A new study has found that a medical-information exchange system that is considered a model for health-care reform efforts saved significant amounts of money and led to better care for patients—including a woman who probably would have died without the system.
The woman was bleeding from her uterus when she came to the emergency room of Saint Francis Hospital in Memphis, Tennessee, in 2009. If her medical records had been unavailable, doctors probably would have ordered an ultrasound, incurring some delay in treatment. But because of the city’s digital information-sharing program—a rarity among U.S. hospitals with different owners—the doctors learned that an ultrasound done days earlier at another facility had detected that the woman had an ectopic pregnancy, in which the embryo becomes implanted outside the uterus. (It’s not clear whether the first institution had failed to follow up or whether the patient refused treatment there.)
This knowledge prompted the Saint Francis doctors to rush the woman into surgery. Her blood pressure dropped to zero on the operating table, putting her in danger of imminent cardiac arrest and death. But she survived, and one of the doctors later wrote in an e-mail to a colleague at Vanderbilt University, where the information-sharing network was originally developed: “Ehealth saved a life tonight.” The study, led by Mark Frisse, a physician and professor of biomedical informatics at Vanderbilt, was published last week in the Journal of the American Medical Informatics Association.
The larger point of the study was to reveal the results of sharing patient data electronically between 12 hospital emergency rooms in the Memphis area. The researchers found that the participating hospitals reduced health-care costs by $2 million over 13 months, largely because doctors avoided needless admissions, CT scans, and other tests after getting insights from the patients’ medical histories. In other hospitals, emergency doctors often fly blind and overtreat or overtest incoming patients.
Earlier studies have pointed to savings that came from adding IT to other elements of the health-care system, such as reducing clerical work through various kinds of software that aid the billing process or prescription-writing. But this was the first to “document rigorously” the cost reductions in patient care, says John Halamka, chief information officer at Harvard Medical School and a leading advocate for health IT networks. The research indicates that the savings for hospitals around the United States could be in the billions of dollars and lead to better medical outcomes, Halamka adds.
The Vanderbilt study comes amid a rapid ramp-up in the use of electronic medical records by U.S. doctors and hospitals, prodded by the incentives in the 2009 stimulus act. In the past three years, the adoption of electronic medical records by U.S. doctors has roughly doubled to around 30 percent. On top of that, 80 percent of hospitals are also beefing up their electronic health records.
This new IT infrastructure sets the stage for sharing patient information as Memphis is doing. And while reducing tests and other procedures might run counter to hospitals’ profit motives, the incentive structure may change under the U.S. health-care reform legislation, which would pay doctors a fixed annual fee to keep patients well, giving them an incentive to reduce duplicate and unnecessary testing.
The Vanderbilt study compared emergency-room outcomes from two groups of patients: one whose clinical background was obtained through electronic records, and another group, with similar conditions or complaints, whose data was not available to doctors.
The study focused on hospital admissions, laboratory testing, head CT scans, body CT scans, ankle and chest X-rays, chest radiographs, and echocardiograms. The clinicians only used the health-exchange information when they felt it was necessary. As things turned out, such checking was done in seven of every 100 patients—but was done at least 20,000 times during the study.
The information-exchange system works like this: unless patients specifically opt out, clinical records from 12 hospitals in the Memphis area, plus any of several county-run health clinics, are automatically sent to a common data center in Nashville run by Informatics Corporation of America. Then, when someone comes into an emergency room or visits a hospital for any other reason, the registration process brings up a screen summarizing how many times that patient—as identified by name and birthdate—has visited any of the other facilities. A doctor can click on that summary to see a chronological report—including the patients’ reported problems, their lab results, and doctor-dictated discharge summaries—from each of the hospitals.
The study found that obtaining such data was associated with a decrease in CT scans and other diagnostics as well as a reduction in hospitalizations compared to the other group, Frisse says. Administering the system costs $880,000 a year, he added, and the benefits should improve as usage increases across more departments. He added that the savings was a tiny fraction of what could be achieved in Memphis alone, especially if every doctor’s office was also linked to health-information exchanges.
In addition to saving the pregnant woman’s life, the system also allowed doctors to learn that a man who arrived in one emergency room had active tuberculosis that had been diagnosed elsewhere. This enabled the doctors to isolate the man from other patients, the researchers say.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.