box10.gif (1299 bytes)







Inside the October 2001 print edition of
Canadian Healthcare Technology:

Feature Report: New developments in telehealth

High-speed tele-homecare trial launched in Maritimes

Several partners have joined forces to test a new telehealth system in the residences of 70 Maritime-based home-care clients. It’s among the largest home telehealth project launched in Canada, and the first to make use of high-bandwidth, Web-enabled videoconferencing.


Integration engines

Saskatchewan District Health has effectively connected 20 different computer systems, integrating acute care hospitals and long-term care centres. An interface engine is used, so that users can share clinical and administrative information.


PACS projects

Implementing a Picture Archiving and Communication System is a large-scale project that can cause disruptions until the bugs are worked out. Once implemented, however, a PACS can deliver major productivity improvements.


Medication errors

It would be extremely useful to physicians if developers of electronic patient records would include software that automatically checks the medications prescribed to patients during an encounter.


NORTHern exposure

The NORTH network, a telehealth system that connects medical specialists with rural delivery centres, is poised for expansion. There are major changes planned on the technological front, as well, as the network shifts from ISDN and Switch 56 to an IP-based system.

PLUS news stories, analysis, and features and more.


High-speed tele-homecare trial launched in Maritimes

By Jerry Zeidenberg

Several partners have joined forces to test a new telehealth system in the residences of 70 Maritime-based home-care clients. It’s among the largest home telehealth project launched in Canada, and the first to make use of high-bandwidth, Web-enabled videoconferencing.

The partners call the system ‘user friendly’, as home-care clients have just two buttons to control – on and off. There’s a camera with a sleeve cover that flips down for privacy. And familiar medical instruments, such as a spirometer and electronic stethoscope, connect to the system using a wireless link.

The five-month-long pilot project is taking place in Halifax and Moncton. It will allow nurses at central offices to make ‘virtual’ visits to the homes of clients, observing and speaking with patients through the videoconferencing system. They’ll also be able to remotely control the medical instruments while patients are using them.

The project involves We Care Health Services Inc., a privately run home care company with 55 franchises across Canada, March Networks Inc., a multimedia technology company with broadband solutions for home care and long-term care, Aliant Telecom, a telecommunications provider, and CANARIE, Inc., the Ottawa-based Internet and data communications development agency.

As well, the project will be independently assessed by Dr. Richard Scott, an associate professor with the Health Telematics Unit at the University of Calgary. Dr. Scott has a background in the evaluation of telehealth projects, and recently worked with the VITAL home telehealth project in New Brunswick.

The corporate partners believe the technology can dramatically improve the quality of care for clients and lower costs for home-care providers.

“We think the technology won’t replace regular visits by nurses, instead, it will augment those visits,” said Bob Webster, general manager of March Networks, in an interview with Canadian Healthcare Technology. He explained that the 70 participants in the pilot project will continue to receive their regular visits from a nurse, but will also participate in two video visits each week.

As a result, they will be in closer communication with their care-givers.

What’s more, they will have the ability to videoconference with the nurses on demand, if they have an emergency situation.

A nurse could recommend immediate medical attention in such cases – such as a trip to an emergency department. Alternatively, the nurse might determine that a trip to hospital is unnecessary in certain situations, thereby saving time and trouble for the client and medical centers.

For home care agencies, it’s believed that telehealth can produce high-quality care at lower cost. “It’s well known that nursing shortages are a problem across Canada, and in most parts of the world,” said Webster. Videoconferencing technology, in theory, enables agencies to provide a greater number of visits with the nurses they have on staff.

Moreover, nurses will spend less time traveling in their cars to see clients – an unproductive use of their time. Instead, they can communicate with patients through the videoconferencing system.

This is especially useful for geographically remote clients, such as those in rural locations, who are difficult to visit.

We Care president John Schram said the technology is the next step in efficient home care, and estimated that telehealth could cut the cost of delivering home care by as much as 50 percent.

March Networks designed the equipment and is investing some $750,000 in the pilot project to demonstrate its effectiveness. CANARIE has chipped in another $250,000.

“We believe there’s a serious business case for this technology,” said Webster. “We believe that demand for it will take off over the next few years.”

Indeed, Webster said March Networks is gearing up to start commercial sales of the system next summer, in both Canada and the United States. Once the units are produced in volume quantities, Webster says March Networks will be able to sell them for about $2,000 to $2,500 each.

According to Webster, the systems are “network agnostic”, meaning they work with any type of high-speed network – including cable, DSL, and others.

The company is using an ASP model to bring down the cost for home care agencies. Instead of selling home care providers a server with software, March Networks is running its own servers and software from a central location.

Home care agencies tap into the server and use it connect with their own customers using a secure, virtual private network.

