A Mobile Relational Agent to Enhance Atrial Fibrillation Self-care

点击翻译按钮获取中文摘要

PROJECT SUMMARY Atrial fibrillation (AF) is a common, morbid condition. Symptoms, complex management, and significant adversity contribute to poor health-related quality of life (HRQoL). Social determinants of health exacerbate morbidity in AF, and limited health literacy compounds the poor patient experience of AF. We propose a single- center parallel group randomized clinical trial to test the efficacy of an embodied conversational agent (ECA) to improve patient-centered care in AF. The ECA is a computer character that simulates face-to-face conversation using voice, hand gesture, and gaze cues to provide education, monitoring and problem-solving. We have used the ECA in multiple health contexts for self-care and demonstrated its success to improve health behaviors and outcomes in individuals with limited computer and health literacy. Here we propose to expand our successful 30-day pilot (n=31) of the ECA and the AliveCor Kardia smartphone heart rate and rhythm monitor in order to evaluate the effect of a 4-month self-care curriculum and assess its 12-month sustainability. We will further test the integration of ECA/Kardia results in the electronic health record (EHR) and monitor alerts and resulting modifications to care. We will randomize 240 patients with AF who are receiving anticoagulation to either (1) the ECA/Kardia intervention, provided by smartphone, and accompanying EHR alerts, or (2) the control, consisting of an AF educational session, diary for recording symptoms and adherence, and a smartphone with a general health application for self-care (WebMD). Our trial will leverage the clinical infrastructure of the University of Pittsburgh Medical Center (UPMC) by recruiting at 8 UPMC clinics that share a common electronic health record. We will focus recruitment on individuals with limited socioeconomic resources, low health literacy, or racial/ethnic minorities. Our aims are: (1) To examine the effect of the ECA/Kardia intervention on patient-centered outcomes. We will evaluate HRQoL with the AF- specific AF Effect on QualiTy of life (AFEQT) measure and general HRQoL with the Patient-Reported Outcomes Measurement Information System-29 Profile at baseline, 4, and 12 months. (2) To evaluate the effect of the ECA/Kardia and EHR alert intervention on anticoagulant adherence. We will quantify adherence using medication possession ratio and pharmacy contact at 12 months, and complementary measures of self- reported non-adherence at baseline, 4, and 12 months. (3) To determine the effect of the intervention on health care utilization at 4 and 12 months using participant interview and the common EHR. Our trial will engage an 8- member patient advisory committee comprised of individuals with chronic AF to guide the intervention's cultural acceptability, recruitment, and presentation of results. Expected Results: In this project we will evaluate a scalable patient-centered intervention to improve HRQoL, improve anticoagulation adherence, and reduce health care utilization in vulnerable individuals with chronic AF. If proven successful, this intervention can be broadly disseminated to improve the care of patients with AF.