A systematic review of mobile health opportunities for atrial fibrillation detection and management

According to current international guidelines, atrial fibrillation (AF) management should be organized in an integrated care model.1 One important component of an integrated care model is usage of technology such as mobile health (mHealth), which can engage patients in their treatment and support healthcare professionals (HCPs) to provide comprehensive and personalized diagnostic and therapeutic processes. Therefore, several handheld and wearable devices, mobile platforms, and support systems have been developed to support detection and integrated AF management. However, many of the available mHealth solutions are not clinically validated. Hence, caution is needed in their clinical use. To date, few systematic reviews have performed analyses of diagnostic accuracy of mHealth devices in screening and detecting AF,2 as well as platforms and applications to empower patients in AF management.3 However, no systematic review has comprehensively evaluated the impact of the variety of mHealth tools developed for patients with AF and HCPs who manage this condition. This study aimed to conduct a systematic review of the available published literature on mHealth solutions including handheld and wearable devices as well as mobile platforms and support systems in AF detection and management. This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses of Diagnostic Test Accuracy (PRISMA-DTA) guidelines.4 The electronic databases PubMed (NCBI), Embase (Ovid), and Cochrane were systematically searched by three independent researchers for articles published until 10 February 2021, inclusive, that evaluated the effects of mHealth solutions designed to screen and monitor AF, enhance patient’s and/or HCP’s education of AF, improve communication between AF patients and HCPs, or to encourage active AF patient involvement in the management of their condition. The main search strategy for each database is available in Supplementary material online, Table S1. Only case-control, cohort, and cross-sectional studies that evaluated the effects of mHealth solutions designed to screen and monitor AF, enhance patient’s and/or HCP’s education of AF, improve communication between AF patients and HCPs, or to encourage active AF patient involvement in the management of their condition, were included. Randomized and non-randomized controlled trials were only considered if demographic and outcome data were available. We excluded duplicates, published conference abstracts, case reports, studies without original data (e.g. reviews, commentaries, editorials), non-English written articles and studies that only included routine methods of cardiac monitoring (pacemaker, cardiac resynchronization therapy, implantable cardioverter-defibrillators). Given that the included studies varied widely in their design, interventions, comparators, and outcomes, no synthesis was undertaken, and we undertook a narrative review. After a systematic literature search, a total of 1483 records were identified. Of these, 159 were deemed potentially relevant, of which 82 and 46 studies aimed at validating handheld devices and wearables for AF detection and/or management, respectively, while 34 studies assessed mobile platforms/support systems (Figure 1). Studies were classified according to the type of devices (wearable and handheld) and technology used to assess AF [photoplethysmography (PPG), electrocardiography,5 and mechanocardiography (MCG)] (Figure 1). The diagnostic accuracy range of PPG-based handheld devices was 91.5–98.5% (sensitivity) and 91.4–100% (specificity). The sensitivity and specificity range of electrocardiogram (ECG)-based handheld devices was 94–98% and 76–95%, and of MCG-based handheld devices was equal to 67% and 99%, respectively. Overall sensitivity and specificity for all validated wearable devices were high and ranged from 67.7–100% and 67.6–100%, respectively (Table 1).6–9 Minority of AF mobile platforms and support systems are FDA/CE-approved and/or evaluated in clinical studies (Supplementary material online, Table S2). Flow diagram for study selection process and type of mobile health solutions for atrial fibrillation detection and management with their sensitivity and specificity considering 12-lead electrocardiogram as the gold standard. This figure summarizes the literature as performed in this systematic review. Sensitivity and specificity range were given when an mHealth solution was clinically validated by >1 study. aSingle-lead electrocardiogram as the gold standard; b24-h Holter monitoring as the gold standard. cNumbers do not add up to 159 as 3 studies assessed in parallel handheld device with mobile platforms/support systems (n = 1) and wearable devices (n = 2). dNumbers do not add up to 82 as 4 studies assessed in parallel PPG- and ECG-based device. Available FDA/CE-approved mobile applications AF, atrial fibrillation; CDSS, clinical decision system support; ECG, electrocardiography; MCG, mechanocardiography; mHealth, mobile health; PPG, photoplethysmography. Available FDA/CE-approved mobile applications AF, atrial fibrillation; CDSS, clinical decision system support; ECG, electrocardiography; MCG, mechanocardiography; mHealth, mobile health; PPG, photoplethysmography. This systematic review suggests that the diagnostic accuracy of mHealth devices differs with respect to the type (handheld vs. wearable) and technology used (ECG vs. PPG vs. MCG). In addition, mHealth solutions with the same technologies may collect data in different ways (intermittent vs. continuous, spot vs. longitudinal assessment) as well as different methods of measurement (handheld vs. wearable), and different study populations which could influence the sensitivity and specificity for AF detection.10 Therefore, there is a need to perform head-to-head comparisons between the different methods of measurement, as their comparative effectiveness is limited, hence unclear. In conclusion, although mHealth solutions are becoming increasingly popular in the detection and management of AF, in particular in high-risk patients11 to institute appropriate treatment, its clinical implications merit further investigation and several barriers to widespread mHealth adaption in healthcare systems need to be overcome. Initiatives such as the TeleCheck-AF project, which was introduced to keep AF patients’ care during the coronavirus-2019 pandemic via teleconsultations coupled with heart rate and rhythm monitoring,12 will help step-by-step to implement tailored mHealth solutions in clinical care pathways. In addition, artificial intelligence-based AF risk prediction algorithms can improve the utility of mHealth solutions in AF detection.13,14 Supplementary material is available at European Journal of Preventive Cardiology online. The data underlying this article will be shared on reasonable request to the corresponding author. Conflict of interest: the Authors declare that there is no conflict of interest.