National HIT Rumblings – A Survey of the News
There were some interesting stories about HIT in the mainstream media
last week. Let's start off with the Wisconsin Technology Network. In a
follow up to their recent Digital Healthcare Conference, they report that despite the benefits, EMR adoption is not easy.
fairly expensive. And there are other significant barriers that need to
be overcome, said Dare, a participant in the recent Digital Healthcare
Conference presented by Wisconsin Technology Network. “It's not an
undertaking for the faint of heart,” Dare said in a post-conference
interview, noting the many well-documented challenges involved in
implementation.
It would be more accurate to just replace the word “can be” in the
above sentences with “is”. Along with a review of the forces driving
adoption, interoperability is tagged as taking the longest to come to
fruition.
Forbes investigates
the role physicians play in EMR adoption and describes a broad array of
EMR adoption efforts.. They mention all the usual issues, too old,
what's in it for me, resistance to practice standards, etc. The good (Inland Northwest Health Services), bad and the ugly (Cedars-Sinai CPOE debacle) are all mentioned.
Stalwart of the mainstream media, Time magazine, weights in with a piece titled, The e-Health Revolution (the title gives me deja vu, as in “bubble“).
They picked a great character to profile for the story, Glen Tullman,
who founded one of the top three insurance claims automation software
vendors and went on to start Allscripts. The story goes on to note that
adoption is driving quarter after quarter double digit growth for both
Allscripts and Cerner. In between the usual statistics, they end up
hitting the nail on the head:
wired. Most health care in the U.S., however, is delivered by small
practices with fewer than 10 doctors, and these physicians don't yet
see any payoff. That's because so far there is none. The cost is high,
about $10,000 to $12,000 per doctor, and most of the benefits accrue to
other players in the system, such as hospitals, employers and insurers.
This is a thought that's echoed by Girish Kumar, President of eClinical Works in an interview
on Matthew Holt's Health Care Blog. eClinical Works is an EMR vendor
targeting the largest and most challenging segment of the physician
market, practices with 1 to 15 physicians. Regardless of how much the
government pays, er, helps, Girish believes that we are still 5 to 7
years away from peak adoption rates. This sounds about right to me,
given that this health care and adoption always takes longer than you
think. While he believes that the EMR market is advancing (rather than
going sideways), he hits the nail on the head regarding physician
adoption:
chart in the outpatient environment, although that varies by specialty.
For example, no dermatologist cares, but an ObGyn or cardiology
practice might care. The big deal in terms of interoperability is
access to lab results.
During my ride on the Internet bubble with e-health connectivity startup, Pointshare,
I was struck at how our health care system is so much like a cottage
industry. Patients, employers, physicians, hospitals and payers –
everyone seems to be disconnected. Patients consume, but employers
pay. Physicians direct care at hospitals but work for themselves. And
payers seem to spend most of their time herding cats, with little
success. The two most widely adopted means of communications in health care are
the telephone and fax machine. (And most of the phone calls are either
to ask where the fax is, or to ask that it be resent, “because the
stack of pages on the fax machine is an inch thick and it would quicker
if you could just resend it.)
Broad EMR adoption is going to be like ERP and CRM implementation in
the commercial world; it will be hard, expensive, and there will be no
quick pay-off. The resulting common framework will provide an
invaluable foundation for automating this cottage industry one manual
transaction at a time. Change will be so profound that in twenty years
we won't recognize our industry.
If the scope of the effort and estimated time frame are too depressing,
you can always read one of any number of breathless press releases and
stories about the impending boom, like this one.
Wristwatch Size Continuous Multi Parameter Monitor Announced

Siemens and Tadiran
Spectralink Ltd report the development of a medical device that monitors patient's health. The device,
which is worn on the wrist like a watch, uses biosensors to
continuously check vital functions, which are then transmitted by an
embedded Siemens GSM/GPRS radio module to a hospital for further
analysis. This innovative device will become available early next year
under the name “MDKeeper TM ”. By using the integrated MC55 Siemens
GSM/GPRS radio module (pictured at right), the device can also support
“panic button” calls and voice calls from the device. Even with a radio
module that's 1.5 inches square, this must be one honking big
wristwatch (the only photo available is of the radio, not the complete
device).
unit measures e.g. [sic] patients’ pulse, cardiac rhythm and blood oxygen
values without inconveniencing them. The user, e.g. at-risk patients
with cardiac or circulatory diseases, patients with chronic illnesses
and seniors can enjoy constant monitoring and support without having to
visit their doctor. The information is stored on a chip, analysed and
transmitted in real time or as needed over the GSM/GPRS network to a
medical centre. “This is the first time that a medical control device
has been combined with mobile communication technology, enabling people
in need to be monitored anywhere anytime and maintain their normal
lifestyle” said Itzhak Beni, President and CEO of Tadiran Spectralink. [Emphasis mine.]
