The State of Central Station Patient Monitoring – Market Drivers
Last week Frost & Sullivan announced that they’re kicking off an “exciting new study” on the state of centralized patient monitoring in the U.S. They’re going to go department by department to analyze centralized patient monitoring in hospitals and ambulatory care. Their goal is to “elucidate key drivers and trends that are transforming the concept of central monitoring.”
The market is indeed changing, although the glacial rate of change typical to health care makes change hard to recognize. Here’s the hypothesis for Frost’s study:
“The future of hospital patient monitoring is being decided at the central stations,” notes Frost & Sullivan Research Analyst Nathan Cohen. “The entire concept of centralized patient monitoring is evolving as clinical information systems are integrated with patient monitoring networks.”
Certainly data acquisition from patient monitors into paperless charting systems (EMRs) is the biggest driver for spending in the patient monitoring market. Imagine dealing with the scenario a 500+ bed east coast hospital is facing:
- Most of their data generating medical devices are 5 to 10 years old and predate any connectivity features like HL7 interfaces;
- 12 central stations supporting 4 different models of patient monitors of various vintages;
- 29 different networks (private subnets) some as old as 15 years – they have Thick-Net, Thin-Net and twisted pair cabling;
- Two separate telemetry systems supporting a total of 300 transmitters;
- Wireless monitors using 2,400-2,484GHz 802.11FH, channelized 608-614MHz WMTS and 174-216MHz VHF (the discontinued telemetry band, now unprotected);
- About 200 spot vital signs monitors; and
- A bunch of stand alone IV pumps and ventilators (i.e., without any connectivity).
The fact that they’ve standardized on one patient monitoring vendor doesn’t really help here, since the vendor’s proprietary solution is for the hospital to “upgrade,” i.e., buy all new equipment that supports connectivity. They’re probably looking at around $20 million for their EMR software – I doubt they’ve got another $10 to $15 million for all new monitors, pumps, etc. – a different solution beyond upgrades will be needed to fix these problems at most hospitals.
While the EMR is the be-all end-all for IT, the clinical side of the house is focused on automation at the point of care (POC) to improve patient safety and staff productivity. Patient monitoring (both continuous and spot) and central stations are key players at the point of care. In addition to monitors though, you have other medical devices (pumps, vents, enteral feeding pumps, dialysis machines, and point of care diagnostics), nurse call systems, wireless phones, and overhead pages. Caregivers now receive life-critical alarms though all of these devices and systems – located behind closed doors in patient rooms, at the central station, or carried by the nurse. All these alarms and alerts are annunciated differently, with different alarm types. Since most of these alerts and alarms are broadcast throughout the nursing unit, caregivers rarely know when an alarm is intended for them (is generated from one of their patients), or one of their peers. A common result of this mish-mash is alarm fatigue, sometimes resulting in failure to rescue and sentinel events. The human factors environment on nursing units is arguably hostile to caregivers and unsafe for patients.
Senior management also has their priorities that impact the delivery of patient care. Most hospitals experience patient flow bottlenecks in their critical care areas: ICUs, step down units and telemetry. An increasingly popular mitigation strategy is the adoption of variable acuity units where certain ventilator patients and monitored patients are kept on their general care units, avoiding admission (or readmission) to critical care units. The concept here is to deliver the appropriate level of care in the lowest cost setting. The effective implementation of this care delivery model can have a significant impact on hospital revenue and operating costs, not to mention staff and patient satisfaction. As more aggressive therapies – and associated surveillance – is delivered outside of critical care areas, the cost of having monitoring techs watch central stations can be significant. Whether the approach is to put one monitor tech per nursing unit to watch patients or centralizing the function and putting all the monitoring techs in one room with a bunch of screens, the cost can be significant. I know of another east coast hospital, this one with almost 800 beds that spends $1 million annually on their central monitoring surveillance.
Now the driving force of medical device data acquisition in support of EMR adoption is properly re-cast as one of several factors driving change at the point of care. It should also be clear that hospitals have to deal with more than just their patient monitoring vendor(s) to respond to the many competing priorities and initiatives around the point of care.
UPDATE: Frost analyst Nathan Cohen comments below, encouraging readers to contact them and participate in their study. He also agrees that, “One-vendor solutions have proven insufficient to meet the needs of complex clinical IT/application environments.” He goes on to describe a recent site where he saw things pulled together by Emergin. Click the “comments” link below to read the whole thing.
NOTE: Since changing content management software the comments made with the the previous software are no longer available.
