Analyzing EHRs
Clinical Looking Glass is a software tool for analyzing electronic health record data. It allows users to quickly define patient groups and clinical outcomes of interest and obtain study results within minutes. (Image courtesy of Emerging Health IT, Inc.)

The Military Health System cares for 9.6 million service members and their families"with costs that make up 10 percent of the U.S. military budget. A large portion is spent on care for those with chronic diseases such as diabetes, heart disease and asthma. Improving both efficiency and health outcomes will have enormous benefits for service members as well as the nation.

Locked within the massive amounts of data being collected in electronic health records today are the keys to improved outcomes and lower costs. If the right analytical tools were available, clinicians could find quick answers as to which treatment options are working best, what measures are preventing hospitalization for chronic diseases, or whether a new drug is causing dangerous side effects.

The Department of Defense is working with a civilian partner to test new analytical software that enables physicians and other users to ask a question and receive a statistically valid answer based in EHR data"within minutes. The Clinical Looking Glass application is obtaining results comparable to those that in the past have taken weeks, months and even years of research along with a team with computational expertise.

CLG was developed at Montefiore Medical Center, a university medical center with four hospitals in Bronx, N.Y. A test project to use CLG with Military Health System data is being conducted with the U.S. Army Medical Research and Materiel Command’s Telemedicine and Advanced Technology Research Center, the U.S. Air Force Surgeon General’s Office and the Navy Space and Naval Warfare Systems Command.

The groups have set up a secure environment to host CLG and evaluate its usability among clinicians in the Military Health System. CLG is expected to be piloted at Walter Reed National Military Medical Center in Bethesda, Md., in September.

CLG grew out of a small team’s drive for information that could possibly control the rampant spread of tuberculosis and HIV at a correctional facility in New York City. Montefiore physician and CLG visionary Dr. Eran Bellin notes that the tool is now a crucial part of Montefiore’s quality assurance efforts and a required part of training for internal medicine residents. Among other things, it has been used to monitor the success of a diabetes treatment model, identify those at risk of a fatal reaction to a certain drug used to treat anemia, and publish papers on asthma, HIV and ER treatment of the elderly.

Says Bellin, “With CLG, we can marshal the huge amounts of data we’re collecting through EHRs in order to vastly improve patient care. We’ve never had a way to analyze and visualize data with such flexibility and ease. CLG also enables clinicians to discover unanticipated events as well as find patterns that may help prevent major health issues.”

Robert Connors of TATRC is managing the DoD project. He notes, “We’ve demonstrated the system to hundreds of military physicians, nurses and administrators, with favorable reviews. The next step is to use it at Walter Reed with real data to conduct quality assurance studies and, ideally, remediate patient care issues at the point of care.”

CLG allows clinical team members to create cohorts (groups of patients sharing common factors) for quality assurance studies, with just a few simple computer inputs. Cohorts can be compared statistically in minutes to discover differences in whatever health outcome the user defines. CLG can save these cohorts for further use, and produce reports suitable for publication in research journals. Privacy protections are built into the system.

TATRC deputy director Col. Ron Poropatich says, “We already know what some of our key costs and problems are. With a tool such as this, Military Health System doctors could, for instance, compare outcomes for different blood pressure medications among our patients to see which of these expensive drugs are really working.”

According to Bellin, no other tool builds longitudinal cohorts on the fly with unlimited temporal relationships, which is what allows for complex queries of data and almost immediate answers.

Bellin says, “Doctors can use CLG for patient remediation management, a new concept that is a core of cost-effective care. They can easily find out whether what they’re doing to treat specific conditions is working, and if not, what they can do differently.”

Page last updated Wed July 6th, 2011 at 15:14