The Computerworld Honors Program
Honoring those who use Information Technology to benefit society
Final Copy of Case Study
Boston, MA, US



Non-Profit Organizations

Technology Area:
Videoconferencing and Image Sharing Technology

Massachusetts General Hospital


MGH TeleStroke Program

Introductory Overview
Every 40 seconds, someone in the United States suffers a new or recurrent stroke.  Every 3 to 4 minutes, someone in the US dies of a stroke.  Stroke afflicts more than 795,000 Americans per year, and it kills more than 143,000 people a year, making it the third leading cause of death behind heart disease and cancer.  In the United States alone, stroke-related medical and disability costs are estimated to be $68.9 billion in 2009.  As impressive as these numbers might seem, the global impact of stroke from a financial and public health standpoint are even more staggering.

Stroke is a form of cardiovascular disease that affects the arteries leading to and within the brain. When a blood vessel that carries oxygen and nutrients to the brain is either blocked by a clot or bursts, a stroke occurs. Part of the brain cannot get the blood (and oxygen) it needs, so brain cells start to die. A stroke is caused either by a clot which obstructs blood flow to the brain (an ischemic stroke), or by a blood vessel rupture that prevents blood flow to the brain (a hemorrhagic stroke).

The public health impact and long-term costs associated with ischemic strokes, which account for 83% of all strokes, can be significantly reduced with rapid assessment and successful treatment with clot-busting drugs like tPA.  This drug must be administered within a three-hour window from the onset of symptoms to work best, so rapid recognition and accurate diagnosis is critical for optimizing outcomes in patients.

Although several bedside tools have been developed to improve the accuracy of rapid diagnosis, this can be challenging for physicians without neurological expertise.  The misdiagnosis rate among primary care and emergency medicine physicians is estimated to be 30%.  Delays in diagnosis, misdiagnosis or complete failure to diagnose acute stroke are driven primarily by the lack of stroke specialists in community hospital emergency departments.  As a result, nationwide only 3 to 5 percent of those who suffer an ischemic stroke are treated with this medication.  A substantial impact on public health would be realized if systems were in place to facilitate more rapid and accurate diagnosis as well as urgent treatment.

The core steps of any acute stroke clinical intervention include a rapid neurological assessment, a review of brain imaging scans, and the assessment of patient eligibility for tPA therapy or advanced surgical interventions.  Telemedicine technology facilitates each of these steps, and helps specialists and hospitals overcome geographic and resource barriers.  It provides specialists with the data they need to urgently assist physicians in distant or under-resourced facilities to make difficult clinical decisions.

Stroke telemedicine - or telestroke - has proven to be effective in remotely performing acute consultative services, delivering neurologic assessments, and directing the treatment of stroke patients.  Exceptional telestroke programs merge the videoconferencing, teleradiology, and Internet technologies to enable multiple highly skilled specialists to collaboratively examine a patient in real-time.  After diagnosis and treatment, some patients remain at the community "spoke" hospital for ongoing care, while others are transferred to the tertiary "hub" hospital facility for more complex and invasive therapies.

Massachusetts General Hospital, founded in 1811 as a non-profit teaching hospital affiliated with the Harvard Medical School, is a recognized leader in the field of stroke telemedicine.  Over the last ten years, it has built a self-sustaining regional telemedicine network of 24 community hospitals focused on providing the right stroke care to the right patients at the right time.  During this period, MGH physicians have performed hundreds of telemedicine consults, saved many lives, and improved outcomes for countless stroke patients.

The Importance of Technology
How did the technology you used contribute to this project and why was it important?
As stated above, the MGH TeleStroke Program relies heavily on the convergence of three mainstream technologies: videoconferencing, teleradiology, and web-enabled relational database software applications.  Together, these technologies can facilitate the timely clinical evaluation of stroke patients.  Without these technologies, emergency department physicians in distant or under-resourced facilities are often unable, or unwilling due to the grave risks associated with tPA, to make difficult clinical decisions regarding stroke patients, which results in unnecessary death or disability.

