Patient Support Program and Remote Cardiac Monitoring via Mobile Phone

Patient Support Systems Applied to Cardiac Monitoring

The healthcare ecosystem has been rapidly evolving over the past few decades, and it seems to be accelerating. Developing more cost-effective methods of addressing chronic diseases, such as cardiac-related illness, is a high priority.  The fairly recent evolution of Patient Support Programs (PSP) is increasingly seen as having the potential to address this need.  


PSPs, which have been around for over a decade, have more recently been focused on integrating devices capable of real-time collection of a broad range of bio-marker data. The systems evolved from pharma systems that supported the protocols associated with high-cost, high-effectivity drugs and now consist of more sophisticated systems collecting information from an array of connected devices such as blood glucose meters, body water analyzers, respiration and cardiac monitors. These systems have created a gateway to the next generation of patient monitoring systems. More specifically, the rapid expansion of advanced remote cardiac monitors over the past few years has created a major market opportunity as well.  Representing over $5 billion dollars in 2022, current projections are that the market could reach as much as $31 billion by 2028 (a CAGR of over 30% per year). 


Healthcare is endeavoring to shift from solely treating disease within hospitals and physicians’ offices, to monitoring a much broader cross-section of the population with the goal of keeping them out of more acute care settings. Integrating these richer sensors with clinical analytics in backend systems enables the identification of underlying trends and escalating interactions and interventions with prospective patients. Early results indicate this approach as the ability to meet the goal of significantly reduced hospital admissions. 


Extending a medical data collection platform from a clinical setting to something much more akin to a commercial system has real implications in product development. Building general-use clinical decision support platforms denotes a significant change in the overall functionality of the software system, requiring a very different set of product drivers, architectural approaches and technology trade-offs. Already, AI-based software systems can be tuned to identify AF in asymptomatic patients. Emerging AI algorithms have made strides in assessing streams of cardiac data to identify subtle shifts in the behavior of the heart, some of which are precursors to a number of different cardiac events. Recent advances in MEMs-based microphones expand the ability to acoustically monitor both heart and lung performance simultaneously and continuously.  


What are the product implications? 

The ultimate goal of these emergent products is to provide a first level of interaction for patients outside the acute care setting. However, potential issues can range from identification of an imminent critical health event, such as an MI, to recognition of early symptoms of a more chronic issue such as AF – to a IT issue or problem with the sensor/collector interface. Designing a sophisticated platform, able to address this range of problems is, to say the least, challenging.  The following characteristics are critical to the platform. 



The ability to observe the health of a distributed system is critical.  If no health events are being reported, how can we be certain that no events have occurred as opposed to a partial system outage?  For a system that is highly observable, it will be very quick to identify where there is a break in connectivity.   


Beyond the potential for a system outage, a system must handle multiple versions and/or generations of components in the patient’s hands.  The system must know which component is being used by each patient.     


Design for Inconsistent Connectivity 

Putting connected devices in the hands of patients in their home setting opens up the opportunity for highly variable connectivity.  Whether the device is using WiFi or Cellular, connectivity is dependent on a consumer infrastructure that is out of your control.  Gaps in connectivity must be expected and designed into the overall system. 



Distributing connected medical devices almost guarantees that eventually someone will try to hack one.   Even if it is just a researcher who uses your product as an example at Black Hat or any other cybersecurity conference, the negative impact of an exposed vulnerability can be significant.  Approaches, such as a Zero Trust model reduce the potential and impact of a malicious device.    


Design for Analytics & AI 

Analytics and AI are very useful in supporting service optimization.  In order to unlock this potential, the system must be designed to support data analysis, data experiments, and data privacy.   


Ultimately, these platforms will operate with a degree of autonomy, interacting directly with patients and issuing medical recommendations.  All of this comes within the framework of evolving FDA regulations. New regulations addressing cyber security, AI-based analytics, Software as a Medical Device (SaMD), add new levels of process rigor requirements, and post-release obligations.  



The new generation of remote cardiac monitors will save enormous numbers of lives, of this we can be sure. However, profitably deploying these new devices will challenge many of the existing approaches to medical product development. The technologies around large-scale, distributed systems are rapidly evolving and the impact on the medical device industry is unique. 


The fundamental changes to system architecture, especially to support more effective cyber security, AI and system risk analysis, are forcing adoption of more sophisticated software development processes. Couple this with the short cycle nature of today’s tech stacks and there are real hurdles to market entry for monitoring equipment manufacturers.  


Existing product development teams must rebuild skillsets to adapt, or products will be late to market and ultimately, not flexible enough to compete. As we have seen, this is far more than learning a new language or platform. The fundamental approach to designing, building, and maintaining a distributed system connecting medical devices represents a sea change for the industry.