Smarter Is: Breaking Data and Disease Down to Size

How one Chinese hospital is using big data analysis to make the most of both traditional and Western medicine

Chronic kidney disease (CKD) is the ninth leading cause of death in the United States, according to the Centers for Disease Control and Prevention.* Modern Western medicine has had little impact in staying the disease’s deadly hand.

But Eastern doctors have found hope for patients with CKD in traditional Chinese medicine (TCM). Over the years TCM has proven effective in delaying the need for kidney dialysis. Guang Dong Hospital of TCM in Southern China treats 10,000 kidney patients a year; if doctors can establish the combinations of treatments that are consistently effective for different categories of CKD patients, they can apply evidence-based treatment plans with confidence, helping to save lives.

Although the Guang Dong Hospital of TCM has access to millions of electronic medical records, much of the patient data will, in the future, be captured in large XML documents that are difficult and time-consuming to sift through and analyze, according to Sarah Knoop, manager for Healthcare Systems Research at IBM Research – Almaden, in San Jose, California.

To address this problem, IBM and the Guang Dong Hospital teamed up to create Healthcare Information Warehouse for Analysis and Sharing (HIWAS), a prototype plug-in–based tool for IBM® InfoSphere® Warehouse. The hybrid solution extracts the existing, complicated XML data and enables it to be converted into more easily analyzed formats, says Knoop. “[These formats are] the most appropriate, efficient storage representation for the business problem at hand leveraging DB2 technology,” she says.

“For clinicians to use this data in the clinical setting, they have to be able to extract it in order to analyze it,” explains Knoop. “However, doing so can be difficult because these XML models are typically complex and more general.”

Knoop also says that HIWAS helps extract valuable information for further analysis so it can be used by IBM Cognos analysis tools. HIWAS converts the excised data into warehouse schemas for hybrid relational-XML or pure relational databases, Knoop explains. Cognos is then able to analyze the data and create reports that highlight the significant relationships between the data. Doctors use those relationships to build a base of clinical evidence that reveals best practices around treatments for CKD, according to Knoop.

With HIWAS, doctors can analyze the data and categorize CKD patients according to those who became sick because of other diseases or in combination with other chronic diseases. This will help doctors understand which combinations of Western treatment plans and TCM approaches are most effective for certain categories of patients, says Lu Yu Bo, president of Guang Dong Hospital of TCM.

“As clinical doctors, we can use IBM technology to find more useful information from the patients’ cases,” says Mao Wei, professor of nephrology at Guang Dong Hospital of TCM. “This useful data can help us to improve our effectiveness in traditional Chinese medicine in the treatment of chronic disease.”

*The Battle Against Diabetes: Progress Made, Challenges Unmet,” by Ann Albright, CDC Congressional Testimony, July 1, 2010.