39 © Springer Nature Switzerland AG 2021 A. B. Bhatt (ed.), Healthcare Information Technology for Cardiovascular Medicine, Health Informatics, https://doi.org/10.1007/978-3-030-81030-6_3 Chapter 3 Remote Patient Monitoring: Delegation of Responsibility Elizabeth A. Krupinski and Jaclyn A. Pagliaro 3.1 Introduction Healthcare is becoming increasingly more technical, not only with respect to the wide array of new devices and tests available to providers and patients, but also the overall healthcare delivery paradigm. The integration of wearable remote monitoring devices and telemedicine is altering the way clinical care is provided and research is conducted. Wearable remote monitoring devices can be deployed to patients and utilized to extend patient data collection outside of conventional clinical encounters [1]. The anticipated impact of remote data collection on patients and health care systems is to increase provider efficacy and efficiency, thereby increasing quality of care [2–4]. This improved quality of life is the result of more reliable and timely methods for collecting, transmitting, and storing personal health data for health care communications. Within this context is the evolution of guidelines for responding to remotely collected and stored data, and reimbursement policies [5, 6] that accommodate remote data management in any reimbursement structures. Recent research supports the use of RPM for adults with cardiovascular disease demonstrating that consistent blood pressure tracking and communication with healthcare teams improves BP related outcomes. RPM lowers both systolic and diastolic blood pressure more than standard in-office BP management. RPM may also promote early detection of atrial fibrillation and improve heart failure volume and weight management. E. A. Krupinski (*) Department of Radiology & Imaging Sciences, Emory University, Atlanta, GA, USA e-mail: ekrupin@emory.edu J. A. Pagliaro Division of Cardiology, Department of Medicine, Massachusetts General Hospital, Boston, MA, USA e-mail: jaclyn.pagliaro@mgh.harvard.edu
40 Wearable devices now add to the RPM field and are being designed to collect data about various physiological, psychological and environmental parameters at the individual and population level which complement standard digital medical data acquisition. As these devices proliferate, data is constantly accessible by a variety of parties (e.g., patients, providers, hospitals, vendors), regardless of the location of the individual from whom the data is being recorded. An increasing number of devices can acquire, transmit and store data to enhance the relationship between patients and their providers. The use of ambulatory monitoring devices in the cardiology setting enables long- term, remote data collection for reliable diagnostics and continuous monitoring of cardiac disease [7]. The device field in cardiology is one of the most advanced of all medical specialties in terms of monitoring and intervention. This provides an existing scaffold for the mechanisms and workflow for remote data collection and analysis, as exemplified in electrophysiology, heart failure, cardiac rehabilitation and most recently in cardio-obstetrics. Remote patient monitoring (RPM) is being used in a variety of applications such as remote monitoring of implantable devices [8] and remote electrocardiogram monitoring [9]. Most studies demonstrate that use of RPM is feasible and can be used to assess device and patient status in a reliable and valid manner. Success may depend on factors such as the type of device (manufacturer, model), patient status, type of information acquired, and even the algorithms used to assess the incoming data (impacting false negative and alarm rates). Thus, some caution should always be used when interpreting RPM data and when to base clinical decisions on the data received. It is particularly important to work with the device manufacturers to understand the criteria used in the various data analysis algorithms used to notify users and providers of significant changes in the data and patient status. Despite the many advancements, there is still much work that needs to be done in terms of research to improve our understanding of actual outcomes and benefits of these tools. It is also critical to determine which patient populations are best suited to benefit from the use of these devices [10]. On the positive side, patients do seem receptive to remote patient monitoring and are generally satisfied with the experience [11, 12]. A key topic that arises when one thinks about all these data is responsibility. When and in what circumstances is it the provider’s responsibility to utilize the acquired data? There is currently no strict legal statute, but there are ways to think about data sources that may be helpful. 3.2 Data and Context As a first step it is useful to distinguish between data acquired strictly within the context of healthcare encounters and data that could be used for research purposes. E. A. Krupinski and J. A. Pagliaro