The MGH videoconferencing (VC) infrastructure consists of a central multipoint control unit (MCU), a Polycom MGC100 and a Polycom gatekeeper running on the corporate IP network. The MCU also acts as the IP VC gateway that allows outside VC systems to connect with internal endpoints.  Most of the MGH endpoints are protected by the corporate firewall, while the MCU resides on the outside in order to provide the gateway functionality. The MCU, while providing IP-based VC services, also supports ISDN connectivity, with approximately 6 Mb ISDN bandwidth.  The MGH fleet of videoconferencing endpoints numbers around 180 units, mostly running across the IP network. These endpoints are located throughout the MGH in conference rooms, offices, auditoriums and clinical areas. They consist mostly of Polycom VSX systems.  During a typical month, the entire fleet of endpoints within the MGH will engage in an average of 1500 videoconference calls. The MGH infrastructure also provides for digital recording as well as on demand streaming video.

The Stroke Telemedicine program currently uses mostly Polycom VSX endpoints with multipoint capability and ISDN / IP connectivity.  Stroke physician consultants from the hub connect to the community hospital unit from within the MGH using a VSX endpoint, or connect from home using a laptop or PC using Polycom PVX software and a webcam. The H.323-compliant software offers the stroke consultant full pan/tilt/zoom (PTZ) far-end camera control, which is critical for performing the examination of the patient.

From a teleradiology standpoint, the spoke hospital transmits a DICOM compatible digital image of the radiology scan (CT and MRI) which typically ranges in size from 10-100 MB via a secure LAN-to-LAN VPN tunnel.  The image is routed to a dedicated Telestroke server and is available to be viewed by the hub based stroke consultant via a web-based image viewer provided by Amicas.  The Amicas application is a Java-based software product that uses standard Web browsers (Netscape Communicator and Internet Explorer) to access and view images at flexible resolution levels.  The Amicas software accommodates HL7, SQL, or HTTP requests to bring reports to the viewer, and a Java Bean or Active X control allows the viewer to be launched from inside many other applications.

The third and final technology that binds these other components together is an internally developed, robustly designed SQL server database with a secure GUI interface that can be accessed via any internet connection.  In addition to serving as a intranet site, this encrypted, password-protected database is a medical record system that accurately documents critical information regarding each clinical interaction, including patient demographics, medical history, and lab values among other things.  The application has embedded in it reminders, warnings, algorithms, and calculators which assist the stroke consultant in making the correct diagnosis and treatment decision.  The Amicas image viewer and radiology images can be accessed directly from the application.  Once the consultation note is digitally signed, the application sends emails to various program staff triggering numerous administrative and financial functions.  Finally, spoke hospital staff with access are able to log on to the application and perform queries on their data and print reports.

Has your project helped those it was designed to help?  

Has your project fundamentally changed how tasks are performed?  

What new advantage or opportunity does your project provide to people?
High quality stroke care is about teamwork: bringing physicians, patients, family and technology together.  MGH's goal is to improve the standard of stroke care and what raise the bar on what is expected in small community hospitals.  The role we play in the community hospital setting is to augment what they already have, not replace it.  We deploy simple, easy to use videoconferencing technologies, allowing us to spend more time focused on improving the process of care.  Our first step in implementing a program at a community hospital is to review their stroke treatment protocols and update them to incorporate the most recent advances in evidence-based medicine.  Once the videoconferencing technology has been purchased and installed in the community hospital, we then go through a rigorous three step training process of talk-throughs, walk-throughs, and mock drills with live patients to ensure that all community hospital emergency room staff are comfortable with the stroke care protocols, the videoconferencing technology, and the administration of tPA medication.  Once implemented, we have a series of maintenance protocols established to ensure that the equipment and various connections are functioning at all a times.

Between 2004 and 2008, MGH stroke physicians performed almost 400 stroke telemedicine consultations, and provided tPA therapy to over 130 patients, for a treatment rate of 33%, which is well above the 3-5% national average.  During the same period of time, we provided 2,300 telephone-only consultations, and treated those patients with tPA 13% of the time.  This telemedicine intervention has saved many lives, and improved outcomes for countless stroke patients.