Also part of the system is software that logs remote visits, and helps the home care agency track and manage client care, staff and business issues. “We know that home care providers have a huge whack of paperwork, and we’re trying to help them with this,” said Webster. He said the software automates a good deal of record keeping, and that medical data – including audio and video clips – are kept on file for reference purposes. “Doctors or nurses may want to go back to it,” said Webster. “We also keep files from the medical instruments, so that you can establish trends for patients.”

When it comes to the equipment going into the homes of clients, Webster said March Networks has tried to make it simple to use and unobtrusive. Instead of a computer and monitor, a gateway box attaches to the client’s TV set. A small camera sits atop the gateway unit, with a shutter that can be flipped open or shut at the user’s discretion.

A box of instruments – including spirometer, blood pressure monitor and electronic stethoscope – is positioned nearby, but there are no cables running to the TV or gateway box. Instead, the medical instruments use a wireless connection to the system.

“The fewer the wires, the better,” said Webster. “People don’t want visitors coming in and seeing a whole lot of cables and equipment lying about. If they do, they assume the person is really in poor shape, which may not be the case.

“By keeping it simple, we feel we will gain more user acceptance,” said Webster.

The camera runs at between 15 and 30 frames per second, enough for real-time-video without any noticeable delays or latency, said Webster. To operate the system, the user needs only to open a sleeve on the camera, and press a green button on the control unit. A red button shuts things off.

The 70 participants in the trial all have chronic illnesses, such as respiratory problems, cardiac illness or cancer. Most will be 55 to 75, which is the typical age range for home care recipients.

For its part, We Care is proactively investing in technological solutions to improve the delivery of home care services. The company is currently investing over $1 million on various information technology applications.



Integration engines enable hospital networks to use best solutions

By Dianne Daniel

Integration is the buzzword as hospitals move forward with plans to share electronic information externally with other medical centres, physician offices, pharmacies and other community-based agencies.

But when embarking on efforts to establish a smooth flow of data from one end of the spectrum to the other, those involved should treat integration as an ongoing process versus a standalone project, cautions Guy Paterson, director of IT and telecommunications for Saskatoon District Health (SDH).

“Integration is not a one-time thing; it’s not a project that starts and ends,” notes Paterson. “You do it and then it’s ongoing maintenance ... How are the changes that are occurring in the organization handled as you’re getting a more complex, integrated information management system out there?”

Since the early 1990s, SDH has chosen to make integration a core competency of its IT department. Today, its underlying infrastructure supports the processing of roughly 100,000 messages each day with nearly 100 per cent data accuracy. Information is sent and received by more than 20 different computer systems within the district’s acute-care and long-term-care facilities, but as that expands to include the community, every medical clinic, pharmacy and doctor’s office becomes a potential customer as well.

To manage the complexity, the district has two full-time IT staff devoted to integration. Monitoring tools are constantly running to alert them when an interface process fails, via pager or e-mail, and testing is done continually.

As SDH manager of IT Perry Kjargaard puts it: “With the complexity of this web we’ve woven, and the demand from users, there’s literally changes requested all of the time. We employ people to manage that process.”

For example, if a hospital opens a new unit, the information about that unit has to be propagated throughout the district. “Suddenly you’re admitting a patient to a bed that didn’t exist before and the receiving departments need to understand that there’s a new bed there,” explains Paterson. “You have to really look at the information flow from end to end, from creation to final use – and end use can be in multiple locations and for multiple purposes.”

While the IT department did a lot of initial work to minimize the amount of data manipulation required to share information among various systems, the development of a data standards council and use of an integrated test plan have also been key to a smooth integration process.

The initial goal was to get as close to Health Level 7 (HL7) compliance as possible, the electronic data interchange protocol that outlines how information is exchanged between systems. The approach at SDH is to allow each business unit to retain control over what Paterson refers to as the transactional layer, the part that deals with selecting best of breed software for the task at hand, whether it’s registering patients, ordering lab tests, performing lab tests or reporting results. This enables departments to choose the most cost-effective and efficient applications.

It then falls to the IT group to support the extraction of information from those systems via an interface engine.

“Our philosophy is to find the best information system to support the needs of those business units, regardless what they are,” he says. “We then look at the extraction of that data – in most cases using an integration engine to put it into some sort of repository or electronic health record,” he says

That philosophy is echoed at Toronto-based electronic Child Health Network (eCHN), a partnership among The Hospital for Sick Children, the government of Ontario, Canada and several member organizations.

ECHN aims to electronically link hospitals, local pediatricians, home care agencies and others that provide children’s health services.

“One of the fundamental requirements we had was we didn’t want to require any of the hospitals to replace any of the information systems that they already had in operation,” notes eCHN chief executive officer Andrew Szende. “We wanted everyone to retain their autonomy and to retain the choice of buying the best of breed information systems, the ones that they wanted.”