This is an intriguing product concept that really gets my mind
spinning on market segments and adoption. With home care reimbursement
in the U.S. almost nonexistent, this might make a better hospital
product. Now with an 802.11 radio and these parameters in a package the
size of a traditional telemetry transmitter…
[Hat tip Wireless Healthcare News]
Read MoreAHRQ Provides Instant Access to Health Statistics Through On-line Database

AHRQ provides users instant access to health statistics and
information on hospital stays from AHRQ's Healthcare Cost and Utilization
Project (HCUP). This project comprises a family of health care databases and
related software tools developed through a federal-state-industry partnership
and sponsored by AHRQ. HCUP includes the largest set of publicly available,
all-payer health care databases in the United States. HCUP data can be queried
through HCUPnet, an on-line system that provides instant access to health
statistics and information on hospital stays from HCUP data. To access HCUPnet,
go to http://hcup.ahrq.gov/HcupNet.asp.
For example, at a hospital I'm working with the fifth most prevalent
DRG is 12 (degenerative nervous system disorders). I can benchmark
their average LOS with national data from 2002 (most recent data). My
client's LOS is 5.47 days longer than the national average for this
DRG. Most of the other DRGs I checked showed them at or below the mean
for 2002. You can get a good overview of the site here.
To speak with an AHRQ expert regarding HCUP data, contact
Bob Isquith at bisquith@ahrq.gov or call (301) 427-1539.
Flaw Found in JCAHO Software
JCAHO revealed yesterday that there
is a flaw in the software hospitals use to help them qualify for JCAHO
accreditation. The software, used by more than 1,000 hospitals, costs
hospitals several thousand dollars annually. The New York Times reports:
Robert Curran, director of clinical excellence at the O'Connor
Hospital in San Jose, Calif., said many hospitals were concerned that
they might have lost quality- control data already entered into their
systems.
The software, which costs members several thousand
dollars a year for hospital groups, and $495 to $1,095 for small
hospitals, is used to help create files showing, for example, that a
hospital has fully informed patients of their rights.
The
problem was a missing identification marker that alerts a hospital to
the 250 standards among the 1,300 that the commission and its auditors
regard as essential. Without the marker, a hospital might overlook
essential categories in which it must verify its compliance.
It's reported
that JCAHO posted this software flaw on their website as soon as they
learned about it. Prompt notification is good, however I could not find
any mention of this on either the JCAHO or Joint Commission Resources web sites. JCAHO is patching the software, hoping to have an update out today.
UPDATE: A software patch is now available from JCAHO here.
Read MoreWMTS Interference Reported at 2005 AAMI Meeting
The following was posted on the Biomed Listserv this morning. Written by Rick Hampton, at Partners Healthcare,
he reports incidents of digital TV and WMTS interference, discussions
on this topic at the AAMI conference last May, and finally requests you
contact him if your facility is experiencing any interference. After
reading this, you might also want to check out a presentation that Rick
made titled, Checklists for RF: Installations and Ad hoc Testing, at MoHCA'S Summit on Electromagnetic Compatibility,
March 26, 2003. And here's an interview of Rick in 24×7 where he talks about WMTS. Emphasis in the following is mine.
disclosed information I had received from two hospitals that had
experienced interference to their WMTS telemetry systems from nearby
digital TV stations operating on TV channels 36 and 38. Some of
the information was received just a few days before leaving for the
conference.
After presenting the information, several of you approached me
afterwards to tell me you knew of other incidents of a similar, if not
identical nature. Interestingly, but perhaps not surprisingly,
FDA has no official knowledge of any events like this. [Here is a link
to the FCC about DTV and medical telemetry.] A search of the
MAUDE database turned up nothing. I also spoke with the folks
from ECRI who would have followed up on such reports and they hadn't
heard of anything either. Both of the hospitals I spoke with
requested I maintain their anonymity.
