, no approved anti-fibrotic pharmacological therapy currently exists. Furthermore, there is limited evidence evaluating the effectiveness of lifestyle interventions, particularly aerobic exercise, in reversing diffuse myocardial fibrosis. Design: A parallel, single-blinded two-arm randomised controlled trial. Method: This study is a randomised controlled trial with repeated measures, recruiting 200 physically inactive individuals with HHD from the community, including participants from Project RESET, a community-based cohort study in Singapore. Participants will be randomly allocated to either the intervention or control group. Participants in the intervention group will receive the 12-month HEALDI intervention, which includes the HEALDI mobile application, wearable device, and remote health coaching. Participants in the control group will receive a wearable device and a basic mobile application without intervention features, used solely for data collection. Data will be collected at baseline (upon randomisation) and at 6, 12 and 24 months. A process evaluation will be conducted using intervention engagement data. In addition, semi-structured interviews with participants and health coaches will explore perceptions of the intervention and behaviour change. A within-trial economic evaluation, from both healthcare system and societal perspectives, will be performed to assess cost-effectiveness. Significance: This study will generate insights into the role of lifestyle modification as a complementary, non-pharmacological strategy alongside pharmacotherapy to halt or slow HHD progression, improving long-term cardiovascular outcomes.
Specific Study Aims 1. To compare the effectiveness of HEALDI on change in diffuse myocardial fibrosis, ambulatory blood pressure, with secondary outcomes including multi-organ clinical parameters (e.g. blood tests, liver and brain imaging) and patient-reported outcomes (e.g. physical activity, dietary practices, perceived stress, sleep quality, medication adherence, social support, psychological health, and health-related quality of life). 2. Examine the cost-effectiveness of HEALDI intervention. 3. Explore the perceptions of the individuals with HHD and health coaches regarding HEALDI. Hypotheses It is hypothesised that participants who received the HEALDI intervention, compared to the control group will have: * Greater regression in diffuse myocardial fibrosis; * Greater improvement in ambulatory blood pressure; * Improved multi-organ clinical paramaters; * Greater adoption and sustained adherence to health behaviours; * Better quality of life; * Lower cost to participants and hospitals. HEALDI Intervention The HEALDI intervention is a 12-month, AI-assisted, health coach-led digital lifestyle intervention comprising a wearable device, HEALDI mobile application, health coach web portal, and administrative web portal. The system functions as an integrated platform to facilitate interaction between participants and health coaches, supporting personalised lifestyle modification. Participants will access the HEALDI mobile application via their personal smartphones and will be provided with a wearable device that continuously collects activity and physiological data. These data, together with participant-entered health metrics (e.g., blood pressure, medication), will be synchronised to a secure cloud server and made available on the health coach portal to inform coaching sessions, including goal setting and progress review. Certified health coaches will deliver 30-45-minute remote coaching sessions over 12 months: weekly during the first month, biweekly for the subsequent 5 months, and monthly for the remaining 6 months. Data Analysis Data will be analysed using IBM SPSS Statistics Version 29.0, with two-tailed p\<0.05 indicating statistical significance. Analyses will follow the intention-to-treat principle. Baseline categorical variables will be compared using Chi-squared or Fisher's exact tests, and continuous variables using independent t-tests or Mann-Whitney U tests, depending on normality. Linear mixed models will assess repeated continuous outcomes, and generalized estimating equations (GEE) will be used for ordinal outcomes. Descriptive analyses will summarise utilisation of the mobile app, wearable adherence, and coaching engagement. Qualitative interviews will be audio-recorded, transcribed, and analysed thematically using Braun and Clarke's six-step framework. Two independent coders will familiarise themselves with the data, generate initial codes and iteratively refine them through discussion.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
PREVENTION
Masking
DOUBLE
Enrollment
200
Proposed Mobile application features include: educational content on HHD and lifestyle behaviours (e.g., physical activity, nutrition, stress, sleep), AI-driven exercise recommendations (frequency, intensity, duration, and type), reviewed and approved by health coaches prior to delivery, Health logging (e.g., blood pressure, medication), Personalised dashboard displaying wearable and self-reported data, Goal setting and progress tracking (short- and long-term), Behavioural nudges (AI-driven and rule-based), Messaging function to facilitate communication with health coaches. Health coaching session involve personalised goal setting and review.
Regression of diffuse myocardial fibrosis
Change in indexed interstitial volume assessed by cardiovascular magnetic resonance
Time frame: Baseline and 12 months upon randomisation
Blood pressure
Change in the average 24-hour systolic and diastolic blood pressure
Time frame: Baseline and 12 months upon randomisation
Left ventricular function
Assessed using cardiovascular magnetic resonance
Time frame: Baseline and 12 months upon randomisation.
