Linking Medicine and Information Technology

 

 

• RSS
• Print

Mark Smith, MD, department chair in emergency medicine at Georgetown University Medical Center (GUMC), stands at the crossroads of a number of seemingly unrelated health care initiatives.

An expert in the areas of medical informatics, disaster medicine, simulation training, complexity theory and systems medicine, Smith melds his more than 30 years of experience as a physician with his ability to design systems for keeping accurate patient medical records.

In addition to his academic role in the School of Medicine, Smith serves as the chair of MedStar Emergency Physicians, which provides emergency medicine physician leadership and staffing at four of the seven MedStar hospitals, including Georgetown University Hospital and Washington Hospital Center.

“One of the reasons I find emergency medicine so interesting is that an emergency department is a fractal of a whole hospital,” Smith says. “People enter the ER, are registered, undergo an initial evaluation, receive a series of diagnostic tests, are administered treatment and then go through a discharge process. It’s a microcosm of a whole hospital admission process, only compressed in time.”

Smith’s prowess with emergency medicine alone is a valuable asset for the medical school’s curriculum, says Stephen Ray Mitchell, MD, dean for medical education at the School of Medicine.

“(He) is really unique in many ways – kind of a 'Renaissance man'. He’s a big picture guy,” he says. “He envisioned the information system at Washington Hospital Center. They actually hired him to make the emergency department run better, and he had a student walking behind every treating doctor asking, ‘What is it that you need right now?’ So when they programmed the system, he was able to say, ‘At this point we need to pull in the blood test, and here we need to go out and get the X-ray.’ ”

All fourth-year medical students participate in a four-week experience in emergency medicine at Georgetown. They receive certification in advanced cardiac life support and clinical emergency department experience.

“Georgetown was one of the first medical schools to require every single student to do emergency medicine – long before ‘ER’ was on TV, so it’s been in our fabric, and Mark has made it an excellent program,” Mitchell says.

Smith, who joined Georgetown in 2000, regards emergency medicine as a valuable lens through which to test innovations in health care. “If it works in the emergency department, it will probably work anywhere,” he says.

Data-Centric Technology

System design is the process of defining and developing a system to satisfy specified data processing needs. It is all about managing complexity, and health care is about as complex as it gets, says Smith. His amalgam of expertise provides him with a distinct perspective on health care reform, particularly expectations centered on the electronic patient record.

Integrating information from a hodgepodge of disparate clinical data sources offers significant potential for reducing errors, inefficiencies and costs, he says.

Currently, medical personnel operate in an environment where information that is needed for clinical decision-making may not be immediately available, Smith explains.

An electronic medical record should provide current, accurate, clinically relevant patient information, in effect making the caregiver aware of everything that is happening to a patient, but designing a universal network of systems to capture, store and deliver that information has proven elusive.

“The electronic medical record is not an end in itself,” Smith cautions. “The end is better management of health and illness. For that we need information presented in such a way that it best informs patient care decisions. The challenge is to integrate all information about a person’s health, disease, and episodic illness across the complete continuum of care.”

He says information has to be available in a way that it’s going to make a difference in care. “Something I read years ago in an article in the Wall Street Journal has stayed with me – ‘In the absence of action, information is overhead.’”

Smith realized early on the importance of integrating information science with medicine. He received a master’s degree in computer science from Stanford University before graduating with his medical degree from Yale University School of Medicine.

“I love the intellectual clarity and complexity of mathematics and computer science, but I wanted my life’s work to involve serving people and to have the texture of a human endeavor,” he says. “That is why I left computer science to become a physician.”

But he never lost his interest in the display and use of information.

“From his years of clinical experience, he knows the practical limitations of a range of theoretical approaches and can guide the development of systems that will have true clinical utility for improved patient care,” says Howard Federoff, MD, PhD, executive vice president for health sciences and executive dean of the School of Medicine.

For decades Smith has believed the most potent force for improving medicine was going to be the application of information technology in the service of hospitals’ care mission. “Optimum information systems would enable care that is safer, faster, cheaper, and better. That has been my driving vision,” he says.

And so was born, Azyxxi, an integrated clinical information system currently in use at Georgetown University Hospital and recently acquired by Microsoft. Smith co-developed the system, which provides a real-time, comprehensive picture of a patient’s health care history.

Physicians and staff providing care for a particular patient have immediate access to current, accurate, critically important information from a multitude of disparate data sources and data types, including laboratory test results, past medical history, X-ray images, full motion angiograms, ultrasound pictures, electrocardiograms, demographic information, scanned documents and more.

Making Data Count

One of the unique aspects of the Azyxxi information system that Smith designed is that it works equally well in providing views across groups of patients as it does in diving down into an individual patient. So it can be used to drive quality improvement and clinical research across an organization.

Smith points out the challenge of designing solutions that can adapt to the unforeseeable needs of a rapidly evolving scientific knowledge base. His general philosophy – learned the hard way after once discarding a piece of “unimportant” data that he wound up needing – is to collect everything.

“You never know what’s going to turn out to be important. Digital storage is cheap, so as long as you aren’t delaying treatment in an emergency situation, go ahead and collect it,” Smith says.

When it comes to using technology to drive safer care, Smith points out that his 25 plus years of running emergency departments has taught him that mistakes occur in medicine because some basic lessons taught in the first year of medical school are not heeded. A change in vital signs – a rising pulse, a falling blood pressure – that signals deterioration in the state of a patient is somehow not always paid attention to by the clinician.

“Well-designed information systems can help correct that problem,” Smith says, “by providing the clinician with ambient information about the state of the patient.”

The electronic medical record and other advances in medicine will only work if the country can find solutions to the current health care crisis, he says.

“We have some major issues with how health care in this country is organized and delivered that are going to have to be dealt with and fixed in order for any new technology to be implemented and work well,” Smith explains. “So we’re not going to be able to succeed with a purely scientific and technological approach without also looking at the way health care is funded, delivered (and) paid for.”

By Frank Reider, GUMC Communications, excerpted from the February 9 - February 22, 2009 issue of the Blue and Gray