If possible, include an example of how the project has benefited a specific individual, enterprise or organization. Please include personal quotes from individuals who have directly benefited from your work.
The MGH Telestroke Program has been extremely successful in helping community hospitals and their communities bring better stroke care to patients. 

According to Ursula Crowell, RN who serves as the Outcomes Manager at SouthCoast Hospitals Group in southeastern Massachusetts, "the (MGH) Telestroke Program has helped SouthCoast support its goal and mission of improving the lives of the communities that we serve. Our increased communication with the neurology specialists through the Telestroke Program has helped us to streamline our processes related to stroke care."

Nancy Edwards, MD from Tobey Hospital in Wareham, MA reported that in the first year of the MGH Telestroke Program they successfully administered tPA ten times, compared to four times in the previous four years.  In her words, "the Telestroke Program is like bringing a neurologist to our department, to the bedside."

Beverly Sylvia was a 51 year-old patient who experienced stroke symptoms and went to her local hospital emergency room which is 50 miles away from Boston.  Using the MGH Telestroke Program, she was evaluated by an MGH physician and determined to be a candidate for tPA treatment.  She received the medication, and recovered completely with no disabilities.  "I was fortunate that I ended up at the right hospital." Sylvia said.  "I know what my father went through.  He was paralyzed on his right side and could not talk right.  Not something you want to have happen to you."

John Looney, Director of Public Relations at Jordan Hospital in Plymouth, MA, commented, "The (MGH) Telestroke Program really has been a great story to tell.  Not only does it directly benefit our patients, but it has really helped to improve our image in the community.  When you speak about affiliations, patients (and) residents of the community can really identify with this.  The expenses associated with this affiliation are very reasonable  and when you compare this to the benefits  you cannot put a price tag on that."

Is it the first, the only, the best or the most effective application of its kind?   First

What are the exceptional aspects of your project?
The MGH Acute Stroke Telemedicine Program, first conceived by Dr. Lee Schwamm in 1996 and piloted in 1998 at Marthas Vineyard Hospital, is arguably the first known program of its kind in the United States.   Currently supporting 24 hospitals across three states, the MGH Telestroke Program is now the largest single hub and spoke telestroke network in the world, and we believe the only financially self-sustaining program in existence in the US.

From 2000 to 2005, the MGH program grew to a network of five community hospitals.  Since 2005, when the Massachusetts Department of Public Health created a Primary Stroke Service certification process, the MGH Telestroke Program has grown at a more rapid pace, adding on average five hospitals per year to its network.  In 2007, the Program entered the bordering states of New Hampshire and Maine, where it now supports six community hospitals.

Leading up to 2003, when the American Stroke Association advocated the use of telemedicine to support a stroke system of care approach, the number of centers initiating telemedicine programs to support rapid evaluation of stroke patients in the US and abroad has grown slowly.  Most programs in the United States are small, hospital-based programs that were developed by an academic medical center or integrated health network.  Besides the MGH program, the most notable are the REACH program developed in 2003 by the Medical College of Georgia (MCG) that supports ten community hospitals, and the state-led telestroke initiative in NY established in 2006 by the state Department of Health to develop a statewide system linking hospitals via telemedicine for acute telestroke consultations.  In this instance, the state government has provided start-up funding and direction to the initiative, and policymakers view the telestroke effort as an initial phase in establishing access to comprehensive telemedicine services for several medical specialties.

In a clear sign of growing acceptance in the US, several for-profit start-up companies have entered the stroke telemedicine market since 2003.  Specialists on Call supports 40 hospitals in a range of medical specialties, including stroke.  The REACH program at MCG was spun off in 2006 as a for-profit entity that markets itself as a turnkey application services provider (ASP).

In Europe, only a handful of countries have well-developed stroke telemedicine programs.  By the end of 2007, there were approximately a dozen telestroke programs operating in Europe in Germany (7), France (2), Finland (1) and the United Kingdom (2).  In aggregate, the 12 existing European telestroke programs reviewed include 18 hub and 77 spoke hospitals.