A critical service provided by the eCHN is the Health Information Network or HiNet, a system that enables healthcare providers to share common data in the form of an electronic health record.

The information shared includes lab results, dictated summaries and images, all of which is intended to provide a continuum of care, linking home care to community agency to regional pediatric centre to a children’s hospital.

Like Saskatoon District Health, the eCHN relies on HL7 messaging standards to ensure smooth data transfers.

While member hospitals usually have an interface engine in place to make outgoing data compliant with the standard, the eCHN uses the Health Data Network (HDN) from IBM Canada Ltd. on top of that to pull the information together into one integrated view.

According to Dr. John Edmonds, eCHN’s network specialist, once the machines on either end understand what it is they’re sending and receiving, the next step is to normalize the data.

For example, hospitals may use different names for the same lab test. When the information arrives at HiNet, the IBM HDN software needs to know they refer to the same thing so that the information gets put into the right `slot’, he says.

Another layer of sophistication is required in the software to ensure the reference range for the lab test is displayed correctly, he adds. “The reference range is different for different ages of kids. The software needs to know how old the child is, what am I going to display as normal now and what was normal when the test was taken.”

Like SDH, the electronic Child Health Network has a team devoted to managing change, called its MED team for Medical Entities Dictionary. The group continually looks at lab tests coming in from different places and has discussions about the best way to handle the information. “It’s not something that’s static,” notes Edmonds.

In operation since October, 2000, HiNet allows users to call up an integrated view of a child’s health record that indicates the most recent test results as well as the patient history.

Members of the network retain ownership and control of their own patient records, but a copy is mirrored in HiNet in real-time so that the most current information is always available to everyone – without the risk of one member making an unwanted change to another member’s record, says Szende.

Current eCHN members include Toronto-based St. Joseph’s Health Centre, Saint Elizabeth Health Care (a home care provider), Orillia Soldiers’ Memorial Hospital, Rouge Valley Health System, Bloorview MacMillan Children’s Centre, of Toronto, and The Credit Valley Hospital, in Mississauga, Ont., as well as roughly 20 office-based physicians.

Information is transmitted via private T1 or leased fibre lines for privacy and security, although Dr. Edmonds is quick to point out the electronic records are much more secure than the paper versions shared normally.

“At the moment a health record department will send out a paper chart, but once we send it by fax or mail, we have no control over how it’s distributed,” he says. “With the electronic version we can keep track of who looks at it.”



Complete commitment, strategic plan key to PACS implementations

By Issie Rabinovitch, PhD

When General Electric Medical Systems Information Technologies recently hosted a one-day press event at their headquarters in Milwaukee, I decided to attend. As GEMS racks up US$9 billion in sales annually and invests US$900 million in R&D per year, it cannot be ignored. I’m happy to report that the mini-conference exceeded my expectations.

The day was fast paced, with numerous presentations by GEMS executives, partners, and customers. The quality was uniformly high, with a few exceptions, but the best presentation (in my opinion) was entitled “Essentials for a successful enterprise PACS implementation,” by Neil D.Johnson, M.D. He’s the Chief Medical Advisor, Clinical Informatics, at Children’s Hospital Center, Cincinnati.

What distinguished this PACS presentation from others I have attended was the experience of the presenter. Dr. Johnson has been involved with his institution’s PACS project since the earliest planning stages, and is still involved day-to-day more than a year after full implementation.

A native of Australia, Dr. Johnson was both a pediatric radiologist in that country as well as an entrepreneur in healthcare technology before moving to the United States. He has authored numerous peer-reviewed publications and made presentations around the world. He is a member of many professional associations, and serves on several national committees. His main clinical focus is vascular and interventional radiology.

Over a decade ago, he brought this unique blend of experience to his job at Cincinnati Children’s, a leading U. S. pediatric institution. It is a 350-bed hospital with seven linked outpatient centers about 20 miles or more away. It performs a total of 148,000 radiology exams and has 375,000 outpatient visits per year.

Dr. Johnson led the center’s development of a PACS (picture archiving and communications system) to replace film used in radiological procedures. The hospital began planning its PACS in 1998, and the undertaking involved 50 people from numerous functional areas.

The RFP stage was reached in January 1999, vendor selection in July, contracting in August, and implementation commenced in December of that year. He told the team that “this is the biggest thing we will ever do”, and in retrospect, he doesn’t think he overstated the case. The PACS project was a major disruption in everyone’s life, and actually lead to a revolt at one stage, which Dr. Johnson dealt with successfully.

The PACS at Cincinnati Children’s replaced a system of radiology film transportation between facilities by truck. It has helped the hospital approach its ideal of “any film, anywhere, anytime”. According to Dr. Johnson, the key to success included a strategic plan, a complete commitment to a filmless environment, and careful selection of vendors for what was assumed to be a 15 or 20-year marriage.