41 3.2.1 Research Context Until a formally recommended structure exists regarding acquiring, transmitting and storing various personal and medical sources of data, we should always apply the Belmont Report’s guiding principles of human research (Table 3.1) to the use of any personal data [13]. 3.2.2 Clinical Context 3.2.2.1 Data Solicited by Providers Data solicited by providers refers to data that patients collect by wearing a device that acquires and transmits data to their healthcare provider to be used for medical decision-making. In this circumstance, the provider “prescribes” a wearable device to acquire the data. Data security must ensure that the data are protected from unauthorized users during transfer, collection and processing, and that they are safely stored to protect identifiable health data from unwarranted access or disclosure. Since RPM is wireless, there is the inherent risk that data can be intercepted, tracked and possibly used in harmful or illegal ways. However, devices that are FDA-approved undergo fairly rigorous evaluation [14] with respect to security, so these risks are minimized (usually via data encryption and authentication). The patient and physician must communicate effectively so the patient understands why, how and when the data are collected, transferred and stored and how they will be used in their care and treatment (Table 3.2). Interestingly, although Table 3.1 Belmont Report’s guiding principles of human research Three guiding principles for protection of human subjects in Behavioral & Biomedical Research Respect for persons The autonomy of people must be respected in terms of allowing them to make choices and providing informed consent. Beneficence Researchers must maximize benefits and minimize risks to those involved. Justice Benefits and costs associated with participation must be fairly and equally distributed among all potential participants. Table 3.2 Recommendations for provider solicited data • Ensure safety of data acquisition • Identify any limitations associated with validity and reliability of data • Recognize variations in use of application with various patient populations • Understand the nature of false alarms • Know how to incorporate the output data into clinical decision making diagnosis and treatment 3 Remote Patient Monitoring: Delegation of Responsibility
42 patients do have concerns about the technology itself and security issues they are often more concerned about loss of contact with providers and whether out-of- pocket expenses will be incurred [15]. Therefore, it is important that physicians are aware of such concerns and work with patients to address them [16, 17]. 3.2.2.2 Data Not Solicited by Providers Data that patients generate without a formal request from a provider can come from a variety of sources. Consumers use wearable monitoring devices such as Smartwatches or Fitness Trackers with the expectation that they will impact their health status and often share their data as an adjunct to their care. This data derived from remote monitoring devices has been found to provide valuable insight into the patient’s overall health to provide comprehensive treatment and care [18, 19]. When a patient brings data from these types of unsolicited data sources to a healthcare encounter (which could be in-person, through a patient portal, email, text, a video encounter, phone or any other mode), the provider has a choice to make. At the same time, it is also becoming apparent that some patients are often willing to acquire various types of health data, including RPM from both provider-approved and off- the-shelf devices (e.g., FitBits [20], and then publicly post that data on social media outlets, complicating the issue even further [19] as most social media companies actually consider any posted data theirs and thus can do with it as they wish. When a patient brings data from these types of unsolicited data sources to a healthcare encounter (which could be in-person, through a patient portal, email, text, a video encounter, phone or any other mode), the provider has a choice to make: should the data be accepted and stored in the electronic health record or other data repository or should the data be declined and thus not stored? Unfortunately, there are few clear-cut guidelines to help with this choice. Whereas the patient claims responsibility when they choose to acquire and present data to their provider, the clinician’s responsibility is less clear. When data was not requested by the clinician, if they decide to accept and store it in the electronic health record they assume a share of the responsibility for the data. This includes its use in medical decision-making and its security (e.g., HIPAA regulations automatically apply). This creates a series of additional responsibilities such as doing the necessary research to understand the device or tool from which the data were acquired, what the limitations of the data acquisition process are, the data validity and reliability, factors that can influence data quality. Without this additional step it becomes difficult sometimes to responsibly incorporate the data into care. If the provider chooses not to accept the data, at first glance it seems that they avoid responsibility. There are several mechanisms by which the provider may need to address the data, however: 1. If the patient provides the data to another provider who uses it to change the course of treatment this may reflect on the original clinician who decided not to accept the data. E. A. Krupinski and J. A. Pagliaro
43 2. If a clinician later decides to use the data, there is little published guidance as to what their responsibility is for previously stored data. 3. If there are abnormal values, there is no clear legal precedence regarding the clinician’s accountability. The next phase of telemedicine and utilization of patient derived data is to offer clinician’s guidance with these emerging decisions. 3.3 Wearable Devices and Remote Patient Monitoring in Cardiology Cardiology is remarkably data driven and most guidelines have algorithms to provide comprehensive patient care. The ability to integrate additional tracking mechanisms into existing workflows has been successful in several healthcare systems. The value of digital health and telemedicine can be best described (Table 3.3) as addressing the trifecta of preventative care, chronic disease management and early detection. 3.4 Patient Driven Tools One of the fastest growing areas of medical technologies is the commercial availability of health monitoring tools. Activity trackers, Bluetooth enabled BP and HR monitors, mobile ECG, and many other devices have flooded the marketplace. Fitting these tools into a digital heart center introduces unique complexities, as questions of data fidelity, false positive and false negative rates, and helping patients distinguish between data and information all play a much larger role in these technologies than other parts of a technology forward cardiology practice. In addition to the question of what problem does this technology solve, for these technologies, we must also ask the question: what harm comes from inaccurate information? Table 3.3 Key areas where digital health and telemedicine can address preventative care, chronic disease management and early detection Prevention – Wearable devices and telemedicine can improve healthcare access to patients who experience barriers to care Chronic disease management – Wearables allow clinicians to continuously manage large patient populations at a distance rather than fewer patients with episodic in-person care alone – Patients with chronic diseases who use wearables and telemedicine also benefit from decreased stress, saved time and finances, while also remaining active members in their local community Early detection – Digital health allows early detection of decompensation leading to rapid and targeted intervention – Algorithms can be applied to the collected data to provide remote clinical care such as medication titrations 3 Remote Patient Monitoring: Delegation of Responsibility
44 3.5 What Harm Comes from Inaccurate Information? If a patient has purchased a health monitoring device on their own, the patient has already signaled that they get some value from the device. So while it’s still important to figure out what problem the device solves for the clinical team, it is much more important to figure out how the information on the device (especially if it’s inaccurate) could bring harm to the patient. One of the best examples of such a technology is the new Apple Watch. The apple watch has two features of interest: one is a calorie/activity tracker, that measures activity in a day in “calories burned” and the other is the ECG monitor, which can offer a single lead ECG tracing. For an activity monitor, the accuracy is likely not that important. When the device says the patient burns 500 calories, if the “true” calorie consumption is 300 or 700 calories, it is unlikely to harm the patient. In fact, if it is relatively precise (consistent day to day) and motivates a patient toward more activity, the accuracy of the activity count is largely irrelevant. Incorporating such a tool in a virtual cardiac rehab program, for example, would confer minimal risk for the patient. Case study example: Thomson et al. examined the Fitbit Charge HR 2 (Fitbit) and Apple Watch devices to assess hea every vigorous intensities based on ECG-measured HR. Apple Watch showed lower relative error rates (2.4–5. 1%) compared to Fitbit (3.9–13.5%) for all intensities. The strongest relationship for both devices with ECG-measured HR was for very light PA and strength of the relationship declined as exercise intensity increased for both devices. Thomson EA. Nuss K, Comstock A, Reinwald S, Blake S, Pimentel RE, Tracy BL, Li K. Heart rate measures from the Apple Watch, Fitbit Charge HR2, and electrocardiogram across different exercise intensities. J Sports Sciences 2019; 37:1411-1419 On the other hand, the ECG tool has a different risk equation. One diagnosis the ECG monitor on the apple watch hopes to make is atrial fibrillation [21]. Here, a false positive diagnosis could result in inappropriate anticoagulant therapy with the associated bleeding risk without potential for benefit. On the other hand, a false negative result could incorrectly reassure us that the patient doesn’t have atrial fibrillation, resulting in inadequately withholding anticoagulation therapy. In one estimation, for every 1 patient correctly diagnosed with atrial fibrillation, 19 people potentially would be inappropriately placed on anticoagulation. In this case, the watch could be used as a screening tool, but care should be used in making treatment decisions based on the data. Since the patient has purchased the device, it has solved at least one problem for them (patient comfort). Still, it’s important that the clinical team determine what (if any) problems the device helps them solve. A clear understanding of what we hope to get from a new device and how we communicate that to patients (Table 3.4) helps E. A. Krupinski and J. A. Pagliaro
45 maximize the benefit for the patient, but also reduces their frustration, as the clinical team can offer up front guidance on what the device can and cannot do. 3.6 Guidance for RPMWorkflow Development and Implementation Cardiology practices should establish an infrastructure for receipt, analysis and clinical application of any form of “remote monitoring”. Practices can create an effective and safe mechanism for the use of remote monitoring and wearable device data. 1. All data (whether or not it was requested by the provider) should be subject to the same security and privacy requirements of any patient health information (PHI). This includes hard copies on site or cloud-based data. 2. Organizations should have in place policies and protocols governing the use of any and all types of remote patient monitoring data. The policy should include how and if unsolicited patient generated data will or could be used and what criteria should be used in the decision to accept or not. 3. If providers are going to recommend certain devices or tools be used by patients, they should have a formal program in place that manages the schedule and timing of data collection by the patient, transfer, storage, retrieval, analysis, and feed back to the patient. 4. Providers should not be responsible for training patients on the actual use of devices that they do not recommend. 5. When a technology or device is prescribed, patients should have a responsibility to submit requested data and the provider(s) should take appropriate action when the patient is non-compliant or is unable to participate appropriately. 6. Protocols should be in place regarding alerts generated from remote monitoring devices (especially when artificial intelligence and related decision support tools are used) [22–25]—when to respond to them, who should respond, how the patient should be contacted and what follow-up measures should be in place. Table 3.4 Ambulatory device capabilities and how we communicate that to patients Ambulatory monitoring capabilities Considerations for patient communication Vital sign data sources – Heart rate – Respiratory rate – Temperature – Oxygen saturation – Blood pressure Wearable data streams – ECG – Sleep assessment – Physical activity Heart rate variability exists, normal ranges should be reviewed with patients FDA approved blood pressure monitors are largely accurate, but checking calibration in the office is useful Oxygen saturation meters can be inaccurate with improper placement At home single lead ECG will not mirror an in-office 12-lead in its entirety for diagnosis. Single lead home ECG or HR monitor devices may have false positives for arrhythmia and follow up with formal ECG or monitoring is recommended before treatment 3 Remote Patient Monitoring: Delegation of Responsibility
