Is Your WLAN Ready for VoIP?
The short answer? Probably not. How about RFID, wireless medical devices? Amost certainly not.
Most hospital WLAN deployments entail throwing up APs (access points) every 100 meters based on eyeballing a floor plan or running a heat map application (that estimates AP placement). Given the frequent tweaking of AP placement after deployment – even for simple data applications like meds administration or charting – one might think there's a lack of awareness as to what's really required to plan and deploy a WLAN in a hospital, and get it right the first time. That's exactly what I explain in a column published in today's Health-IT World News email newsletter, titled Wireless VoIP Reveals WLAN Challenges.
I spoke with Bruce Van Nice of Trapeze Networks and Geri Mitchell-Brown from SpectraLink for the column. Bruce and Geri mentioned they had an educational challenge with both hospitals and resellers. According to Geri, “Many in IT think of a wireless LAN as an extension of wired Ethernet, but the introduction of the RF link to wireless devices adds a new level of complexity, like finding optimal AP placement and dealing with things like multipath interference.” The most frequent problem that Geri's seen is when customers don't ask for help before implementing the system – then SpectraLink frequently gets calls complaining of “noisy handsets” or dropped calls, which are really network deployment issues.
Not mentioned in the column is that SpectraLink is the wireless VoIP vendor used by Memorial Hospital. SpectraLink has been selling Wi-Fi VoIP for over six years. They've got an interesting program for certifying WLAN infrastructure vendors. They evaluate the WAN vendor's implementation of technology standards, performance metrics, and then run a pre certification test plan – twice, once by the WLAN vendor in their lab and once by SpectraLink in their lab. Each vendor ends up with the other's equipment in their lab so either or both of them can reproduce and fix customer problems. This lab approach is standard operating procedure in the IT systems integration world – but it's virtually unheard of in medical device integration, which is too bad.
Performance metrics has a big impact on resulting application performance. Not too long ago I was involved in an evaluation of WLAN infrastructure vendors for a medical device vendor. Trapeze Networks was on the short list because of their mobility strengths, in particular, handoffs between APs and roaming across subnets. Many mobile hospital users roam across wide areas – respiratory techs and monitored patients going to surgery or diagnostic departments come to mind. Applications in hospitals, besides VoIP, require low latency and reliable handoffs.
Also not mentioned in column is that your VoIP needs assessment should look at wireless voice system's “intelligence” like the ability to ask for the “closest case worker” rather than just dialing numbers. And if your goal is to design and deploy your WLAN once, be sure to consider wireless device proliferation and the resulting capacity that will be required – also don't forget wireless medical devices. At the last HIMSS Welch Allyn introduced a wireless vital signs monitor, and GE and Spacelabs introduced wireless patient monitors – all on 802.11b. The smart pump vendors were out in force as well. Medical device support is similar to VoIP; roaming and latency are important, as are a number of other things.
If you've had any experiences like these (or would like to avoid them), give me a call. You can subscribe to the Health-IT World newsletter here.
UPDATE: I guess it would help if I mentioned the key WLAN applications that present the biggest differences in requirements for WLAN planning and site surveys: they are broad scale bedside data apps (like CPOE), indoor positioning/RFID, wireless VoIP (or VoWIP), and wireless medical devices.
Some wireless medical device applications may be pretty wimpy now (like updating drug formularies in infusion pumps) but the end game includes life threatening alarms (LTAs), continuous monitoring, and someday device interoperability with other devices and information systems. When you implement one of these applications, your hospital network will become part of an FDA regulated medical device (covered by their 510(k) and intended use statement). Relax though, it's not as scary as it sounds; but it will mean substantial changes.
Analogic Introduces New Non-invasive Cardiac Monitoring Features
What first read as a re-introduction of the LIFEGUARD II patient monitoring line, turns out to be new non-invasive cardiology capabilities (press release).
Electrical Bioimpedance (TEB) to measure continuous cardiac output, and
is the only patient monitor that has ICG totally integrated with all
other standard, non-invasive vital signs.
Non-invasive
hemodynamic measurement is a valuable, established tool in managing
cardiac patients, and the LIFEGARD II with ICG is the most
comprehensive non-invasive cardiac output monitor available today.
In the monitoring world, this is pretty cool stuff. I wonder if the ICG is also available as an OEM module? With this introduction, they've even updated their web site with LIFEGUARD II product info!
JCAHO to Make Hospital Data Public
In a surprising demonstration of business acumen, JCAHO has abandoned plans (reg. req.) to sell hospital data and will instead make the data public.
The JCAHO is the outfit that inspects and then accredits hospitals. Without accreditation, hospitals would lose a major portion of their reimbursement (not to mention garner a lot of bad publicity). Who knows who came up with the original ideal that JCAHO President Dennis O'Leary called, “commercialization of data.” The data in question is data that's gathered from hospitals during the accreditation process – something hospitals spend lots of money to prepare and then pay lots more to JCAHO to take and analyze.
The cynic in me thinks that hospital protests about “privacy violations” had more to due with bad feelings over the fact that JCAHO wanted to profit further from work hospitals had already paid them to do. In any event, data that JCAHO gathers surely contains some interesting information. They will garner tremendous good will and possibly a much greater awareness among the general public by making the data public – that beats angering your customer base any day.
Open Source Software Will Save Money
Only in health care would a foundation need to do a study to document the value of open source software. Used for years in other commercial and industrial market segments, open source software is also frequently built into “proprietary” systems from HIT and even medical device vendors. The potential impact needed to spur IT adoption in health care is commoditization, where lower priced products see more rapid market adoption. The adoption of industry standards also be a boon to adoption.
The report said that industry must rethink its current practices and
“look at ways to add value to commonly used software rather than
producing competing, though functionally equivalent applications.”
With open standards in place and an effective open-source
community established, collaboration could fill the gaps in EMR
(electronic medical records) technology and network infrastructure.
“Where those standards exist, open source is a good candidate
for implementing them at low cost because there will be less motivation
for proprietary vendors to invest in competing products,” said the
report.
You can download your own copy of the report here (pdf).
General Dynamics Contracted for $1.1 Million Nurse Call System
In news that triggered “$900 screwdriver” flashbacks, the U.S. government has contracted with General Dynamics to provide design and installation services for the nures call system at Bassett Army Community Hospital. The Army Corps of Engineers is building a new 43 bed hospital at Bassett to replace the existing structure. For those interested in savoring the absurdity, that's over $25,000 per bed for design, cabling, installation, and hardware.
Certainly someone at General Dynamics is more interested in publicizing new contracts than ensuring they cast General Dynamics (and the Army) in a good light. There must be more to this than a simple nurse call system.
