Imagine the possibilities if health care entered the digital age. X-rays, MRIs, prescriptions, and patient health records could all be in digital form, and could be moved electronically when and where they are needed. Teleconferencing could allow your current doctor to enter into real time discussion with a local doctor about a particular medical problem you are experiencing half a world away, but with access to all records. Remote monitoring could free up hospital beds.
Information technology has made huge inroads in how most businesses are run, but progress in health care has been slow indeed. A recent trip to my doctor confirmed this. He needed to see a recent x-ray, and I was asked to pick it up at the place where it was taken, and bring it with me for the appointment. At the appointment, the nurse pulled out a large file of paper dominated by handwritten notes that constituted my medical record. And when I left I was given a piece of paper with illegible marks to take to the pharmacy so my prescription could be filled.
Convenience and cost savings are obvious reasons why digitization would bring value. Dr. David Brailer, the national coordinator for health information technology, argues that electronic health records could save $140 billion per year in the U.S., 8 percent of health care spending, as reported in an interview with Tim Mullaney, BusinessWeek E-Business editor, March 28, 2005. According to the Institute of Medicine, $300 billion was the estimate for one year for treatments that may not improve health, may be redundant, or may be inappropriate.
Perhaps a more important motivator is improved health care, particularly in the area of safety. According to the Institute of Medicine:
- 44,000 to 98,000 Americans die each year from inpatient medical errors
- 770,000 Americans are injured or die each year from adverse drug events
It can be argued that lower costs of health care make this care more widely accessible and hence lead to improved health care, but we have to be careful not to take this argument too far. Clearly, more complete and accurate records could go a long way toward addressing both cost and safety issues. Since the technology is available and has been used in other industries, why has not more progress been made in health care?
One important reason is the past underinvestment in IT in the industry, according to Brailer. A substantial new investment would be required to put in the infrastructure to support this new paradigm. A second reason is the fragmentation of the industry; there has not been a force to integrate the factions. Brailer’s appointment to the position of czar for health care IT is an effort to push toward a common solution, but the small amount of funding ($125 million per year) may hardly make a dent in the problem.
Technology, cost, and people issues have slowed the application of information technology in health care.
A major technical issue concerns standards for medical records. It would seem to be relatively straightforward for a single hospital or clinic to computerize its records. Lessons from other industries would suggest this is not as easy as it appears. Common terms and definitions are needed to form a very precise language for communication between computers. Human communication is much more forgiving. To allow these records to be exchanged between hospitals, pharmacies, and ultimately across international boundaries, these agreements would have to extend beyond what any single hospital or clinic could control.
The financial industry has already accomplished this standardization, and their Electronic Funds Transfer network and protocols enable the transfer of money many places in the world. This standardization was achieved through the same kind of laborious process of agreement on protocols and data definitions—but a financial record is fundamentally simpler than a medical record.
Systematized Nomenclature of Medicine (SNOMED) represents the combined effort of the NHS and the College of American Pathologists to achieve common terms, and is widely distributed in the U.S. There are other nomenclature efforts in Europe. The National Institute for Standards and Technology is engaged in a coordinating role toward the development of more comprehensive standards (www.NIST.gov). Further, there are some aggressive health care facilities such as the Cleveland Clinic and the Hackensack Medical Center in Hackensack, New Jersey, that are moving forward to demonstrate the cost savings (and life saving) benefits of the effective use of information technology.
Privacy and Security
Even given the ability to exchange records, there remain issues of privacy and security. In the U.S., the Health Insurance Portability and Accountability Act (HIPAA) has added significant costs and paperwork to health care facilities to assure privacy of medical information. In some cases the legal concerns have caused hospitals to take a particularly conservative stance, and these policies would have to be thought through carefully for a digital world.
The health care security issues are quite different from those in financial transactions. First, those who would hack into health records would not profit from them in the same way that they might from accessing financial records. But they could interfere with employment opportunities, if a company were able to collect my medical history and sell it to a future potential employer. Potential for a costly disease might cause an employer to look for another candidate. Second, tampering with medical records could be a matter of life and death, not just dollars. A small change in an MRI, or a drug prescription, could be deadly. Hence, those thinking about the security of medical records need to protect from different scenarios than those who work on financial records security.
All of us have had experience trying to use software designed by a computer “geek” but never properly tested by an average user.
A few years ago, Boeing, NIST (www.NIST.gov), and other industry partners got together to see what it would take to put a “usability” stamp of approval on software. Users who were experts in their application area (engineers, finance, HR, etc.) were brought into a laboratory to try to work with new software. Their keystrokes were recorded, the cameras caught their facial expressions, and the microphones captured their words as they attempted to get their jobs done. The goal was to test how intuitive it would be for a new person to use the software. Feedback from such sessions was used to improve the software human-computer interface.
It comes as no surprise that usability issues are also a problem for doctors dealing with electronic medical records. A University of Pennsylvania study reported by Robert Langreth in Forbes.com (March 8, 2005) showed that the introduction of some health care software actually increased the odds of prescription drug problems. “The system didn’t make sense in terms of the way doctors actually worked,” according to Ross Kopel, lead author of the study, as reported in the Langreth article.
Usability is a fixable problem, with some good tools and strategies being developed at NIST and elsewhere. For the medical field, usability testing should be a requirement for all new technology solutions.
Beyond the technical issues briefly described here, there are a host of nontechnical issues that stand in the path of the successful electronic transformation of health care.
A big one is cost. First is the cost of hardware and software. David Brailer estimates that hospitals generally spend about 2-3 percent of revenue on IT compared with significantly greater expenditures from other industries. Dealing with this investment when all of the medical costs are under pressure will be a significant challenge. But a second financial question relates to how revenue will flow from a more open system. Today, hospitals strongly compete with each other for medical dollars. While it is desirable for the patient to have more open access to medical care, it is less clear how this will affect particular institutions. Institutions often act parochially in such situations.
A second reaction comes from the medical professionals. They have been under enormous pressure in recent years, particularly in the U.S. Declining incomes, increasing costs, particularly for insurance, have left many of them much more resistant to other changes—from writing to data entry and from some control to much less control in this emerging patient-centered environment. There are very innovative technology users in the medical profession, but it will take a “tipping point” for this technology to have a more widespread impact.
Other industries have faced the technical, cost, and people issues inherent in transforming themselves through technology. In this sense, the health care field has a better roadmap to follow. Like other industries, facing the significant transition effort and costs is formidable, and a barrier for many. It will take some early adopting champions (like Dr. Gerard Burns, featured in the BusinessWeek cover story, March 28, 2005). It will take the standards effort described earlier. It will take some government seeding of the effort, some pressure by funders of the medical field, and some pressure by patients. It will take some innovative, and useful, products coming from the vendors. Ultimately, I believe we can expect to reach a “tipping point” sometime in the next five years.
Perhaps it is good that health care has lagged behind so many other industries. Health care is a matter of life and death, and it’s important that we get it right. Making the problem more difficult is the conflicting measures of right. For some administrators, lowering costs becomes the driving definition. Unfortunately, lowering costs and improving health care and safety are not perfectly aligned. We need to be watchful of those implementations that reduce cost and increase risk.
Hopefully, the most near-term progress will be those areas where cost and safety are not in conflict.
Al Erisman is executive editor of Ethix, which he co-founded in 1998.
He spent 32 years at The Boeing Company, the last 11 as director of technology.
He was selected as a senior technical fellow of The Boeing Company in 1990,
and received his Ph.D. in applied mathematics from Iowa State University.