AAMI Exhibits
As John Pantano of Radianse noted, “technology has come to health care.” The exhibit floor was notable for the many RFID vendors, and even a connectivity systems integration firm, HCTSI. As is typical for a show with medical device vendors, Emergin was to be found in many booths – Philips, GE, and Cardinal to name a few. I like the fact that exhibit hours do not overlap educational sessions, so I didn't feel the conflict I usually feel at HIMSS or RSNA between checking out the vendor news and attending the sessions.
Agility Healthcare Solutions, an RTLS vendor, was my fist visit. They claim the largest hospital market share, measured in top-to-bottom full deployments (i.e., not pilot sites). This claim may be true, but from what I heard from their competitors the hospital market is taking off and the “top-o-the-heap” designation will likely switch. The Agility system uses a dual frequency tag (433 MHz and 307 kHz) and claims a 5 year batter life. Agility has an interesting pricing model where they install the infrastructure and provide tags for a monthly fee. This is great if you don't want to pop for a capital expenditure up front. This does tend to hid the true cost of the solution, and makes price comparisons difficult. Since HIMSS (this past February) Agility has released a new touch screen client that takes the RTLS application from power users like IT and materials management into the realm of diagnostic departments and patient care area users.
Passing the FLUKE Biomedical booth I thought I recognized an Ekahau tag. FLUKE has indeed OEM'd the Ekahau RTLS, and this was a “pre-introduction” of the FLUKE branded solution. Word was the contract with Ekahau was signed at 9pm the night before the AAMI conference opened – nothing like cutting it close.
The word at the Radianse booth was that they've just closed their sixth big whole-house RTLS deal last week. Their patient flow application has evolved considerably since HIMSS – they've got their first patient flow sale and will install in early fall. They have a new positioning algorithm in beta that's giving them meter resolution. It seems RFID tags are shrinking and Radianse showed a new smaller version of the existing tag that's small enough to fit on top a Philips patient monitoring module – available this fall.
In other RTLS buzz, ECRI has an RTLS report coming out in the fall – and not every RFID vendor will be receiving a passing grade. The report looks at 4 or 5 vendors as a first evaluation of this new technology. There was some grumbling that it was taking so long, but it seems unrealistic to expect a speedy analysis of any technology as new and complex as RTLS.
Draeger showed their new (as in not yet approved by the FDA – not for sale) Infinity TeleSmart telemetry system. I was blown away. What they really have is a patient worn monitor in telemetry drag. The unit is the size of a typical telemetry unit, with a very nice ECG waveform display, and local alarms. The arrhythmia analysis is done in the telemetry unit, so if the patient goes out of range of the central station, alarms are still generated and annunciated. Power management is pretty cool too. The unit sports a 24 hour batter life using an enclosed rechargeable battery. There is a bedside trickle charger to keep batter levels up, and a slick central station charger for use between patients. When you place the unit in the central station charger, it automatically discharges the patient from the central station. The central charger provides a 70% charge in one hour. The unit also uses Draeger's mono-lead, greatly reducing clutter. The OneNet part of the product name denotes that it runs on an 802.11 b/g wireless LAN, like their other patient monitors. Given the problems hospitals have been having with WMTS channel capacity, this could be a real advantage. Overall, the device struck me as miles ahead of the competition; something I would have expected Welch Allyn to come out with.
Ultrasound-based RTLS vendor Sonitor was exhibiting. They entered the hospital market through a science experiment at Brigham & Womens Hospital developed by a group from Harvard and MIT. Like many non-Wi-Fi competitors, their tags run about $20 and have a 5 year batter shelf-life. Because their system works on ultrasound, power consumption is high. Like Wi-Fi based RTLS they use a motion sensor to activate the tag. According to CEO Terry Aasen, the tags will get a 2 to 3 year batter life when deployed on an infusion pump.
Prosec Protection Systems have sold infant protection systems for 10 years. They claim over 400 hospital installations. They OEM their system from Verichip.
