IBM Partners With Cleveland Clinic For Better Medical Data Access

The first project will focus on improving outcomes and treatment for patients with abdominal aortic aneurysms by letting doctors electronically tap into research discoveries at the bedside.
Cleveland Clinic and IBM this week said they're collaborating to provide the clinic's patients with more customized treatments by allowing doctors to electronically tap into research discoveries at the bedside.

Under the pact, IBM and Cleveland Clinic will develop a "translational medicine platform," or infrastructure that ties together patients' electronic health-record data with the clinic's clinical, genetic, and other research data.

The first project will focus on improving outcomes and treatment for patients with abdominal aortic aneurysms, a condition that often shows no symptoms until the aneurysm is about to rupture, in which case the survival rate is less than 50%.

"If we can predict what patients are at risk for these ruptures, we can identify the patients who need surgery" or other treatments," says Dr. Kenneth Ouriel, chairman of the Division of Surgery and the Department of Vascular Surgery at Cleveland Clinic.

The infrastructure being built at Cleveland Clinic will allow researchers to access "blind" patient data from all repositories in the clinic, including lab, genetic, imaging, and even financial information, Ouriel says. "It will pull information from all areas of the institution and give it a friendly front end," he says.

The goal is to allow clinic researchers to look for variables such as a patient's age, drug history, and genetics and how they contribute to risks as well as outcomes to treatments. Initially, this work will focus on aneurysms, but later on other illnesses, such as prostate cancer, Ouriel says.

The clinic's doctors within a year will be able to tap into this research when treating aneurysm patients, Ouriel predicts.

Without the use of the pattern-recognition and data-mining tools being applied to patient's electronic health records and the clinic's data repositories, medical students would need to scour thousands of charts to manually compile variables and data for analysis by researchers, he says. "In this age of the electronic health record, that's foolish," he says.

The application of data-mining, pattern-recognition and other technologies to clinical data, and especially genomic information, to help doctors improve care for individual patients is an important way the practice of medicine is being transformed by IT.

The work between IBM and Cleveland Clinic follows a similar customized medicine collaboration revealed in August between IBM and Mayo Clinic in which the two are applying data-mining and pattern-recognition technologies to electronic health records of more than 4 million Mayo Clinic patients. A goal of that project is to give Mayo doctors by the end of the decade the ability to, for example, electronically tap into research findings about the best treatment outcomes for cancer patients who might have specific traits such as rare genetic mutations.

"This is IT meeting molecular medicine," says Mike Svinte, IBM's VP of information-based medicine.

Ouriel says the technologies being applied to data at Cleveland Clinic might have allowed the institution to discover the correlation between heart attacks, strokes, and patients taking the prescription arthritis drug Vioxx, which was pulled from the market this week by Merck & Co.

Hypothetically, Ouriel says, "we could've looked at patients with heart attacks and seen a common variable of being on Vioxx."