Abstract Body (Do not enter title and authors here): Background: Coronary heart disease (CHD), the global leading cause of death (8.9 million annual fatalities), requires combined lifestyle and medication management. We developed iCARE - a personalized mHealth system delivering visualized recommendations to improve health behaviors and medication adherence.
Purpose: To investigate health behavior trajectories and evaluate effectiveness of iCARE in CHD patients.
Methods: This multicenter randomized controlled trial enrolled CHD patients (≥18 years, angiography-confirmed, Android smartphone owners) from two Chinese tertiary hospitals, randomizing them (1:1:1) to: 1) iCARE group receiving personalized mHealth interventions via visualized content (comics/videos/images) plus usual care, 2) text message group receiving text-based interventions, or 3) control group with usual care and daily logging. Primary outcomes were 12-month health behavior trajectories (medication adherence, physical activity, diet, smoking cessation) assessed at discharge, 3-, 6-, and 12-month intervals using validated scales and digital biomarkers, analyzed via generalized linear mixed models (GLMM) to identify temporal patterns and group×time interactions.
Results: Between September 2019 and May 2024, we screened 3,000 patients, enrolling 1,149 participants: 378 in the iCARE intervention group, 383 in the text message group, and 388 in the control group. In this 12-month randomized controlled trial, trajectory analyses using generalized linear mixed models revealed significant Group × Time interactions for all continuous behavioral outcomes (all p<0.001). The iCARE group demonstrated greater monthly increases in healthy diet rate (β=0.41, 95%CI[0.25-0.57]) and medication adherence rate (β=0.38, 95%CI[0.22-0.54] ), while the text message group showed superior improvement in regular exercise rate (β=0.35, 95%CI[0.20-0.50]). Both interventions significantly reduced smoking rates versus control (iCARE: β=-0.43, 95%CI[-0.58,-0.28]; text: β=-0.31, 95%CI[-0.46,-0.16]). Peak improvements occurred at 6 months (e.g., maximum between-group difference in medication adherence: 12.4 percentage points, 95%CI[9.1-15.7]). Substantial between-individual variability was observed (slope variances: 0.12-0.19), particularly in smoking trajectories.
Conclusion: The iCARE system effectively improves dietary, exercise, and medication adherence in CHD patients, with the greatest behavior change typically occurring at 6 months.