The MGH Telestroke Program is unique and exceptional in many ways, but two critical differences that distinguish us from other programs are worth noting.  First, we do not sell a hardware solution, but instead advocate that our spoke hospitals use low-cost, non-proprietary videoconferencing hardware and software from established vendors that utilize the latest industry standard codecs.  In doing so, we are not required to spend large sums of money on product research and development, and we ensure that all our various devices communicate effectively with very little technical support.  Second, our program maintains its academic roots, supporting the vital mission of training current and future generations of clinicians to treat future stroke patients.

These two points help to illustrate a key difference between our program and all others: Unlike other programs, our system architecture permits multipoint videoconferencing.  Using this invaluable feature, other medical and surgical specialists can join in and collaborate on any VC call as needed to provide optimal care to the patient, and medical trainees and providers can participate in any call for training and educational purposes.

What were the most important obstacles that had to be overcome in order for your work to be successful? Technical problems? Resources? Expertise? Organizational problems?
The MGH Telestroke Program faced many challenges during its implementation.  They included infrastructure funding, reimbursement for services, physician licensure and credentialing, medical liability, and compliance with privacy and security laws.

Infrastructure funding:
The development of any telemedicine program requires capital investment in infrastructure, including the purchase and maintenance of computer hardware and related software, as well as a secure means of transmitting data compliant with current federal privacy standards.

The MGH Telestroke Program approach is to minimize infrastructure costs at the community hospital level, where financial and human resources are most scarce.  We place the more sophisticated, expensive technologies (e.g. fixed videoconference stations, firewall traversal devices, videoconferencing bridges, image storage servers) at the hub site, and place the simpler, less costly systems (mobile videoconferencing units) at the spoke sites where financial resources, technical support, and training capabilities are limited.  By utilizing low cost, open-source, interoperable videoconferencing devices manufactured by established vendors, the program is able to minimize costs compared to other programs that offer proprietary technologies.

MGH Telestroke Program hospitals typically finance their infrastructure through a combination of capital funds, operating funds, philanthropy, and grants.  One community hospital in our network hosts an annual golf tournament called Strokes for Strokes which raises enough money to sustain the program.  MGH also assists hospitals to identify grant funding from state and federal government agencies (HRSA, FCC, and the USDA) as well as foundations and other sources. 

Technical Challenges:
Problems with the technology, including non-connecting or malfunctioning devices, can be major barriers to successful telemedicine programs.  They result in distrust by users and low levels of satisfaction.  The ability to access adequate bandwidth to transmit large amounts of data quickly, accurately and securely is also a big challenge for remote community hospitals.  Many rural areas do not have access to the consistent low-latency, high-speed bandwidth required to support reliable, high-quality video transmission and reception over open, standards-compliant networks.  Also, DICOM compatible digital radiology scans (CTs and MRIs) ranging in size from 10-100 MB in size must be sent to the hub hospital rapidly for viewing by the stroke expert.

Until recently, the MGH Telestroke Program has encouraged community hospitals in rural areas to utilize dedicated ISDN lines for the transmission of high quality video and audio data. While these lines are more costly to install and maintain, they allow for high reliability and better performance.  Now that reliable public broadband networks are beginning to reach rural areas, we have begun to switch community hospitals in our network over to IP-based technology.  Our strategy of using devices manufactured by large vendors that support both IP and ISDN based communication modes has eased this transition and allowed our clients to switch over to IP connections with little or no additional investment or change in their process.

Compliance with Privacy and Security Laws:
The Health Insurance Portability and Accountability Act of 1996 is a federal law that regulates the sharing of patients protected health information amongst health care providers.  This type of information exchange is essential to the routine operation of telemedicine programs. 

The MGH program has worked extensively with equipment vendors, communications network providers, and internal IT staff to ensure the security of the live transmissions between hospitals, the security of data storage associated with the consultations, and the technology networks that consulting physicians may use to access this information remotely.  Our database application is heavily encrypted and secured, and maintains detailed audit logs of all activity by user.  Also, our executed contracts with each community hospital carefully outline all required information security standards and protocols, as well as the permitted uses of patient data by each entity.

Often the most innovative projects encounter the greatest resistance when they are originally proposed. If you had to fight for approval or funding, please provide a summary of the objections you faced and how you overcame them.
Physicians are important gatekeepers in telemedicine adoption and diffusion.  The ease of use of any telemedicine technology is vital to the success of any program.