46 7. Policies should also be in place for all data sources regarding how long the data should be stored. Some resources are available to providers such as the American Medical Association’s “Digital Health Implementation Playbook” that provides a very nice summary of RPM and other digital tools available and how to integrate them into patient care [26]. Cardiac specific workflow for RPM, which incorporates patient solicited data mandates a clear approach to appropriately protect patient data. First, we must define what data points are clinically relevant and how it will be identified, collected and communicated to the appropriate care team member. Next, organize remote patient monitoring data into existing practice workflows. Finally, do not create new workflows and additional clinical and administrative burden. For successful implementation (Fig. 3.1), engaging the staff is essential to a successful transformation to a digital RPM inclusive practice. Staff (clinical and administrative) training should include knowledge on how to explain the device role in management, order, deliver and engage with the device from a patient perspective, help troubleshoot and have access to the device manufacturers helpline. Practices will also need to be able to schedule device checks and data downloads, oversee data acquisition, be familiar with patient information materials and most importantly, know when to contact clinicians. A successful patient-clinician partnership is the first and most important outcome to achieve during a transition to increased RPM usage in the practice. Patients need to be equal partners in device use, understand clinical endpoints that are being measured and be facile with methods of communication with the clinical team. Although some recommend patient selection for compliance and technological proficiency, the goal of expanding care to the individual patient is to improve access, therefore we must meet patients and help overcome their barriers. Addressing social determinants of health is a foundational part of the patient engagement strategy. Once a successful partnership is achieved and demonstrated, outcomes then include a range of clinical, experiential and business outcomes (Table 3.5). The National Institute for Standards and Technology [27] is also addressing some RPM issues and has posted a request for “technology vendors to participate in the development of an example solution for securing the telehealth remote Structure For successful RPM, data points and workflow must be well-defined Process All members of the practice, administrative and clinical, will require training and ongoing technical and clinical support for successful implementation of RPM Outcomes Patient and physician satisfaction, use of the RPM system, increased access and achievement of clinical endpoints can be sequentially evaluated as an RPM program is established. Fig. 3.1 Steps for successful implementation of RPM use E. A. Krupinski and J. A. Pagliaro
47 patient monitoring ecosystem” to create a Cybersecurity Practice Guide. The European Commission [28] is also developing frameworks for mHealth guidance that includes RPM. RPM implementation requires several steps for success. Adequately supporting the technology, workflow and staffing for an RPM program is essential for sustainability. Unlike the use of synchronous telemedicine for expanding access and improving outcomes by reaching more patients, the use of RPM requires well- selected populations with effective patient-clinician communication and adequate staff resourcing to demonstrate clinical efficacy and improve outcomes. Similar to clinical trials, establishing an RPM program requires inclusion and exclusion criteria for patient enrollment, as well as the equivalent of shared decision making, wherein patients understand the goal of RPM, and the importance of their active role to enable success. The initial onboarding for all telemedicine modalities, whether synchronous video, text-based interactions, or RPM including wearable data needs to be a personal one-on-one thorough explanation at the initiation of the program, with continued opportunity for contacting the team for troubleshooting on demand. Even one poor interaction with telemedicine risks creating an aversion to its use. However the stakes are higher with RPM, in that one poor interaction may result in less than optimal care. Key Steps in Implementation of RPM (Table 3.6): • Administrative staff, leadership and clinical staff must all be involved in design review prior to implementation • Patients should understand the goal of RPM for their healthcare and their role in a successful model of care • Real-time, at the elbow help for administrative, workflow and technical assistance must be in place not only at implementation but in some form for the duration of chronic disease management using RPM. Table 3.5 Clinical, experiential and business outcomes for successful RPM implementation Clinical outcomes • Blood pressure monitoring • HF exacerbation management • Admission and readmission Reduction • GDMT achievement regardless of cardiac diagnosis • Compliance with medications • Improved clinical markers of disease • Less progression of disease Experiential outcomes • Patient, staff and provider satisfaction • Accessing healthcare appointments • Successful engagement with interpreter services • Coordinated multidisciplinary visits patient engagement with technology Business outcomes • Decreased cancellations and no shows • Increased clinical productivity • Fewer hospitalizations and ED utilization Telemedicine utilization Reimbursement 3 Remote Patient Monitoring: Delegation of Responsibility
