In Rural Pennsylvania, Modern Medicine Seeks New AnswersIn Rural Pennsylvania, Modern Medicine Seeks New Answers

A decade ago, Geisinger Health System began digitizing medical records for its more than 2 million patients in rural Pennsylvania. Today, Geisinger has a treasure trove of electronic clinical data that very few U.S. health care providers have. A new five-year alliance with IBM aims to help Geisinger leverage that data to improve quality of care and patient outcomes, as well as provide more personalized treatment plans.

The collaboration between IBM and Geisinger includes building a "clinical decision intelligence system" with a standards-based infrastructure that integrates financial, administrative, claims, and clinical systems. The plan calls for clinical data warehouses and analysis tools to help Geisinger boost the quality of care.

"Quality is the top goal of this," says Dr. Ron Paulus, chief health information technology officer at Geisinger, which runs three regional hospitals in Pennsylvania, a nonprofit health plan, and a 650-physician group practice. Huge gaps exist between evidence-based best practices and what actually happens in the delivery of health care, he says.

Targeted Treatment
Geisinger has invested about $70 million over the last 10 years deploying its electronic medical record system from Epic Systems, which has been in use by all Geisinger physicians for the last five years. The next steps of tapping into the clinical data will help Geisinger doctors provide better customized care for patients based on factors that include treatment outcome data, best practices, genomic profiles, and patients' personal preferences based on finances, belief systems, or other considerations.

Decision support and analysis tools could help a Geisinger physician treating a cancer patient decide on the best course of treatment based on outcome data of other patients who have the same health problems and type of cancer and possess similar genomic markers. The system might also help doctors and patients decide on the best treatment plans based on medications covered by health insurance. Or this: Doctors could determine the cost and impact of using different drugs to treat a given ailment--say, standard heparin therapy instead of newer anti-coagulants to prevent deep venous thrombosis in orthopedic surgery patients.

Such analysis could help tackle larger societal issues such as the rising cost of health care, Paulus says. The mining and analysis of Geisinger clinical data also could help better identify patient candidates for drug and other clinical trials.

"Most of the world is just getting to the point where they're saying, 'We should install electronic medical record systems,'" says Paulus. "We're moving up to the next level." A study released last week by Massachusetts General Hospital and George Washington University estimates that 24% of U.S. doctors have installed basic e-health record systems, but only 9% use more advanced systems that can help reduce costs and medical errors.

IBM is working with a number of larger health care providers in projects that also aim to personalize and improve care, including the University of Pittsburgh Medical Center and the Cleveland Clinic. What makes the alliance with Geisinger unique is the health system's rural setting; its patient demographic, which includes multiple generations of families that receive most of their care from Geisinger doctors; and its decade-long use of e-health records.

The dollar value of Geisinger's contract with IBM wasn't disclosed. Paulus says the alliance should provide "tangible output" to Geisinger physicians, including new decision-support and analysis tools, in less than two years. IBM expects versions of the system being developed at Geisinger also will be made available to other health care providers.