The adoption of Value Based Care and Patient Centricity have emerged as key element in how health care is delivered and paid has opened doors to exciting changes in healthcare . The emergence of Internet of Medial Things (IoMT) and the concept of Digital Health or more specifically connectivity between patients, clinicians, machines and care environments is further expanding opportunities to better engage patients and to more consistently adopt evidence-based medicine in diagnosis and treatment decisions.
The application of the Internet of Medical things along with scalable and advanced analytics is addressing critical need to improved outcomes, reduced costs and ultimately, provide greater access to high-quality care for more people across the globe. These events, already in progress, have been dramatically accelerated by the challenges that the Global Covid Pandemic.
The ubiquity of low cost and reliable sensing technology as well as reliable and low latency communications infrastructure, like 5G for example, for both consumer and clinical devices has created an exponential increase in the pace and scale of innovation. As stated before, the Covid Pandemic has accelerated almost all aspects of their Digital innovation and transformation in healthcare. From diagnosis to treatment recommendations to the ability to extend access to healthcare via remote or telemedicine engagements has become a “new normal” Such innovation is driving efficiency, reduced costs and real world applications personalization of care and convenience. This is especially true in the area of Remote Patient Monitoring or RPM.
When thinking about the application of remote sensing of physiological parameters one cannot help but begin to understand the challenges of scaling such a scheme to millions if not hundreds of millions of patients across the globe. Traditional Internet of Things approaches may not be appropriate. An important element in facilitating both scale and the integrity of RPM will be a move to the “EDGE”, that is the ability to sense and apply analytics at various points throughout the network including the deployment of analytics, alerting and decisioning at the point of the sensor. A critical element in development and deployment of RPM schemes at the edge will be the concept of Software as a Medical Device (SaaMD).
SaaMD is a formal concept defined by regulators as “Software that can diagnose conditions, suggest treatments, and inform clinical management”. IoMT and SaaMD will accelerate digitalization in clinical research, development and care delivery and drive the development and Adoption of Digital Health What then does RPM look like in the context of digital health, SaaMD, and IoMT. Formaly RPM is the collection and analysis of patient physiologic data that are then used to develop and manage a treatment plan related to a chronic and/or acute health illness or condition. Such services must be provided by clinical staff and importantly are not considered the be “diagnostic” rather RPM services are about collecting patient data remotely. Theses data, in the hands of clinitians, can be used for both monitoring exisiting conditions as well as providing insight in to the over all health of the patient.
There are some key rules as to how RPM should be delivered with regard to reimbursement;
- The RPM device must digitally (i.e., automatically) upload patient physiologic data.
- Data cannot be self-recorded or self-reported by the patient .
- Use of an RPM device to digitally collect and transmit a patient’s physiologic data must be reasonable and necessary for the diagnosis or treatment of the patient’s illness or injury
- The RPM device must be used to collect and transmit reliable and valid physiologic data that allow understanding of the patient’s health status to develop and manage a plan of treatment.
One very interesting note about the devices used to collect data, they do not need to be FDA approved “Medical Devises” and they do not need to be formaly prescribed by a physician. They simply need to be a product that is manufactured to a level of quality where the measurements the device makes is consistent and accurate, like a commercially available weight scale for example.
The Centers for Medicare & Medicaid Services have Multiple CPT (Current Procedural Terminology ) Codes available for Billing Remote Patient Monitoring. There are currently 5 RPM specific codes that clinitians can use for RPM services. They include codes for the collection and interpretation of physiologic data, for the actualRemote monitoring of physiologic parameter(s), for the devices Devices that supply daily recordings or programmed alerts transmission, each 30 days and two codes specific for the monitoring and analysis of RPM data.
Remote physiologic monitoring treatment management services (20 min.) The benefits of RPM are quite clear and are being demonstrated more and more every day and provide benefits to payers in the form of risk mityigation. Specific areas of interest include the reduction in costs associated with chronic illness. Imroving post/acute care outcomes and reduced readmissions. RPM also supports patient compliance and adherence to treatment wich also results in Improved outcomes. RPM is also a path to more personalized/precision care and provides more immediate visibility into patient health status. The list of use cases for RPM is alo growing as the technology and its deployment matures.
