Design and rationale of the artificial intelligent dialogue System assisted comprehensive Management of secondary prevention Among post coronary aRtery bypass graft patienTs (SMART): protocol for a randomised controlled trial for postcoronary artery bypass grafting management.

Abstract

Cardiovascular risk factor management (ie, hypertension, dyslipidaemia and diabetes) in post-coronary artery bypass grafting (CABG) patients is suboptimal, with a high prevalence and low control rate due to various barriers, including a lack of self-management awareness and inadequate healthcare resources. Artificial intelligence (AI) interventions are promising for improving lifestyle management and secondary prevention; however, their effectiveness in post-CABG patients remains unclear. We aimed to describe the protocol of the artificial intelligence dialogue system-assisted comprehensive management of secondary prevention among post-coronary artery bypass graft patients (SMART) assessing the efficacy and safety of an AI-based dialogue system, namely 'Smart Family Doctor', on blood pressure, lipids and glucose control. The SMART study is an open-label, parallel-group, randomised controlled trial. 536 post-CABG patients with uncontrolled hypertension, diabetes or dyslipidaemia are 1:1 randomly assigned to either the health manager or the intervention group with the implementation of the 'Smart Family Doctor' and followed up for 6 months postoperatively. The 'Smart Family Doctor' runs on the WeChat service platform, which includes an AI-based dialogue system of individual intelligent questions and answers, physiological indicators monitoring, medication reminders and automated message sending. Data on physical examinations, biological samples and self-management behaviour will be collected at baseline, 3 and 6 months after randomisation. The primary outcome was the change from baseline to 6-month follow-up in cardiovascular risk factors (hypertension, dyslipidaemia and diabetes) control rates. For the primary analysis, we will use generalised mixed-effects models to compare the control rate at the 6-month follow-up between the intervention and control groups, accounting for study sites as random effects. The study was registered at http://www. gov (NCT06615609). The central ethics committee of Fuwai Hospital approved the SMART trial (2024-2410). All collaborating sites obtained local approval or accepted central ethics approval. The findings of this study will be disseminated through peer-reviewed publications, presentations at national and international conferences, and potentially public engagement initiatives. NCT06615609.