We consider local physician endorsement to be the cornerstone of success for the MGH Telestroke Program.  Spoke hospital physicians are more likely to become enthusiastic participants in a telemedicine program if they play an active role in its implementation.  To achieve acceptance, we use physician champions at the hub and community hospital to engage stakeholders and promote the development of the program.  Also, using non-proprietary VC devices with multipoint capability, we are able to utilize them for network-wide professional medical education events.  These both help to promote physician-to-physician interaction, build trust, increase awareness, and help to change attitudes and increase utilization of the program.

This philosophy and feedback mechanism resulted in a dramatic change in our program model early on.  As mentioned earlier, to support the program we developed a robust SQL server relational database accessed via a GUI interface and secure web portal.  This database accurately documents each clinical interaction.  In the early stage of implementation, much of the data (e.g. patient demographics, medical history, critical lab values) was entered by the spoke hospital physician where the patient was.  However, based on feedback from the physicians that this interfered with their care to the patient, we transferred responsibility for data entry to the stroke consultant.  That decision enabled our software developers to make the application more dynamic. They began embedding reminders, warning screens, algorithms, and calculators into various parts of the application, which assist the stroke consultant in making the correct diagnosis and treatment decisions.  This feedback from the community physicians, and the flexibility of our application, allowed us to make significant improvements to the care process and increase adoption of our program.

Has your project achieved or exceeded its goals?  

Is it fully operational?   Yes

How do you see your project's innovation benefiting other applications, organizations, or global communities?
As a result of our success, we have recently been approached by other MGH clinical services who want to leverage the infrastructure we have deployed in these community hospital emergency rooms.  These other clinical services (trauma, obstetrics, pediatrics) are similar to Neurology in that they are often not available in smaller community hospital settings.  Because the technology we selected is non-proprietary and flexible, any specialty can use it successfully for their consultations.

In addition, several other academic medical centers, hospital networks, foundations, associations, and government agencies in the US and abroad have sought our assistance in designing and developing their own stroke telemedicine networks.  We have successfully guided the implementation of similar "hub and spoke" stroke telemedicine networks in Seattle, Washington (Swedish Medical Center) and New Haven, Connecticut (Yale-New Haven Hospital).  The Yale program was able to get their first spoke hospital up and running within two months of signing the contract.  We are in the process of assisting hospitals to implement their own telestroke programs in Oklahoma City, OK (Integris Health) and Richmond, Virginia (University of Virginia Health System/Virginia Telehealth Network).

How quickly has your targeted audience of users embraced your innovation? Or, how rapidly do you predict they will?
At its inception, we never envisioned the MGH Telestroke Program supporting as many as 24 hospitals.  Yet we have been able to expand to that level over the last nine years by maintaining a flexible hardware and software infrastructure and deploying low cost, open-source, interoperable videoconferencing endpoints and infrastructure devices manufactured by large, stable, successful vendors.

When a hospital enrolls in our program, we typically require two to six months to set them up and go live.  More hospitals continue to learn about and embrace our program.  We are currently in discussions with seven additional community hospitals that are interested in joining our New England network.

In addition, we are in discussion with several other hospitals and health networks through out the US, and with select government agencies in Europe, to provide technical assistance in the development of their own stroke telemedicine networks.

Continued adoption of the stroke telemedicine models globally will have a profound positive impact on the treatment and recovery of stroke patients, ultimately reducing the financial and public burden of the disease on society.  The MGH Telestroke Program is extremely proud to be considered a pioneer in the field, and we are honored to have been nominated for this prestigious award.

Digital/Visual Materials
The Program welcomes nominees to submit digital and visual images with their Case Study. We are currently only accepting .gif, .jpg and .xls files that are 1MB or smaller. The submission of these materials is not required; however, please note that a maximum of three files will be accepted per nominee. These files will be added to the end of your Case Study and will be labeled as "Appendix 1", "Appendix 2" or "Appendix 3." Finally, feel free to reference these images in the text of your Case Study by specifically referring to them as "Appendix 1", "Appendix 2" or "Appendix 3."

Currently Uploaded Appendices: