Prognosis Good for Text-based Data Analysis

University of Louisville researcher digs into unstructured text to cut patient costs, save lives

Cary, N.C. — December 15, 2003 — The University of Louisville has been using business intelligence provider SAS' Text Miner software to analyze unstructured text in hospital billing, medication orders and physician's chart notes. The university said its research revealed potential savings in patient costs related to risk adjustment, as well as improved the quality of patient outcomes.

Dr. Patricia Cerrito, who holds a doctoral degree in biostatistics and is a professor of mathematics at the University of Louisville, said she and her team research cancer, genetics, heart disease and environmental health. For her research, Cerrito uses SAS analytics including Text Miner, Enterprise Miner and STAT. Cerrito said the integration of the analytic tools allows her to tap into analytic functionality, since she can move between tools to explore structured data and unstructured text.

Cerrito explained that the ability to unlock the intelligence held in documents is a major advance. By using the analytics tools to examine large hospital databases for relationships between physician practices and patient outcomes, Cerrito is solving patient care problems that have existed for years.

For example, prior to the use of SAS Text Miner, variation of physician practice patterns were surveyed but were not analyzed for critical insights. Cerrito analyzed relationships between particular physician practices and thousands of patient outcome records and learned that the prescription of certain medications resulted in prolonged patient hospital stays.

In another project, Cerrito and other researchers determined that Text Miner's automatic extraction of patient information from the pharmacy order database could enhance or replace manual extraction from patient charts. Using Text Miner insights, Cerrito regularly presents and publishes research results that reduce hospital costs and improve patient care worldwide.

Cerrito said the benefits of her research results hit close to home when her husband, John Cerrito, received open-heart surgery at Jewish Hospital, Louisville. His surgical team made medication decisions that were based on Cerrito's research that connected the prescribing of antibiotics to diabetic patients undergoing open-heart surgery and their post-surgical outcome. Having used the analytics software to detect a relationship between high glucose levels and risk for infection, and then seeing her husband's positive outcome first-hand, Cerrito is certain that the surgical team's research-based medication decisions worked to his benefit.

Cerrito said SAS has done a good job of integrating its analytic tools. "Text Miner highlights relevant patterns in documents such as clinical reports, and it quantifies text-based information. We move seamlessly to Enterprise Miner to combine and analyze this unstructured text with structured data such as demographics and laboratory values. And, we augment that analysis with SAS/STAT. That's why we standardized on SAS for our research," she said. "During any given work day, these benefits add up to significant return on investment at the University of Louisville."

Loading