If you spoke with me regarding another interference event to WMTS
telemetry systems from nearby TV stations, please email me privately or
call me with your contact information. I would like to gather
this information and present it to FDA, FCC, ASHE, and ECRI as I
believe there are actions that could be taken to prevent this from
recurring in the future.
For those unable to make it to the AAMI conference, here are the basics of the interference events:
Kansas, 2003
In mid-2002, the hospital purchased a telemetry system, installed in
September 2002, consisting of 54 channels of telemetry operating from
608.2025 – 610.7250 MHz. At the time of the initial site survey,
the nearby NTSC analog TV38 transmitter was operating at a reduced
power.
The DTV36 transmitter was not yet activated at the time of the initial
site survey. As a result, all channels assigned were in the lower
portion of WMTS band and data from site survey used to design a
broadband antenna system.
Approximately 1 year after installation, the hospital began
experiencing intermittent, severe dropout on all channels. When
the Clinical Engineering department was unsuccessful in troubleshooting
and isolating the problem, the manufacturer was called in to diagnose
the problem. Upon performing a spectrum analysis in February of 2004,
the manufacturer found interference from both the NTSC analog TV38
transmitter operating with increased power and a newly installed DTV36
transmitter.
The manufacturer implemented corrective action in March 2004.
This included the installation of additional filtering and changing the
operating frequencies of all transmitters to the middle 1/3 of the WMTS
band. As a result of the additional RF filtering, the useful
portion of the WMTS band has been reduced, limiting future expansion.
New Jersey, 2004
In Mid to late 2003, the hospital upgraded its existing Part 15
telemetry system to the 608-614 MHz UHF WMTS band. A total of 96
telemetry channels were installed, with 48 channels in Telemetry and 48
additional channels in other hospital areas.
Beginning in November 2003, the hospital began experiencing excessive
dropout on multiple channels. Investigation by the manufacturer
showed high background noise and interference across the entire WMTS
band.
In January 2004, the manufacturer of the antenna subcomponents
performed an onsite inspection and determined the amplifiers and
antennas had too much gain. The manufacturer began replacing the
amps and antennas with lower gain versions. Unfortunately, this
caused the problem to only get worse with the Telemetry unit now
experiencing serious problems. Most telemetry channels were not
functioning properly. The manufacturer then determined the
bandwidth of the new amplifiers & antennas was too broad and unable
to reject energy from the nearby DTV36 & DTV38 stations. Two
more weeks passed before the entire antenna system could be completely
redesigned and installed to preclude overload from nearby TV36 &
TV38.
In the meantime, lower than usual patient census allowed the hospital
to use wired monitors and the remaining unaffected telemetry channels
to continue operation for the duration of mitigation activities.
This system was finally made fully functional across the entire 608-614
MHz band.
In both instances, the Chief Engineers
at the TV stations were contacted and described as being cordial, but
not particularly helpful. When the FCC was contacted, the
hospitals were redirected to ASHE, since ASHE is the official
coordinator for WMTS. When contacted, ASHE advised the hospitals
to continue working with the manufacturers to mitigate the
problems. It is important to note here, the WMTS rules contain no
regulatory recourse for institutions suffering adjacent channel
interference from nearby TV stations. It is up to the facility
and the telemetry manufacturer to solve the problem, either by
additional filtering or moving to a different band.
Louisiana, February 2005
At least one other possible interference event was reported to me, but
with no details. Based on a search of FCC's database, the most
likely areas where this would have occurred is in the vicinity of
Alexandria, New Orleans, West Monroe, or New Iberia. I've sent an
email to all Louisiana hospitals registered with ASHE, however the
report remains unconfirmed. I have received only a handful of
replies, all indicating no knowledge of any such event.
Based on what I've learned from the
manufacturers and hospital CEs, the two main contributing issues
leading up to these events are inadequate antenna design and a
misunderstanding of the WMTS regulations.
