Skip to content

Kum awarded $1 million to facilitate use of ‘Big Data’ to improve care while protecting patient privacy

big data research

Thanks to a new $1 million funding award, a Texas A&M School of Public Health research team will explore new ways to integrate diverse datasets for improving patient outcomes through data-driven research that is effective and respectful of patient privacy. The award from the Patient-Centered Outcomes Research Institute (PCORI) will support this research project for the next three years.

The project, led by Hye-Chung Kum, PhD, associate professor in the Health Policy & Management Department in the Texas A&M School of Public Health with a joint appointment in the Computer Science and Engineering Department in the Texas A&M College of Engineering, will study how to best balance the quality of integrated data used in health research with the privacy rights of the patients the data describe.

For years, researchers have integrated and used so-called secondary data such as hospital discharge information and insurance data to improve health care, but a critical task to reap the benefits of big data for population health is data integration, a process known as record linkage, or RL. RL identifies the same patient records from many different data sources without a common identifier. Sometimes called patient matching, it is often conducted via a hybrid system of automatic linkage using software, and cleaning and resolving uncertain links manually.

Due to the sensitive nature of the data that are needed for RL, there are concerns about privacy, leading to many obstacles to getting full access to all required data. Some concerns are real, but many are only perceived obstacles. This privacy concern can be overcome by suppressing identifying information in data sources, but this comes with the costs of reduced RL accuracy and missing data that leads to suboptimal outcomes. Additionally, many patients are unaware of where their data will be used in the future, making patient consent a tricky issue.

“Information privacy is a complex problem that requires a holistic approach,” Kum said. “The only viable solution is at the intersection of technology and regulations where technology and regulations are reconciled and applied to safely use only the needed identifying information for optimal linkage.”

The research team will work to determine how much information is actually needed for quality data integration and study the relationship between RL quality and how much patient data is suppressed.

“We will also develop tools to analyze privacy risks involved when using medical data and communicating the technology to patients,” Kum added.

Patient-centered health care means that patients are expected to be active participants in their care decisions. To do this, patients will need better information so they can make informed choices. “Part of this involves knowing how much of their personal data could be used by researchers, what the risks are to their privacy and what steps researchers are taking to protect them,” Kum said. “Transparency is one of the most effective mechanisms for privacy protection.”

Kum will conduct the study with a multidisciplinary team that will use information privacy protection techniques, human computer interface design, and input from health care researchers, legal experts and patients. Additional researchers include Alva Ferdinand, DrPH, JD, and Cason Schmit, JD, both from the Department of Health Policy and Management at the School of Public Health as well as Eric Ragan, PhD, from the Department of Visualization in the Texas A&M College of Architecture.

PCORI is an independent nonprofit, nongovernmental organization located in Washington, DC, that was authorized by Congress in 2010. Its mission is to fund research that will provide patients, their caregivers and clinicians with the evidence-based information needed to make better-informed healthcare decisions. For more about PCORI, visit: https://www.pcori.org.

Media contact: media@tamu.edu

Share This

Related Posts

Back To Top