Left ventricular mass
Assessed using cardiovascular magnetic resonance for changes from baseline, indexed to body surface area (g/m2).
Time frame: Baseline and 12 months upon randomisation.
Left ventricular volume
Assessed using cardiovascular magnetic resonance.
Time frame: Baseline and 12 months upon randomisation.
Diastolic and systolic function
Assessed by echocardiogram
Time frame: Baseline and 12 months (upon randomisation)
Valvular function
Assessed by echocardiogram
Time frame: Baseline and 12 months (upon randomisation).
Heart function
Assessed by electrocardiogram
Time frame: Baseline and 12 months upon randomisation
First occurrence of cardiovascular event
Heart failure, Myocardial Infarction, Strokes and Death
Time frame: 2 years
Markers of liver fat, inflammation and stiffness
Assessed by liver fibroscan
Time frame: Baseline, 12 months and 24 months upon randomisation
Markers of liver fat, inflammation and stiffness
Assessed by liver magnetic resonance
Time frame: Baseline and 12 months upon randomisation.
Brain structure
Assessed using brain magnetic resonance to characterise brain microstructure and extracellular space.
Time frame: Baseline, 12 months and 24 months upon randomisation.
Cognitive function
Cognitive function will be assessed by the Montreal Cognitive Assessment (MoCA), on a 0-30 scoring scale. Thresholds are: Severe cognitive impairment: 0-9 Moderate cognitive impairment: 10-17 Mild cognitive impairment: 18-25 Normal cognitive performance: 26-30
Time frame: Baseline and 12 months upon randomisation.
Cardiac and inflammatory markers
Cardiac and inflammatory markers, including but not limited to C-reactive Protein and Cardiac Troponins, will be assessed by blood specimen.
Time frame: Baseline, 12 months and 24 months
Circulating biomarkers of inflammation and fibrosis
Assessed using blood specimen, inclusive of but not limited to, PICP (procollagen type I carboxy-terminal propeptide) and PIINP ( procollagen type III amino-terminal propeptide).
Time frame: Baseline and 12 months upon randomisation.
Liver function tests
Assessed by blood specimen
Time frame: Baseline, 12 months and 24 months from randomisation
Lipid profile
Assessed by blood specimen
Time frame: Baseline, 12 months and 24 months upon randomisation
Renal function markers
Renal function markers, including but not limited to creatinine, will be assessed by blood specimen
Time frame: Baseline, 12 months and 24 months from randomisation
Glycemic control
Assessed by fasting blood specimen
Time frame: Baseline, 12 months and 24 months upon randomisation.
Body Mass Index
Change in Body Mass Index (BMI) will be assessed using weight (kilograms) and height (meters) combined to report BMI in kg/m\^2.
Time frame: Baseline, 12 months and 24 months upon randomisation.
Body composition: Body fat
Body fat (% or kg) will be assessed by Bioelectrical Impedance Analysis (BIA) and dual-energy X-ray absorptiometry (DEXA)
Time frame: Baseline and 12 months upon randomisation.
Body composition: Muscle Mass
Muscle mass (kg) will be assessed by Bioelectrical Impedance Analysis (BIA) and dual-energy X-ray absorptiometry (DEXA).
Time frame: Baseline and 12 months upon randomisation.
Body composition: Total body water
Total body water (% or Litre) will be assessed by Bioelectrical Impedance Analysis (BIA) and dual-energy X-ray absorptiometry (DEXA).
Time frame: Baseline and 12 months upon randomisation.
Waist Circumference
Change in waist circumference
Time frame: Baseline, 12 months and 24 months upon randomisation.
Energy expenditure
Assessed by whole body calorimetry
Time frame: Baseline and 12 months upon randomisation.
Depressive symptoms
Change in depressive symptoms will be assessed by the Patient Health Questionnaire-9 (PHQ-9), on a scale of 0-27. The following PHQ-9 ranges will be used: Minimal: 0 - 4 Mild: 5 - 9 Moderate: 10 - 14 Moderately severe: 15 - 19 Severe: 20 - 27
Time frame: Baseline, 6 months, 12 months and 24 months upon enrolment.
Anxiety symptoms
Change in anxiety symptoms will be assessed by the Generalized Anxiety Disorder-7 (GAD-7), on a scale of 0-21. The following ranges will be used: None/minimal: 0 - 4 Mild: 5 - 9 Moderate: 10 - 14 Severe: 15 - 21
Time frame: Baseline, 6 months, 12 months and 24 months upon enrolment.