In both cases, the WMTS installation utilized a highly amplified,
broadband antenna system, with filtering inadequate to attenuate the
out-of-band radio energy from the nearby TV stations. The antenna
design is a holdover from Part 15 days when telemetry allocations
allowed operation on unused TV channels from 174 MHz – 668 MHz.
In those days, interference from a new nearby station could be remedied
simply by tuning the receivers and transmitters to a new, unused TV
channel further away from the new transmitter. Now, however, with
the WMTS allocation, there is no place to move without requiring the
wholesale removal and replacement of the affected system.
Further, the filters commonly used in these antenna systems are
inadequate for the job. Of the two events described, one hospital
used a legacy antenna system installed some time ago, while the other
had a new antenna installed with the old design. It should be
pointed out here that simply retuning the transmitters and receivers to
the UHF WMTS band may not provide adequate protection from digital TV
interference with this type of antenna system.
The second contributing factor was an inadequate understanding of the
WMTS rules by both the hospital CEs and, apparently, the manufacturers.
While one of the CEs stated he received notification letters from the
TV stations announcing the pending activation of the TV transmitters
(the other has no recollection of receiving the letters), both CEs
state they did not realize remedial action might be necessary to
prevent interference to WMTS systems. I
wish to point out here there is a common misunderstanding of what it
means for the WMTS to be “protected” by FCC. Many people believe
1) all WMTS systems are designed to prevent interference, 2)
interference to WMTS, of ANY kind, will not be tolerated by the FCC,
and 3) the FCC will take action against any interloper. It is
obvious from examining the rules and these cases, none of these beliefs
are correct.
In light of these events, I strongly recommend any hospital utilizing UHF WMTS telemetry systems take the following steps:
1) Foremost, don't panic. While being near a TV36 or TV38 station
certainly increases your risk dramatically, it does not guarantee you
will experience interference, as there are several factors that enter
into the interference equation. There are currently 29 DTV36/38
stations on the air with allocations for another 55, a number sure to
grow. There is time for a levelheaded collection of information.
2) Review all coordination letters you may have received from nearby TV
stations to ensure there are no current plans for activation of a TV36
or TV38 station in your vicinity.
3) Because the FCC's Media Bureau is constantly processing requests for
new DTV allocations and requests for moving from one channel to
another, you should visit their website on a monthly basis to ensure
you haven't missed any coordination letters. You can perform a
directed search for your area at the “TVQ TV Database Query” at http://www.fcc.gov/mb/video/tvq.html. Alternatively, you can check the updated list of DTV stations at http://www.fcc.gov/oet/dtv/start/dtv2-69.txt.
(Note: Pay close attention to the preparation date of any prepared
lists you reference. Many links on the FDA and FCC websites are out of
date and lead back to the original allocation list dated 1998.)
4) Contact the manufacturer of your WMTS telemetry system to ascertain
the type of antenna system and receivers in your system. Request
the manufacturer provide you with a detailed list of upgrades, along
with their costs, needed to prevent your system from suffering from
“adjacent channel interference” from TV36 and TV38 signals.
5) Should you find a new TV36 or TV38 station near activation in your
vicinity, contact the TV station and respectfully request they delay
beginning broadcasts until you have had a chance to perform remediation
activities to your system, if needed. I would suggest you copy
your correspondence to ASHE, FDA, and FCC.
6) If your facility has already experienced interference to your WMTS
system, report it to the FDA and ECRI. If, for whatever reason,
your administration wishes not to report such issues to FDA, PLEASE
consider reporting them to ECRI or, failing that, to me. I will
honor requests for anonymity as I have for the other two hospitals, but
understand it is vitally important for these types of events to be
reported.
If you have any questions, please feel free to contact me.
Rick Hampton
Wireless Communications Manager
Partners HealthCare System
One Constitution Center, OCC210
Charlestown, MA 02129
Office: 617-726-6633
Email: rhampton@partners.org
UPDATE: I confirmed with Rick that all of the above interference
problems were with channelized WMTS, not the frequency hopping (GE) or
DECT-based (Philips) “smart-hopping” WMTS implementations more recently
available. It's also worth noting that DTV produces “side lobe
interference” where signal from channels 36 and 38 bleed over into the
WMTS band — all allowable by the FCC. As a result, markets with both
TV channels face additional pressure in implementing more than a modest
number of fixed channels in WMTS.

