IBM, health group sign deal to mine patient data to improve care
Geisinger Health System and IBM said today that they will partner to develop a system for integrating patient data with billing, claims, financial and other systems as part of an effort to improve patient care.
Under the multiyear agreement, the two companies will design and put into place a tools called the Clinical Decision Intelligence System, which will mine data from Geisinger's electronic health record (EHR) system and other sources to create customized treatment plans, ensure that patient care meets national standards and identify best practices.
While large hospitals and health care systems across the country have begun the move to automate patient data through EHRs, there are only a handful that have started to integrate and mine patient data for best practices and suggested treatment plans. The New York-Presbyterian Healthcare System in August 2005 began rolling out an IT system that generates suggested care plans for physicians based on data about previous patient outcomes. The Mayo Clinic and IBM in August 2004 said they were starting to use a DB2 database to help physicians treat patients.
The foundation of Danville, Pa.-based Geisinger's project -- which will take place over the next 18 to 24 months -- is an enterprise data warehouse that will integrate data on about 2.5 million patients from Geisinger's 650 physicians, three hospitals, its health insurance company and its research center, said Ronald Paulus, Geisinger's chief technology and innovation officer.
Although he declined to provide the cost of this project, Paulus said Geisinger so far has invested $70 million in building its EHR using software from Epic Systems Corp. This project will allow the health care provider to help close the gap -- present in health care nationally and internationally -- between what treatments are known to be helpful to patients and what care the patients actually receive, he said.
"Once we get this data organized within the warehouse, we are ready to ... thoughtfully analyze and mine that data so we can tease out those best practices locally and implement them directly into the Epic electronic record," Paulus said.
Because the health system has 100% adoption of its EHR, it has reached a "quasi endpoint in how far we will get in simple decision support functions," he added.
"The big gap is not installing systems but changing the way we care for patients fundamentally from the bottom up," he said.
After the warehouse is in place and monitoring treatment against standard best practices, the health system will begin to look at additional ways to analyze its information, such as researching the best medicine for patients based on their genetic make-up and more quickly identifying patients who are good candidates for clinical trials, Paulus said. The data also could be used to identify the best treatment plan for patients based on their preferences, such as their willingness to have surgery or to risk the side effects associated with certain medicine, he added.
Brett Davis, IBM's global solutions executive for health care, said the Geisinger project eventually could be used as a model for future efforts at other health care organizations because its breadth of services is representative of much of the health care market. For example, Geisinger serves patients from rural and suburban areas, includes a large physician practice and performs some clinical research.
"They have those basic building blocks -- that EHR -- that allows a project like this to move forward," Davis said. "The rest of the market is going through a clinical transformation ... to get those clinical systems in place."
Subscribe to RSS feed