Health Behaviour: Physical Activity (Wearable)
Change in physical activity-related wearable data such as accelerometery.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Physical Activity (IPAQ-SF)
Change in physical activity behaviour will be assessed by the International Physical Activity Questionnaire-Short Form (IPAQ-SF) to serve as a recall estimate of time spent performing physical activity (moderate or vigorous) and inactivity (sitting). The 7-items include: 1. Vigorous activity (days) 2. Vigorous activity (minutes per day) 3. Moderate activity (days) 4. Moderate activity (minutes per day) 5. Walking (days) 6. Walking (minutes per day) 7. Sitting (days)
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Physical Activity (MPAM-R)
Change in physical activity behaviour will be assessed by the Motives for Physical Activities Measure-Revised (MPAM-R), consisting of 30-items, each on a 7-point Likert scale (1 = Not at all true for me; 7 = Very true for me), for a total score ranging between 30 - 210. A higher score will indicate higher motivation, while a lower score will indicate lower motivation.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Physical Activity (Stage of Change for Physical Activity Questionnaire)
Change in physical activity behaviour will be assessed by the Stage of Change for Physical Activity Questionnaire. The scale consists of 4-items designed to categorise individuals into the five stages of change: pre-contemplation, contemplation, preparation, action, maintenance, based on the Stages of Motivational Readiness for Change Model.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Sleep Quality and Duration
Change in Sleep Quality and Duration will be assessed using the Pittsburgh Sleep Quality Index (PSQI) and wearable-derived metrics. The scale consists of 19-items across seven components: (1) sleep duration, (2) sleep disturbance, (3) sleep latency, (4) daytime dysfunction, (5) sleep efficiency, (6) sleep quality, and (7) sleep medication usage. Each component will yield a score from 0 - 3, and will be tallied to yield a score from 0 - 21, with higher scores indicating worse sleep quality. Good sleep quality: 0 - 5 Poor sleep quality: 6 - 21
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Sleep (Wearable)
Change in sleep-related wearable data
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation
Health Behaviour: Perceived Stress
Change in perceived stress will be assessed using the Perceived Stress Scale-10 (PSS-10). The scale consists of 10 items rated on a 5-point Likert scale (0 - 4, 0 = Never; 4 = Very often), totalling a score of 40. The following score range will be used: Low stress: 0 - 13 Moderate stress: 14 - 26 High stress: 27 - 40
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Social Support
Change in social support scores will be assessed using the Multidimensional Scale of Perceived Social Support (MSPSS). It consists of 12-items on a 7-point Likert scale ranging from 1 (very strongly disagree) to 5 (very strongly agree), for a maximum possible score of 84. A higher score will indicate higher perceived social support.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Medication Adherence
Change in medication adherence will be assessed using the Medication Adherence Report Scale (MARS). The scale consists of 5-items on a 5-point Likert scale ranging from 1 (never) to 5 (always), with total scores ranging between 5 - 25. Higher scores indicate higher medication adherence.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health Behaviour: Dietary Behaviour
Change in dietary behaviour will be assessed by the Health Promotion Board Singapore Dietary Questionnaire published in the 2010 National Nutrition Survey. The scale consists of 8 items assessing the consumption frequency of fruits and vegetables, wholemeal products, sweetened drinks and food bought from eateries.
Time frame: Baseline, 6 months, 12 months and 24 months upon randomisation.
Health-related quality of life
Change in health-related quality of life will be assessed using the EQ-5D-5L. The EQ-5D-5L questionnaire will be used to evaluate the health-related quality of life of an individual to derive a 5-digit health profile code (e.g. 11211) and utility score (0 = worse than death; 1 = full health).
Time frame: Baseline, 6 months, 12 months, 24 months from enrolment.
Work Productivity and Activity Impairment Questionnaire: General Health V2.0 (WPAI)
Work Productivity and Activity Impairment Questionnaire: General Health V2.0 will be used to assess absenteeism, presenteeism, and work productivity loss. Activity impairment will be measured on a 0-10 scale, with 0 representing no problems and 10 representing severe problems. Total work impairment will be multiplied by the hourly wage to derive the valuation. This will make up part of indirect costs, which will be used for cost effectiveness analysis.
Time frame: 6 months, 12 months and 24 months upon randomisation.
Cost-Effectiveness Analysis (CEA)
Cost effectiveness Analysis (CEA) will be assessed via the incremental cost effectiveness ratio (ICER). Below details the steps required for conducting CEA: 1. Direct costs will adopt a healthcare system perspective, while indirect costs will be from a societal perspective. 2. Quality-adjusted life years (QALY) will be derived from a cost-survey analysis (direct costs: healthcare resources used will be valued using Singapore-specific unit costs; indirect costs: WPAI questionnaire and informal healthcare costs) and EQ-5D-5L (Singapore-specific value set). 3. ICER will be derived from the difference in mean costs divided by the difference in QALY between the intervention and the control group.
Time frame: 6 months, 12 months and 24 months upon randomisation.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.