Search

BackgroundCardiovascular disease (CVD) is the leading cause of death worldwide. Despite strong evidence supporting the benefits of cardiac rehabilitation (CR), over 80% of eligible patients do not participate in CR. Digital health technologies (ie, the delivery of care using the internet, wearable devices, and mobile apps) have the potential to address the challenges associated with traditional facility-based CR programs, but little is known about the comprehensiveness of these interventions to serve as digital approaches to CR. Overall, there is a lack of a systematic evaluation of the current literature on digital interventions for CR. ObjectiveThe objective of this systematic literature review is to provide an in-depth analysis of the potential of digital health technologies to address the challenges associated with traditional CR. Through this review, we aim to summarize the current literature on digital interventions for CR, identify the key components of CR that have been successfully addressed through digital interventions, and describe the gaps in research that need to be addressed for sustainable and scalable digital CR interventions. MethodsOur strategy for identifying the primary literature pertaining to CR with digital solutions (defined as technology employed to deliver remote care beyond the use of the telephone) included a consultation with an expert in the field of digital CR and searches of the PubMed (MEDLINE), Embase, CINAHL, and Cochrane databases for original studies published from January 1990 to October 2018. ResultsOur search returned 31 eligible studies, of which 22 were randomized controlled trials. The reviewed CR interventions primarily targeted physical activity counseling (31/31, 100%), baseline assessment (30/31, 97%), and exercise training (27/31, 87%). The most commonly used modalities were smartphones or mobile devices (20/31, 65%), web-based portals (18/31, 58%), and email-SMS (11/31, 35%). Approximately one-third of the studies addressed the CR core components of nutrition counseling, psychological management, and weight management. In contrast, less than a third of the studies addressed other CR core components, including the management of lipids, diabetes, smoking cessation, and blood pressure. ConclusionsDigital technologies have the potential to increase access and participation in CR by mitigating the challenges associated with traditional, facility-based CR. However, previously evaluated interventions primarily focused on physical activity counseling and exercise training. Thus, further research is required with more comprehensive CR interventions and long-term follow-up to understand the clinical impact of digital interventions.

BackgroundDespite the promise of machine learning (ML) to inform individualized medical care, the clinical utility of ML in medicine has been limited by the minimal interpretability and black box nature of these algorithms. ObjectiveThe study aimed to demonstrate a general and simple framework for generating clinically relevant and interpretable visualizations of black box predictions to aid in the clinical translation of ML. MethodsTo obtain improved transparency of ML, simplified models and visual displays can be generated using common methods from clinical practice such as decision trees and effect plots. We illustrated the approach based on postprocessing of ML predictions, in this case random forest predictions, and applied the method to data from the Left Ventricular (LV) Structural Predictors of Sudden Cardiac Death (SCD) Registry for individualized risk prediction of SCD, a leading cause of death. ResultsWith the LV Structural Predictors of SCD Registry data, SCD risk predictions are obtained from a random forest algorithm that identifies the most important predictors, nonlinearities, and interactions among a large number of variables while naturally accounting for missing data. The black box predictions are postprocessed using classification and regression trees into a clinically relevant and interpretable visualization. The method also quantifies the relative importance of an individual or a combination of predictors. Several risk factors (heart failure hospitalization, cardiac magnetic resonance imaging indices, and serum concentration of systemic inflammation) can be clearly visualized as branch points of a decision tree to discriminate between low-, intermediate-, and high-risk patients. ConclusionsThrough a clinically important example, we illustrate a general and simple approach to increase the clinical translation of ML through clinician-tailored visual displays of results from black box algorithms. We illustrate this general model-agnostic framework by applying it to SCD risk prediction. Although we illustrate the methods using SCD prediction with random forest, the methods presented are applicable more broadly to improving the clinical translation of ML, regardless of the specific ML algorithm or clinical application. As any trained predictive model can be summarized in this manner to a prespecified level of precision, we encourage the use of simplified visual displays as an adjunct to the complex predictive model. Overall, this framework can allow clinicians to peek inside the black box and develop a deeper understanding of the most important features from a model to gain trust in the predictions and confidence in applying them to clinical care.

BackgroundAs smartphone ownership continues to rise, health care systems and technology companies are driven to develop mobile health (mHealth) interventions as both diagnostic and therapeutic tools. An important consideration during mHealth intervention development is how to achieve health equity despite demographic differences in smartphone ownership. One solution is through the recirculation of loaner smartphones; however, best practices for implementing such programs to optimize security, privacy, scalability, and convenience for participants are not well defined. ObjectiveIn this tutorial, we describe how we implemented our novel Corrie iShare program, a 30-day loaner iPhone and smartwatch recirculation program, as part of a multi-center mHealth intervention to improve recovery and access to guideline-directed therapy following acute myocardial infarction. MethodsWe conducted a prospective study utilizing a smartphone app and leveraged iOS enterprise features as well as cellular data service to automate recirculation. ResultsOur configuration protocol was shortened from 1 hour to 10 minutes. Of 200 participants, 92 (46.0%) did not own an iPhone and would have been excluded from the study without iShare. Among iShare participants, 72% (66/92) returned their loaned smartphones. ConclusionsThe Corrie iShare program demonstrates the potential for a sustainable and scalable mHealth loaner program, enabling broader population reach while optimizing user experience. Implementation may face institutional constraints and software limitations. Consideration should be given to optimizing loaner returns.

The explosion of mobile health (mHealth) interventions has prompted significant investment and exploration that has extended past industry into academia. Although research in this space is emerging, it focuses on the clinical and population level impact across different populations. To realize the full potential of mHealth, an intimate understanding of how mHealth is being used by patients and potential differences in usage between various demographic groups must also be prioritized. In this viewpoint, we use our experiences in building an mHealth intervention that incorporates an iOS app, Bluetooth-enabled blood pressure cuff, and Apple Watch to share knowledge on (1) how user interaction data can be tracked in the context of health care privacy laws, (2) what is required for effective, nuanced communication between clinicians and engineers to design mHealth interventions that are patient-centered and have high clinical impact, and (3) how to handle and set up a process to handle user interaction data efficiently.

The current revolution of digital health technology and machine learning offers enormous potential to improve patient care. Nevertheless, it is essential to recognize that dermatology requires an approach different from those of other specialties. For many dermatological conditions, there is a lack of standardized methodology for quantitatively tracking disease progression and treatment response (clinimetrics). Furthermore, dermatological diseases impact patients in complex ways, some of which can be measured only through patient reports (psychometrics). New tools using digital health technology (e.g., smartphone applications, wearable devices) can aid in capturing both clinimetric and psychometric variables over time. With these data, machine learning can inform efforts to improve health care by, for example, the identification of high-risk patient groups, optimization of treatment strategies, and prediction of disease outcomes. We use the term personalized, data-driven dermatology to refer to the use of comprehensive data to inform individual patient care and improve patient outcomes. In this paper, we provide a framework that includes data from multiple sources, leverages digital health technology, and uses machine learning. Although this framework is applicable broadly to dermatological conditions, we use the example of a serious inflammatory skin condition, chronic cutaneous graft-versus-host disease, to illustrate personalized, data-driven dermatology.