IBM, Mayo Clinic to develop trial, research database

The Mayo Clinic is teaming up with IBM to put the health center's huge patient records database to work in helping to more quickly identify patients for clinical trials.

In an announcement today, the Rochester, Minnesota-based clinic said it will work with IBM to create a database system that could also help speed treatment decisions garnered using data from some 6 million patient records. The patient records will be stripped of names and other personally identifiable information in the database, according to Jeff Auger, director of strategy for IBM Life Sciences, which is developing the project with the Mayo Clinic.

The new database system will be built in three phases that will take about 18 months, Auger said. Patient information will be loaded into an applications layer on top of IBM's DB2 database program. A prototype version is running on a midrange IBM eServer running AIX, while the final hardware will most likely be a multiple Regatta server p690 configuration deployed across the Mayo Clinic, he said. The system is being designed to provide links to other public and private data sources using IBM's DiscoveryLink data integration software.

The database will help Mayo Clinic doctors look beyond an individual patient to see how similar patients with similar medical histories and cultural backgrounds have been diagnosed and treated for similar problems, Auger said. The Mayo Clinic has about 2,400 physicians and scientists in more than 100 specialty areas. Eventually, they could use the information in the database to draw meaning from patient information as detailed as genomic data.

The first phase will be completed in July and will take current medical records and make them ready for DB2 use.

The second phase, which will begin in July, will extend the database structure for new information that will include genetic data which is now being collected at the clinic.

Dr. Piet de Groen, the Mayo Clinic's project director for the database, said in a statement that the project will "immediately benefit our research activities by providing investigators with rapid access to clinical data for use in ongoing research activities."

Dr. Robert Beck, a vice president and chief information officer at the Fox Chase Cancer Center in Philadelphia, said improved diagnostic and treatment uses for patient databases are constantly being sought in the medical field as a way of improving care. Other hospitals and medical institutions are working on similar projects, but none have a database as large as the Mayo Clinic, he said. Fox Chase's 100-bed hospital is the fourth largest in the U.S. devoted entirely to cancer care.

One shortcoming, Beck said, is that although large patient databases provide a wealth of historical information over the years, they may not be as effective for treatment information, because new drugs and treatments wouldn't have been available more than five or 10 years back into the database.

Join the newsletter!

Error: Please check your email address.

More about IBM AustraliaMayo Clinic

Show Comments