48 3.7 Conclusions There is no doubt that RPM techniques and wearable devices will continue to be developed and implemented by providers and patients in a wide variety of clinical applications, including cardiology [29]. Although the potential for improving patient outcomes with these devices is high, the evidence on outcomes to date is still inconclusive in many respects [30, 31]. More large-scale studies (e.g., [32]) are required to better assess the impact of RPM and wearables devices in broader, more heterogeneous groups of patients with a variety of cardiovascular conditions and degrees of severity. As we proceed down this path as a healthcare community that includes not only providers but also patients and their caregivers, we need to continue to develop and refine our perspectives and guidance surrounding the ethical, sage and effective use of virtual data no matter what its source. Remote patient data ownership must be considered, but it should not represent a barrier or hindrance in its use as a tool to improve the efficacy and efficiency of clinical decision making and the care and treatment of patients. References 1. DeVore AD, Wosik J, Hernandez AF. The future of wearables in heart failure patients. JACC Heart Fail. 2019;7(11):922–32. 2. LiaoY, Thompson P, Peterson S, Mandrola J, Shaalan M. The future of wearable technologies and remote monitoring in health care. Am Soc Clin Oncol Educ Book. 2019:115–21. 3. Malasinghe L, Ramzan N, Dahal K. Remote patient monitoring: a comprehensive review. J Ambient Intell Humaniz Comput. 2019;10:57–76. 4. Queiros A, Alvarelhao J, Cerqueira M, Silva AG, Santos M, Rocha NP. Remote care technology: a systematic review of reviews and meta-analyses. Technologies. 2018;6:22. Common-sense steps to implemention an RPM wearable device program: • Establish program structure, process and outcomes to be measured • Center the program around specific patient populations • Document data acquisition and data use in the clinical decision-making process • Guarantee security and privacy • Support data entry by patients, providers and staff where applicable • Provide links to patient records to facilitate full data inegration • Incorporate clinical guidelines in establishing criteria for responding to alarms based on the data (for example, agree upon published acceptable false positive and false negative rates) • Document procedures for data alarms including who is screening, in what time frame is a response needed, what to do when a patient does not respond, how to document the response • Determine who is responsible for the data (acquiring, transmitting, storing) and engage your legal tream regarding clinical responsibility and interaction with industry • Determine how long RPM data should be collected and stored (and how much of it) Table 3.6 Key common sense steps in implementation of RPM E. A. Krupinski and J. A. Pagliaro
49 5. Berridge C. Medicaid becomes the first third-party payer to cover passive remote monitoring for home care: policy analysis. J Med Internet Res. 2018;20:e66. 6. Centers for Medicare & Medicaid Services (CMS). 2019 Physician Fee Schedule and Quality Payment Program. 2018. https://www.cms.gov/newsroom/fact-sheets/finalized- policy-payment-and-quality-provisions-changes-medicare-physician-fee-schedule-calendar. Accessed 13 Dec 2019. 7. Sana F, Isselbacher EM, Singh JP, Heist EK, Pathik B, Armoundas AA. Wearable devices for ambulatory cardiac monitoring: JACC state-of-the-art review. J Am Coll Cardiol. 2020;75(13):1582–92. 8. Cheung CC, Deyell MW. Remote monitoring of cardiac implantable electronic devices. Can J Cardiol. 2018;34:941–4. 9. Li J, Yang P, Fu D, Ye X, Zhang L, Chen G, YangY, Luo H, Chen L, Shao M, Li C, LiuY, Zhou Y, Jiang H, Li X. Effects of home-based cardiac exercise rehabilitation with remote electrocardiogram monitoring in patients with chronic heart failure: a study protocol for a randomized controlled trial. BMJ Open. 2019;9:e023923. 10. Dickinson MG, Allen LA, Albert NA, DiSalvo T, Ewald GA, Vest AR, Whellan DJ, Zile MR, Givertz MM. Remote monitoring of patients with heart failure: a white paper from the Heart Failure Society of America Scientific Statements Committee. J Card Fail. 2018;24:682–94. 11. Artico J, Zecchin M, Zorzin Fantasia A, Skerl G, Ortis B, Franco S, Albani S, Barbati G, Cristalli J, Cannata A, Sinagra G. Long-term patient satisfaction with implanted device remote monitoring: a comparison among different systems. J Cardiovasc Med. 2019;20:542–50. 