I ran into fell connectologists at the HCTSI booth. This new company does systems integration and medical device connectivity for hospitals. Many of their staff came from Quovadx. As a Capsule Tech reseller, they've completed a recent project for Clarian in Carmel (that's CAR-mel, and not car-MEL) Indiana. We talked about the medical device connectivity business and they noted that it seems many hospitals are waiting for wireless connectivity. The hassles of dealing with serial cables with proprietary pin-outs and multiple dedicated connectors is problematic for some hospitals. A wireless term server would be nice but expensive – the real solution will be network connectivity that's integrated into the medical device. Another sticking point were home health devices that depend on the patient to send data – doesn't happen. To date, most hospital connectivity experience is limited to single vendor integrated systems, like wireless patient monitors – where hospitals are shielded from the true cost and complexity of connectivity. Now most hospitals are looking to integrate medical devices into EMRs or departmental systems for the ICU, ED or OR. This type of connectivity takes hospitals (and vendors) into a new realm where needs assessment and vendor selection processes are not known. HCTSI notes that many hospitals express sticker shock when considering the cost of a multi vendor connectivity project. Finally, ROI was mentioned as a barrier to adoption. Until you've implemented an EMR with medical device connectivity, you don't know the pain and additional costs to be incurred by expecting clinicians to manually enter data from patient connected devices.
Philips showed some new products and features. Philips had an evening reception (great food) to introduce their new low acuity monitoring product line to the biomed community. The products range from a vital signs monitor (OEM'd from Collin), to a C1 baby monitor, up to the VM4, VM6 and VM8. The VM series are Philips devices manufactured at a Philips plant in China. The monitors sport 2 to 4 waveforms and all the usual parameter. The units include 96 hour trending and a 4 hour Li battery. The VM 4 through 8 has an Ethernet connector (no wireless) and a USB port for software upgrades. The VM units output HL7 over TCP/IP. This was most surprising. I'm not aware of any other medical device that outputs HL7 directly. Standards like HL7 are not plug and play, requiring configuration for every different information system and site. Unfortunately I couldn't find any technical folks at the reception that could enlighten me as to how each individual monitor is configured, or if they expect HIT systems to simply conform to their implementation of HL7. Moving the industry to a standard HL7 configuration would be a good thing, and if anyone has the clout to make it happen, it's Philips. Philips was also showing an interesting new positioning feature for their DECT-based WMTS wireless network. Since their system knows which access point (AP) a patient monitor is associated with, they have taken that data and used it to return a zone or area inwhich the device is located. They can also deploy APs at choke points to capture when devices leave an area. For missing devices, the system saves the last six known locations providing some digital cookie crumbs that can be used to backtrack and hopefully find a misplaced device. Philips is coy about locating patients (they don't claim that) but if the monitor has an ECG, it must be connected to the patient, right? They're also working on a “key finder” feature that would facilitate finding devices that exit through the laundry chute or trash.
DISCLAIMER: All of the above was gathered wandering around drinking wine during the opening reception. Any inconsistencies, errors, or omissions are completely my responsibility – but as a reader of this blog (whether from a hospital, vendor or elsewhere) me and all of your fellow readers are counting on you to provide corrections, additional information and alternative points of view. You can make your contribution via the comments link below, or via email.
UPDATE: I forgot that one of the things that Paul Coss told me about their new wireless capabilities was a site survey service that Philips now offers – something no hospital should be without prior to installing a Philips telemetry system.
AAMI 2006 – Day One
I usually avoid keynotes like the plague, but I'm glad I caught this one. Mae Jemison is a polymath (biomedical engineer, physician, astronaut) and expert at reframing assumptions about life and health care in surprising ways. She explored health care, examining the intentions and objectives of the industry, and suggesting alternate metrics of performance. Where other industries track performance based on quality or success, health care tracks things like clinical procedures simply on the number of procedures performed with no consideration on patient outcomes. Sure quality is a secondary consideration, but to her point, doesn't that in itself say something? Jemison is funny too. An entrepreneur herself, her definition of an entrepreneur is, “someone who wants to make a lot of money, do it their own way, and isn't so broke they need a real job.” Commenting on the considerable changes being wroght in health care, she observed, “committees are cul-de-sacs where ideas are lured and quietly strangled.” Jemison provided the right note of inspiration, motivation, insight and direction for kicking off this year's conference.
