Curated by UCL

What Could Total Connectivity Mean for Medtech and for Healthcare?

In the very near future, healthcare systems could be able to connect almost all medical devices to their networks. By 2025, most hospitals will have the ability to network connect more than 90% of their devices, predicted David Niewolny, director,…

In the very near future, healthcare systems could be able to connect almost all medical devices to their networks. By 2025, most hospitals will have the ability to network connect more than 90% of their devices, predicted David Niewolny, director, market development, healthcare, at Real-Time Innovations (RTI). Today, he estimated that “roughly 50% of the world’s installed medical devices and 90% of the devices that have been introduced in the last year have the ability to connect to a larger network. The reason for the large discrepancy is that the refresh cycles on many medical devices are ten years and in some cases more.”

Niewolny said the percentage of connected medical devices will continue to increase as device manufacturers launch products with more and more connectivity. He began working with medical device vendors to include wired technologies in 2006-2008 (USB and Ethernet) and began adding wireless technologies (WiFi and Bluetooth) in 2012-2014.

MD+DI asked Niewolny a few questions about the explosion of medtech connectivity, the efforts still needed to ensure patients benefit from such connectivity, and some of the challenges medical device manufacturers can expect to face along the way.

What are the benefits of such connected devices? For instance, how specifically can they improve patient outcomes, reduce medical errors, and lower medical costs?

Niewolny: There are a number of benefits of connected medical devices. First and foremost is that they greatly increase patient safety. Medical errors are the 3rd leading cause of death behind heart disease and cancer and are responsible for 210,000-440,000 deaths per year. A fully connected healthcare system will capture additional data that will allow multifactorial analysis that cannot be done by humans today. In collaboration with artificial intelligence (AI) and machine learning (ML), clinical decision support algorithms will pave the way for automation, and we will see a continual improvement in patient outcomes along the way. These devices will also reduce unnecessary alarms and reduce “alarm fatigue” by hospital staff. Currently there are 10,000 alarms per day per hospital and more than 85% don’t require attention, meaning that staff often get in the habit of ignoring alarms, resulting in adverse events. Connected medical devices will also provide vast improvements to hospital operations. The initial benefit to users of interoperable, connected healthcare systems will be operational. Providers will get access to real-time information that they have never had before, allowing them to improve work flows and business analytics.

Have there been any studies or reports on how any connected devices have improved patient outcomes, reduced medical errors, and lowered medical costs?

Niewolny: The most comprehensive study was completed in 2013 by West Health, titled “The Value of Medical Device Interoperability.” The focus of the study is on interoperability, not just connectivity alone, but it is important to note that the potential to improve patient outcomes and lower cost drastically decreases with only connectivity. Connectivity and interoperability are what allow the development of true systems of systems, which are the foundation of an IIoT system.

What role would such devices play in value-based systems?

Niewolny: I believe that an interoperable, connected, real-time system is a necessity in order to move to a value-based system of care. The EHR was created as a billing tool, not a patient health record, and the data captured is reflective of that. Because of that, there is not enough data entered to true cost procedures or measure outcomes. As I mentioned above, the introduction of a real-time, interoperable, connected healthcare system would give providers access to so much more information, allowing them to much better quantify the value they are providing to patients. The first step is to access the data; the next steps are to analyze the data and then implement continuous improvements. We can and will begin this process with EHR data, but the impact to value-based care will pale in comparison to what will be done when a real-time system is put in place.

What are all the challenges with such connected devices and how can they be addressed?

Niewolny: In no particular order, the most significant challenges in implementing connected healthcare systems are reliability, security, interoperability, and scalability. Reliability is a challenge because the data must reach its destination. Device manufacturers must adhere to the FDA’s Cybersecurity Guidelines, and devices need to be added and removed from the network easily, but securely, making security another challenge. Interoperability is another challenge because devices need both syntactic and semantic interoperability, and this will be driven by either standards and the creation of an ecosystem or by a single, major player who drives the adoption of its platform. Lastly, scalability is an issue to be addressed because you need the ability to scale thousands of nodes per gateway with little to no impact on performance (latency or payload). Adoption of a data-centric architecture significantly reduces the risk of system failure due to any of the above challenges.

What is meant by ‘data centric?’

Niewolny: When I speak about data centricity, I am speaking about an architecture in which the data within the system is the primary focus. In a data-centric system, the data are centralized and accessed frequently by other components in the system. The components access a shared data structure and are relatively independent, in that they interact only through the data store, which in the case of RTI’s Connext DDS is the Connext Databus. Data centricity at its core provides a more reliable (eliminating single points of failure) and more secure (securing data, not just the network) system. Other benefits of data centricity include syntactic interoperability and true real-time performance. As the IIoT moves from the simple monitoring applications being implemented today to applications needing automation, data centricity become exceptionally important.

Will connected devices play a role in home-based care, and if so, are there other challenges in that environment?

Niewolny: Connected devices play a big role in the market today and will continue to play an even bigger role reach year. Most of the connected health devices in the market today are consumer market focused. We will continue to see a convergence of consumer wellness devices and clinical medical devices targeted for home use. These connected devices will improve the patient experience, at a lower cost, without any decrease in the level of care. These devices need to be considered when building out a complete healthcare IIoT system. In the future, healthcare will not be only delivered within a clinic or hospital. Instead, the home will become an integral part of the care continuum and connected devices will be a big part of that.

Can you explain why real-time connectivity, standardization, and scalability are key; what challenges there are in achieving these goals; and how these goals can be achieved?

Niewolny: Real-time connectivity provides a number of benefits, including a much richer, more complete data set for real-time analysis. It also provides the foundation for automation, an example being with a vital signs monitor controlling an infusion pump. A challenge there is with less than millisecond latency. Standardization provides a foundation for an interoperable ecosystem of devices that can be connected to systems of systems. Some challenges are that medical device business models are not currently aligned to support an open ecosystem. A true healthcare IIoT system will span the entire payer, provider, patient ecosystem, providing access to data to anyone authorized to view it. With scalability, there are challenges like maintaining performance (latency and bandwidth) as well as the fact that the number of edge nodes is in the hundreds of millions

Any other points to share?

Niewolny: There are real challenges to implementing an edge-computing healthcare IIoT system that will provide a highly reliable, automated, and semi-autonomous system for real-time clinical decision support. The adoption of an open, data-centric architecture like Data Distribution Service (DDS) significantly reduces the risk of system failure and provides an interoperable foundation for an ecosystem of connected devices to build upon.