12. Catalan-Matamoros D, Lopez-Villegas A, Tore-Lappegard K, Lopez-Liria R. Patients’ experiences of remote communication after pacemaker implant: the NORDLAND study. PLoS One. 2019;14:e0219584. 13. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report: ethical principles and guidelines for the protection of human subjects of research. Bethesda, MD: US Government Printing Office; 1978. 14. United States Food & Drug Administration. Device software functions including mobile medical applications. https://www.fda.gov/medical-devices/digital-health/device-software- functions-including-mobile-medical-applications. Accessed 13 Dec 2019. 15. Walker RC, Tong A, Howard K, Palmer SC. Patient expectations and experiences of remote monitoring for chronic diseases: systematic review and thematic synthesis of qualitative studies. Int J Med Inform. 2019;124:78–85. 16. Maher NA, Senders JT, Hulsbergen AFC, Lamba N, Parker M, Onella JP, Bredenoord AL, Smith TR, Broekman MLD. Passive data collection and use in healthcare: A systematic review of ethical issues. Int J Med Inform. 2019;129:242–7. 17. Mittelstadt B. Ethics of the health-related internet of things: a narrative review. Ethics Inf Technol. 2017;19:157–75. 18. Bierman AS. Functional status: the sixth vital sign. J Gen Intern Med. 2001;1611:785–6. 19. Petersen C, DeMuro P. Legal and regulatory considerations associated with use of patient- generated health data from social media and mobile health (mHealth) devices. Appl Clin Inform. 2015;6:16–26. 20. Thomson EA, Nuss K, Comstock A, Reinwald S, Blake S, Pimentel RE, Tracy BL, Li K. Heart rate measures from the apple watch, Fitbit charge HR 2, and electrocardiogram across different exercise intensities. J Sports Sci. 2019;37(12):1411–9. 21. Turakhia MP, Desai M, Hedlin H, Rajmane A, Talati N, Ferris T, Desai S, Nag D, Patel M, Kowey P, Rumsfeld JS, RussoAM, Hills MT, Granger CB, Mahaffey KW, Perez MV. Rationale and design of a large-scale, app-based study to identify cardiac arrhythmias using a smartwatch: the Apple Heart Study. Am Heart J. 2019;207:66–75. 22. Balthazar P, Harri P, Prater A, Safdar NM. Protecting your patients’ interests in the era of big data, artificial intelligence, and predictive analytics. J Am Coll Radiol. 2018;15:580–6. 3 Remote Patient Monitoring: Delegation of Responsibility
50 23. Faden RR, Kass NE, Goodman SN, Pronovost P, Tunis S, Beauchamp TL. An ethics framework for a learning health care system: a departure from traditional research ethics and clinical ethics. Hastings Cent Rep. 2013:S16–27. 24. Geis JR, Brady AP, Wu CC, Spencer J, Ranschaert E, Jaremko JL, Langer SG, Kitts AB, Birch J, Shields WF, van den Hoven van Genderen R, Kotter E, Jochoya JW, Cook TS, Morgan MB, Tang A, Safdar NM, Kohli M. Ethics of artificial intelligence in radiology: summary of the Joint European and North American Multisociety Statement. Radiology. 2019;293:436–40. 25. Wang L, Alexander CA. Big data analytics in biometrics and healthcare. J Comput Sci Appl. 2018;6:48–55. 26. American Medical Association. Digital implementation playbook. 2018. https://www.ama- assn.org/amaone/ama-digital-health-implementation-playbook. Accessed 25 Dec 2019. 27. National Institute for Standards and Technology. Securing the telehealth remote patient monitoring ecosystem. https://www.nccoe.nist.gov/sites/default/files/library/project-descriptions/ hit-th-project-description-final.pdf. Accessed 25 Dec 2019. 28. European Commission. Privacy code of conduct on mobile health apps. https://ec.europa.eu/ digital-single-market/en/privacy-code-conduct-mobile-health-apps. Accessed 25 Dec 2019. 29. American Heart Association. Using remote patient monitoring technologies for better cardiovascular disease outcomes guidance. 2019. https://www.heart.org/-/media/files/about-us/ policy-research/policy-positions/clinical-care/remote-patient-monitoring-guidance-2019.pdf? la=en&hash=A98793D5A04AB9940424B8FB91D2E8D5A5B6BEB. Accessed 25 Dec 2019. 30. Brahmbhatt DH, Cowie MR. Remote management of heart failure: an overview of telemonitoring technologies. Card Fail Rev. 2019;5:86–92. 31. Noah B, Keller MS, Mosadeghi S, Stein L, Johl S, Delshad S, Tashjian VC, Lew D, Kwan JT, Jusufagic A, Spiegel BMR. Impact of remote patient monitoring on clinical outcomes: an updated meta-analysis of randomized controlled trials. NPJ Digit Med. 2018;1:20172. 32. Shufelt C, Dzubur E, Joung S, Fuller G, Mouapi KN, Van Den Broek I, Lopez M, Dhawan S, Arnold CW, Speier W, Mastali M, Fu Q, Van Eyk JE, Spiegel B, Merz CNB. A protocol integrating remote patient monitoring patient reported outcomes and cardiovascular biomarkers. Digital Med. 2019;2:84. E. A. Krupinski and J. A. Pagliaro
RkJQdWJsaXNoZXIy MTYzOTI3MA==