The first session I attended was “New WMTS Band Interference Issues and the Future of Telemetry Equipment.” This was a panel presentation that included John Collins, ASHE; Rick Hampton, Partners; Terry Hinkle, RF Monolithics; and Nancy Pressly, FDA. It seems that many hospitals that migrated from VHF to 608 MHz are having RF performance problems. In my experience, some of this is trying to cram too many transmitters into a channelized system (>240 transmitters), and part of the problem is allowable interference from adjacent digital TV stations. The presenters expect most markets (at least the 100 largest TV markets) to have both channels 36 and 38 by 2009. According to John Collins, the FCC created the 3 WMTS bands (608-614, 1395-1400, and 1429-1432) with the expectation that at least one of those bands would not be available to a hospital, due to co-primary users (radio astronomy) and adjacent frequencies (high power military radar and digital TV). The upper 2 WMTS bands can also be occupied by Itron utility meter reading radios. Makes you wonder why so few vendors using WMTS have products that use the two upper bands, doesn't it? John also emphasized the importance of registering your WMTS use, in order to facilitate frequency coordination. He noted that only 2000 hospitals in the US have registered their WMTS with ASHE. Rick Hampton emphasized the importance of managing and coordinating RF use in the hospital, and with facilities outside the hospital. He also showed examples of intentional interference (digital TV and water meter readers) and unintentional interference (Rick hates paper shredders). Nancy Pressly broached the subject of reporting problems to the FDA (which many hospitals mistakenly do not do). Nancy also mentioned that the FDA will have a guidance document on wireless out in the fall. The bottom line: 15 years ago, you could throw up a telemetry system according to the manufacturer's specs and it just worked. Now a days, things are more complex, and designing and deploying a traditional telemetry system takes real engineering – RF site surveys, a truly balanced antenna, and fiddling with expensive narrow band filters. The contemporary analog is the hospital WLAN. Most WLAN resellers (and their hospital customers), typically throw up a WLAN by placing access points (APs) every 100 meters – and for simple data apps like meds admin or computers on wheels (COWs) they just work. Once you try wireless VoIP, or wireless medical devices, you've got to do some real work. Just like today's telemetry system, WLANs require extensive site surveys, design, deployment, and post installation fine tuning.
The next session was “Physiologic Alarms: Techniques for Standardizing Alarms in Your Institution.” Tobey Clark, from the University of Vermont Technical Service Program, talked about alarm notification issues, and Alan Lipschultz, of Christiana Care, described how they implemented standardized alarms throughout their 2 hospital system. Alarm notification is a pet peeve of mine, and an area where clinical practice and technology can greatly improve patient safety. Tobey described some ACCE efforts that are very important. Alarm standardization is clearly a best practice for hospitals. Tobey mentioned a number of interesting facts, like research that's shown that individuals can't learn more than 6 different alarm sounds. Another study showed caregivers could not identify even one half of common ICU critical alarm sounds when played back. The alarm fatigue problem results from alarms that are inconsistent with the clinical situation. A 2006 paper in the Am. J. Emerg. Med showed that 99.4% of alarms were determined to be false with less than 1% of all alarms resulting in a change in patient management. Many manufacturers' “better safe than sorry” philosophy has led to an increase in the number of alarms, volume and degree of irritation. In addition to all this, many alarms are set improperly – at the wrong limits, alarm priority, volume, or even turned off. The Biomed Listserv was even quoted:
“Do you test them all?”