Changing vendors sooner than 15 years, according to Dr. Johnson, is a prospect too horrible to contemplate. Therefore, the PACS supplier must be a stable company with long-term vision and commitment, substantial resources for development, and an aggressive approach to upgrades. Dr. Johnson and Cincinnati Children’s chose GEMS, but there are other vendors that fit the profile. In Dr. Johnson’s view, smaller companies are more likely to disappear, and are therefore a dangerous choice.

A PACS is a BIG Information System, the emphasis being Dr. Johnson’s. It is more than just being filmless. He made the point, several times, that having the ability to pass images around (image distribution) isn’t all there is to PACS.

Whenever he reads about institutions that claim to have installed a PACS within a few weeks, he is absolutely confident that they have simply accomplished something far more limited. Installing a PACS takes more than weeks or months.

At Cincinnati Children’s, 90 percent of reports are signed off within 24 hours. If a report is marked as urgent, it is handled within one hour. Some film is still used, but every image is available in digital form.

There are some radiology procedures at outpatient clinics that require a radiologist to be in attendance. Until PACS, these physicians would only read a few pictures per day, but now they can be far more productive, since they can read pictures that were taken anywhere. Wherever the doctor is, including at home, he or she can read the film. The PACS solution provides after hours service, uniform service, any place, any time.

There is a national shortage of pediatric radiologists, with over 100 advertised positions currently unfilled. Dr. Johnson mentioned the case of a city in the U.S., comparable in population to Cincinnati, that was about to lose all of its pediatric radiologists in a labour dispute. The hospitals in that city were considering outsourcing the work to Cincinnati Children’s, whose efficient PACS gives them the capacity to handle this extra work.

Dr. Johnson also had some valuable tips on working with consultants. He admitted that his experiences in Australia with consultants had been less than successful, but his experiences in the U.S. were decidedly more favourable.

According to Dr. Johnson, consultants can be very valuable in implementing a PACS, especially if they are neither part time nor beginners, if they have national experience and are part of a team with multiple skills. Dr. Johnson has found that consultants are most valuable for project organization and discipline.

Issie Rabinovitch, PhD, is a Toronto-based computer consultant.



Doctors’ offices would benefit from automatic drug checking

By Dr. Alan Brookstone

Canada’s aging population is creating a poly-pharmacy nightmare for physicians across a range of clinical specialties. One of my greatest frustrations in family practice is my inability to easily detect or predict dangerous drug interactions particularly in my elderly patients who are on multiple medications.

The number of pharmaceutical products on the market makes it virtually impossible to predict serious drug interactions given the constraints of time and the limited information that is available at the point-of-care.

The average consultation time for an office visit is 10 minutes, and even though tools such as Palm-based ePocrates qRx 4.0 provide drug interaction checking capability, using my Palm during the office visit requires that I have to double-task.

During an average consultation I have to take a history, examine my patient, develop a diagnosis and treatment plan and write a prescription. As a result, I have little time available to perform a drug interaction check to look for potential side effects.

To be useful, this information should be automatically and transparently presented to me as part of the patient encounter.

The patient medication profile needs to be resident in my clinical record system without the need to re-input any information other than a change in dosage or the addition of a new medication at the time of the clinical interaction. Anything less is too cumbersome.

As a result, I depend to a large degree on my clinical knowledge, my intuition and the friendly local pharmacist, who calls my office all too frequently to confirm a medication dosage or warn of a potential interaction with another medication.

How do they do it? Through sophisticated drug databases that automatically prompt the pharmacist with respect to potential interactions based on a patient medication profile.

According to an August 2001 National Post news article, it is estimated that approximately 5,000 to 10,000 Canadians are dying per year due to medical errors.

This number is approximately one-tenth of that reported in the United States and is in proportion to similar studies in the UK and Australia. Electronic Medical Record systems are an important means for reducing office based prescribing errors.

In a medical office, time is the greatest limiting factor for physicians. If a tool does not integrate into the daily patient workflow, is difficult to use, is not instantaneous in response or adds to workload, it is simply not going to be used.

Physician offices currently lack the technology necessary to detect adverse drug interactions and suggest possible alternatives. As a result, we cannot take advantage of technology to attack this problem because the majority of physicians run purely paper-based medical offices.

Standalone tools do not work, as they require a change in workflow in order to be adopted by physicians and provide limited end user benefit. Therefore, I strongly believe that adoption of a physician-based electronic record management system will require a more complete solution than that which is currently available.

Implementation will also require considerable workflow re-engineering in order to be acceptable to the physician user.

Features will need to include immediate access to clinical data and flexibility in the mechanism of data entry including keyboard, voice recognition and handwriting recognition. I strongly believe that physicians require a solution that applies to every aspect of daily workflow.

Dr. Alan Brookstone is a physician based in Richmond, B.C.