Today there are many examples of RPM across a number of indications, for example;
- Chronic Care Management
- Congestive Heart Failure
- Neurodegenerative Disorders
- Behavioral Health
- Post Acute/Discharge Care – Readmission
And the list of available sensors is growing as well, further driving the adoption of RPM across disease states;
- Glucose Meters
- Heart Rate
- Respiration Rate
- Blood Pressure
- Oximeter – Blood Oxygen Levels
- Scales for Weight Monitoring
- Activity monitoring
- Gait and Ballance Assessment
- Continuous Dementia Surveillance Monitors
- Grip Strength
One element of RPM that can be overlooked is the value of the new forms of data that can now be collected from patients in a more continuous way. Not just when the visit a doctor but possibly continuously over long periods of ttime. Such a scenario challenges the current approach to analytics. That is that the current “analytics lifecycle follows a sort of linear time line and primarily relys on after the fact analysis. New forms of streaming data have been introduced into this analytic flow, but we still rely on a historical and restrospective system.
A new approach to the Analytics life cycle, and one that takes advantage of high dimensionality and high frequency streaming data, where the streaming sensor data are both used to analyse historical trends and patterns but those data are also used to develop analyses and models that can be applied “in stream”. That is scoring the data as it is moving.
Such a system provides real time analysis and perdition and offers the opportunity for more immediate action based on the streaming analysis. More so such a system can incorporate the streaming data and scorning to enrich and rebuild analytic models. In effect an analytic system that is a true leaning system and one which is adaptive and real time.
The Application of such an analytic approach would drive value and improve outcomes through a progressive and “intelligent” stepwise path. From status quo Monitoring to more intelligent and enhanced awareness of the patient that would drive automation and intelligence in terms of alerting, to near real time continuous monitoring to an intelligent and learning that facilitates the delivery of true personalized and predictive care.
Taking advantage of a variety of sensors that can be deployed in a home setting one can begin to measure and analyse paramaters that would both track and predict issues that a clinitian would need to be aware of. Such telemetry and data would be used not only to monitor individual patients and their particular health situation but to also develop new and more sensitive analytic approaches to RPM that would benefit entire patient populations .
Importantly such analysese could be deployed in real time and used, where appropriate for continuous monitoring with the application of event stream processing. Further, such monitoring could be incorporated in to an intelligent decisioning work flow where alerts and recommendations as to actions can be taken can be highy automated, facilitating the scale and speed of such RPM aproaches
Perhaps equaly important to sucessfuly deploy such an approach would be the application of a highly robust system to managing the administrative process of development deployment and monitoric of the automated analytics. Specifically one would require a transparent ability to ensure model governance and transparency, to easily validate models to ensure high-quality predictionsand automate monitoring model performance. Such an aproaqch would close the loop on of data collection, model development, model development into production and model management. The results would be to facilitate and expand the value of the application of remote patient monitoring to improve outcomes.
RPM has always presented to care givers, patients and payers numerous advantages with regard to convenience, cost and outcomes. The current pandemic has put these value propositions front and center. This, along with the growth and development of a variety of autonomous and digital sensing technologies that can be deployed remotely and securely as well as the ubiquity of connectivity available to patients and providers and a formal and detailed reimbursement structure has moved RPM from a “nice to have” to a “must have”.
RPM wil be a critical technology moving forward and it will be a significant component of The adoption of Value Based Care and Patient Centricity and will meaningfuly increase acces to care, reduce costs and improve patient outcomes.
View this video to see an indepth presentation on this topic form SAS Global Forum 2021.
About SAS in IoT
SAS empowers organizations to create and sustain business value from diverse IoT data and initiatives, whether that data is at the edge, in the cloud, or anywhere in between. Our robust, scalable, and open edge-to-cloud analytics platform delivers deep expertise in advanced analytics – including AI, machine learning, deep learning, and streaming analytics – to help customers reduce risk and boost business performance. Learn more about our industry and technology solutions at www.sas.com/iotsolutions
Mark Wolff, Ph.D.
Chief Health Analytics Strategist, IoT, SAS
Principal – Global IoT Commercialization, SAS