Thursday, December 31, 2009


Answering this question first requires an understanding of the legal definition of a medical device. Section 201(h) of the Federal Food, Drug, and Cosmetic Act, defines a medical device as, "... an instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including any component, part, or accessory, which is ... [either] 1) intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals ... [or] 2) intended to affect the structure or any function of the body of man or other animals." In short, a medical device is a physical, mechanical or software product that is intended for medical use. Note that the definition includes components, parts and accessories. With a medical device system, the network, server, client devices and related software are all part of the regulated medical device.
For many years after medical device systems first came to market, medical device manufacturers were responsible for the installation, service and ongoing support of their entire systems -- including the network. Medical device manufacturers maintained this control by installing physically separate networks, resulting in "islands of information." Recently, there has been increasing pressure to deploy medical device systems on enterprise networks. There are various reasons for this change. For medical devices that are deployed throughout the enterprise, it is not cost effective to install separate hospital-wide networks for each make and model of medical device that's widely deployed. Such an approach could easily result in multiple networks for patient monitors, telemetry systems, smart infusion pumps, and point of care diagnostic testing devices -- four networks in addition to the enterprise network. Medical device connectivity applications like alarm notification or advanced clinical documentation are also pulling medical device systems out of their private network environments in order to break down those islands of information and get data into other systems.
If you have a medical device, including any component, part or accessory that is running across a portion of your enterprise network, your network has become part of that regulated device. With the adoption of devices like wireless patient monitoring and smart infusion pumps, the only way to really determine if your enterprise network is a medical device (or part of one) is to do a comprehensive survey. Each medical device system should be identified, inspected, reviewed and documented.
So what if your network is part of a medical device system? The potential patient safety impact is such that the FDA hosted a meeting to consider regulating the deployment and use of networked medical devices in 2005. The outgrowth of this study group is IEC 80001, a "voluntary" end user standard for networked medical devices. (Voluntary is in quotes because it is expected that adoption of the standard will be mandated by some organization once the standard is completed in 2010.)
Medical device systems running on hospital enterprise networks do require an organizational response. The main justification is that unforeseen or unmanaged network problems -- problems that may have nothing to do with the medical device system -- could cause a failure that results in a patient injury or death. The challenge is that by moving medical device systems onto shared networks, the industry has created a chimera, a type of system that has characteristics for IT networks and biomedical devices but is different and thus something new in the hospital. In short, neither existing policies and procedures in IT or Biomed are sufficient to properly manage networked medical devices (hence the push to create IEC 80001).
There's been a lot of talk about moving the reporting relationship of Biomed from Facilities (where most Biomed departments traditionally report) to IT. In fact, the lines on the org chart have very little to do with meeting the patient safety challenges of networked medical devices. Besides the survey mentioned above, there are several other industry practices emerging for dealing with this issue.
Operating Framework
The first challenge is providing an operational framework for IT and Biomed to effectively collaborate. Issues include:
  • Who "owns the clock" on networked medical device issues, information system support for these issues (IT help desk apps and Biomed asset management apps both have holes);
  • The creation of a set of operating policies and procedures for managing this IT/Biomed chimera of networked medical devices, including bring IT discipline to medical device system installations and      configurations, and applying BIomed risk management to the enterprise network and networked medical devices;
  • The implementation and application of the above operating policies and procedures to existing medical device systems, including gathering detailed specifications and regulatory details from medical device manufacturers, and audits of current network and system environments; and
  • The application of all the above to the purchase and implementation of new medical device systems, including messaging middleware, medical device connectivity solutions and any other application that directly utilizes medical device data (all of which the FDA considers medical devices).
By removing these medical device systems from their private networks, onto the enterprise network, medical device manufacturers have lost control of the operating environment for their systems after their initial installation. The responsibility to maintain that operating environment -- which is increasingly your enterprise network -- falls on the health care provider.
In hindsight, perhaps the best medical device system is the one that IT never has to know about. But the reality of technology adoption is that medical devices are becoming just another information appliance on the enterprise network, albeit a rather specialized one. Continued workflow automation at the point of care means that there's no going back to the good old days of stand alone medical device systems. Future blog posts will delve into other medical device connectivity issues, how they impact the enterprise, and the standards and industry practices that are evolving to meet the challenges outlined above.

No comments:

Post a Comment