Wireless Patient Monitors, WMTS or ISM?
The FCC and FDA designated two portions of spectrum for wireless telemetry, WMTS (wireless medical telemetry service) and ISM (the Industrial, Scientific and Medical band). Since the inception of WMTS in June of 2000, and the release of WMTS and ISM based medical devices, there has been a lot of talk about one band being more “protected” from interference than the other.
There are two kinds of RF interference, intentional, and unintentional. In the previous string, the talk has been about intentional interference from HDTV, TV, and PLMR (but NOT ham radios).
By far the most frequent (and troublesome) form of interference is unintentional. Bad fluorescent light ballasts, elevator motors, microwave ovens and other RF radiators in and around hospitals.
Vendors have made a lot of hay about how WMTS is “protected.” It is protected, but only from intentional interference. And even then, you must identify the source and go through a formal process to resolve the situation. This could take weeks or months to resolve.
WMTS is a designation of frequency ranges, a piece of RF real estate. Vendors can develop any kind of communications schemes within that band that they want. There are no standards built into WMTS to ensure coexistence, maximize capacity or minimize the impact of interference, intentional or otherwise.
The ISM band (approved by the FDA and FCC for medical telemetry) is also a piece of RF real estate, but it includes standards that everyone must follow to ensure (or at least facilitate) coexistence, enhance capacity and minimize the effect of interference (either power level restrictions or protocols like 802.11a/b/g). This is why GE (and soon Philips) is employing ISM-like protocols (spread spectrum frequency hopping) in their WMTS products.
ISM has many times more bandwidth than WMTS, thus providing greater theoretical capacity, and more room to avoid interference. In larger facilities, this may be an issue. Also, WMTS coexistence between two or more vendors is a big issue. I’ve seen a GE system hop all over a channelized Philips system, rendering the Philips system ineffective. WMTS infrastructure, i.e., access points, are proprietary — although the network cable pulled to them is not.
In the real world of patient monitoring and surveillance, WMTS and ISM are substantially equivalent. Which is best depends on the details of your particular situation, and the vendor(s) involved.
The real criteria come into product features like scalability, reliability, ease of use, and support for your clinical workflow. Oh ya, and the devices that hang off the network.
Philips Telemetry to Use New Hospital WLAN Technology
According to an updated page on their website, Philips has abandoned their “channelized” radios for a frequency hopping, bi-directional, spread-spectrum WLAN in the WMTS band. They’ve joined GE and Datascope with similar WLAN technology in WMTS. They couldn’t resist coming up with their own name, so they’re calling it “Smart-hopping Technology.” From this diagram, it seems that they’re using a private LAN connecting their access points to their central stations. Apparently they can bridge to the Hospital LAN at the central station or a server for HL7 connectivity.
This means that for all of you who recently upgraded your Philips Telemetry system to move to WMTS, you will need to install a new network infrastructure that duplicates the antenna system of the older channelized system if you expand your system. There is no press release or other information offering any more details at this time.
[Update] I got a nice email from a contact at Philips who said that Philips has not publicly announced their new telemetry network (although the page is still up on their site).
[Update] More info on Phiip’s new wireless network. They will continue to offer their channelized radios (in the 608-614Mhz band) for traditional Telemetry deployments and international markets. The new network uses the upper 1.4Ghz WMTS band, and with greater capacity for networkd devices, is geared for hospital-wide monitoring deployments.
Philips has developed a component radio module that can be integrated into a range of monitors. The radio and APs are based on the DECT protocol (Digital Enhanced Cordless Telephoney). This standard from Europe is used for private wireless phone systems, like SpectraLink phones. PMS is leveraging a core competancy in DECT technology that Philips corporate has developed. You can find an introduction to the standard here (requires registration – email me and I’ll send you the .pdf). They chose DECT because, like voice, continuous monitoring requires requires low latency and virtually no dropped packets.
Leveraging DECT provides an industry standard, with associated economies of scale and robustness. However, by creating their own WMTS implementation of DECT, Philips has created a proprietary network. Change vendors and some of the network infrastructure will have to be replaced. Multi vendor coexistance at 1.4Ghz is also a question.
The radio is currently integrated into their telemetry monitors, but won’t be available in multi parameter monitors until this summer. Current capacity of the network is 128 devices. This will be expanded to 512 devices by this summer, and is theortically expandable beyond 512.
Infrastructure is made up of (from the edge in) access points (APs), controllers (thin APs?) and switches. The APs are power over Ethernet (POE) where power and network connection are on the same wire.
