Self-monitoring using Mobile App to improve symptoms and Reduce rehospitalization and morTality in Heart Failure (SMART-HF trial)i

Abstract Background Considering the uncertainty of the benefits of mobile apps in patients with heart failure (HF), we developed a mobile health platform and aimed to evaluate this intervention in terms of improving symptoms and outcomes in HF. Methods Seven tertiary hospitals in South Korea participated in this prospective, randomized, open-label, nationwide, multicenter, pilot trial (SMART-HF trial). Patients aged ≥20 years hospitalized for acute heart failure who were able to use a smartphone were enrolled before discharge. Eligible participants were then randomized 1:1 to either the intervention group (AIDA/KT-app with devices) or the control group (AIDA/KT-app only). The mobile health platform consists of three parts: 1) smartphone app for patients (2) dashboard system for physicians (3) clinical decision support system. Only in the intervention group, the applications were automatically paired with a Bluetooth blood pressure monitor, weight scale and body water analyzer. The intervention group were able to enter their HF symptom scale into the app and receive feedback or alerts on their input. In the control group, patients could only enter their blood pressure, weight, and HF symptom into the app, and there are no connected devices and no feedback from the app. Results 84 patients were randomized to the intervention group (n=43) and control group (n=41). Overall, 62.8% of the patients were male, and the median age was 63.0 years. At 4 weeks, dyspnea score by questionnaire (range 0-10) was 1.5±2.3, and 2.0±2.4 in the intervention and control groups, respectively, and was not significantly different (P=0.415). The total HF symptom score (range 0-12) including dyspnea, fatigue, edema, and palpitation entered by the participants in the apps was not significantly different between two groups (2.0±2.1, and 1.6±2.1, P=0.454). In the intervention group, the level of body water composition showed trend toward decreasing during study period (initial level 7.46 -> final level 6.71, P=0.10). App adherence during the study period was higher in the intervention group compared with control group (70.6±32.5% vs. 47.8±32.0%, P=0.004). App satisfaction scores tended to be higher in the intervention group (7.1±2.6 vs. 6.2±2.9, P=0.147). Secondary composite endpoints including death, rehospitalization, and urgent HF visit were not significantly different between two groups (4/40 vs. 3/38, P=0.999). Conclusions Our mobile health platform showed feasibility and efficacy for HF patients although clinical outcomes were not significantly different, maybe due to small sample size. Considering that motivating patients to use the mHealth tool more frequently may lead to greater clinical benefit, further large research with a positive user experience is needed for HF patient care.