A search of the FDA's MAUDE database shows alarm related deaths are increasing. In response to this patient safety situation, ACCE has created the ACCE Healthcare Technology Foundation (known as the AHTF) to tackle alarm issues through education and awareness, clinical engineering certification, and researching ways to improve clinical alarm management and integration. The AHTF is looking specifically at alarm design, environmental factors, the role of clinical engineering, and care management. This was the perfect segue to Alan Lipschultz's talk on alarm standardization efforts at Christiana Care. Alan started with a litany of alarm problems, especially nuisance alarms. He described how they assembled a team to evaluate alarms and define standard alarm limits in order to reduce alarm fatigue and zero in on clinically relevant alarms. A great example of this was SpO2 alarms. Manufacturers frequently sent their alarm default at 92%, and physician orders often aim to maintain 92% or better saturation. Because SpO2 fluctuates wildly, alarms set at 92% frequently become nuisance alarms. The Christiana alarm team decided to set default SpO2 alarm limits at 87% (89% in PACU where are less mobile), resulting in a drastic reduction in SpO2 nuisance alarms. Alan provided many more great examples of the alarm policies they've developed. Their patient alarms group is a subgroup under their patient safety committee. To be successful, Christiana learned the team must include strong clinical leadership, and have the backing of both key medical staff and nursing management. Improving alarm management was a great opportunity for Clinical Engineering to take a lead role. They found that once they started working on alarm standardization, different groups of clinicians started talking to one another and to question practices that “have always been done that way.”
Pictured right is Dara McLain, RF interference guru from Philips, during educational session on RF wireless medical devices.
Since the meeting started around noon today, there were a limited number of sessions. More later!
UPDATE: Be sure to check out AAMI Exhibits and AAMI – Day Two.
Point of Care Diagnostics Connectivity
Last year Abbott Diagnostics introduced a new portable whole blood testing system, the Precision PCx for the point of care (POC) diagnostics market. Abbott also has a data management system that supports the Precision PCx.
Early connectivity solutions in a product category have a lot of latitude – end users have little or no previous experience, and early adopters don't worry about too many point solutions cluttering up their environments. As vendors add more sophisticated features – a wireless radio for the Precision PCx comes to mind – vendors may unintentionally obsolete previously released technology. And as the market matures, market requirements start to evolve.
The POC diagnostics market is starting to shape up like previous medical device market segments. A few vendors will get it right, and many will end up spending much more time and money getting to market then they anticipated. Early adopter hospitals will also fall into the lucky (or smart) few and the many that face unanticipated costs and other problems.
Pictured right is the Abbott Precision PCx.
MobileAccess Raises $11 Million in Funding
Distributed antenna system (DAS) vendor MobileAccess, raised $11 million in a round of financing lead by Ziegler Meditech Equity Partners. They're going to use the funds for “aggressive growth” and “undertake a major initiative to further penetrate the healthcare market for wireless information delivery” – no details were provided.
According to the press release, MobileAccess now has over 1,000 clients. The company claims to provide the
flexibility to simultaneously support any mix of current and future
wireless services and features a “Wire It Once” architecture that
enables customers to add new services at any time, in any place,
without disturbing the existing system or services.
The MobileAccess system is pretty compelling, but the DAS market is relatively new and unproven. Early adopters are buying in and gaining some advantages on their investments.
Pictured right is a Mobile Access equipment rack that is deployed in wiring closets throughout a building. Powered antenna elements are distributed from this module via coax cable. Racks are connected to a MobileAccess “head end” (typically in the basement) via fiber optic cable.
Distributed Antenna Systems Tame RF Deployments
As wireless proliferates in healthcare, many
IT departments are looking for an enterprise solution that will better
manage RF technologies, improve performance, and
lower operating costs. Distributed antenna systems (DAS) from MobileAccess and others, promise some degree of enterprise management for RF systems.
Last week Health-IT World published another column of mine, this one dealing with DAS and one hospital's experience with MobileAccess. Check it out to learn how Northwestern Memorial did their vendor selection, and their experience with MobileAccess.
The jury's still out on DAS, and exactly which RF systems should full under a DAS umbrella. Some hospitals plan to put everything on their DAS, and others limit DAS use to supporting centralized transmitters like paging and cell phone transmitters.
If cell phone coverage is an issue at your hospital, be sure to read about Northwestern Memorial's successful approach. Who would have though carriers could be so accommodating?
