This story is from the category Health
Date posted: 15/08/2010
Looking out his office window in Seattle, Thomas Payne can see two hospitals that use the same electronic record system as his own. And yet, says Payne, medical director of information technology services at the University of Washington, they all still exchange information by fax or paper.
That situation is the norm in today's fragmented and distributed health-care universe, where electronic medical records (EMRs) must increasingly draw on information from the multiple health-care institutions--clinics, hospitals, and specialists--where people receive care. But giving physicians access to the right information at the right time could dramatically streamline medical care. An estimated $77.8 billion, or about 5 percent of health-care costs, could be saved each year in the U.S. if a fully interoperable record-sharing system were in place, according to a 2005 study in Health Affairs. Most of the savings would come from preventing duplicate testing.
Michael Zalis, a radiologist at Massachusetts General Hospital, has experienced the drawbacks of the splintered system firsthand. He says gathering data on particular patients is a huge time sink for physicians, taking as much as 20 percent of their time. Even with an EMR, medical records are spread over multiple databases that don't necessarily communicate with each other. Older, less relevant test results could turn up as frequently as new ones, says Zalis, requiring the doctor to visually scan all of them. The problem will likely grow as many of the nation's hospitals and doctor's offices use new stimulus funding to implement EMR systems.
Zalis aims to alleviate that problem by applying lessons from search engines to medical databases. He likens the current EMR situation to the early days of the World Wide Web: larger and larger stores of data, with increasing need for indexing and searching capabilities. Before good search engines like Google arrived, finding the right webpage could be a challenge. Many links and pages had little relevance, or were so out of date they were almost useless.
Over the past five years, a team led by Zalis and his Massachusetts General Hospital colleague Mitchell Harris developed a program called Queriable Patient Inference Dossier. The program combines a search engine with a programming system to automatically pull data from various EMRs and databases and process the information. While this may sound simple, it's actually a major improvement for doctors, as most EMRs have little to no built-in search capabilities. They described their system in the August 2010 issue of Journal of the American College of Radiology.
While Google's PageRank system works by giving more weight to pages that are linked to more often, EMRs don't have links and therefore cannot employ that approach. Instead, the dossier system has the ability to "learn" certain types of searches from its users, understanding that a search for "squamous cell carcinoma" and another search for "lung cancer" are actually seeking the same information.
The program, which is employed by a network of Boston-area hospitals, now has more than 800 registered users and posts more than 10,000 pages of medical-record information per day. Zalis says other large health-care organizations are beginning to use it, though expanding its use is likely to be a challenge. While the Dossier program can integrate with other EMR systems to provide advanced search capabilities, doing so requires permission from the different hospitals and medical centers involved, as well as adequate export capabilities. Some hospitals say that the vendors of EMR systems have made it difficult to access information in the databases from outside their proprietary programs.
See the full Story via external site: www.technologyreview.com
Most recent stories in this category (Health):
01/05/2013: Microelectronics: Taking the heat off microfluidic chips