Earlier this week, we published part one of this two-part series in which we highlighted how the combined use of claims data and electronic health record data can drive better research outcomes. However, we also noted that combining these two datasets comes with inherent challenges due to differences in coding and categorization of records.
In an effort to overcome these challenges, two pilot projects funded through a public-private partnership with the Patient-Centered Outcomes Research Institute (PCORI), the U.S. Food and Drug Administration (FDA), and the Regan-Udall Foundation (participating in an advisory capacity) took on an ambitious effort: creating an infrastructure capable of linking PCORnet—one of PCORI’s most innovative initiatives that offers researchers electronic health record data—and the FDA’s Sentinel program, which offers mostly claims data. Part one of this series explored the pilot project focused on the congenital Zika virus, and this article will explore the second pilot, which focused on monitoring exposure to antimicrobials.
An Initiative to Monitor Exposure to Antimicrobials
The challenge: Antibiotics and similar drugs, together called antimicrobial agents, have been used for the last 70 years to treat patients who have infectious diseases. Unfortunately, these drugs have been used so widely and for so long that the infectious organisms the antibiotics are designed to kill have adapted to them, making the drugs less effective. Exacerbating this problem is overuse and misuse of these medicines. In hospitals, approximately 30 percent of antibiotics are unnecessary or prescribed incorrectly. Sometimes patients are treated with antibiotics for diseases or conditions that are not caused by bacterial infections. With better data, we can better inform hospitals and doctors on over-prescribing of antibiotics and encourage them to only give antibiotics when necessary. This study team piloted a methodology to create antibiotic use reports from PCORnet and Sentinel data for sending such reports to the Centers for Disease Control and Prevention’s National Healthcare Safety Network. The National Healthcare Safety Network compares rates of antibiotic use across similar hospitals and reports these comparisons back to hospitals.
The process: The team’s primary focus was to develop a PCORnet technical infrastructure in partnership with Sentinel that would generate more complete information on how antibiotics are being used and prescribed. The structure provides a framework that can benefit healthcare facilities locally as well as through reporting to the National Healthcare Safety Network. Once developed, the team implemented this augmented data structure at five healthcare institutions across three PCORnet partner networks (CAPriCORN, Mid-South, and REACHnet).
“PCORnet and Sentinel both have Common Data Models to standardize data, but neither were sufficient for the goals of our project,” said Bill Trick, principal investigator for CAPriCORN. “For example, these models may tell us that antibiotics were dispensed at a hospital, but not the location in the hospital, which was important to our research. As a result, we had to work in tandem with the Sentinel team to extend both Common Data Models and build something unique to this project.”
“We are proud of our outcomes
with this work and see it as
a stepping stone for future researchers.
Bill Trick – Principal Investigator, CAPriCORN
The result: The team effectively developed a final data model, dictionary, and analytic questions that are now available for use beyond the project institutions. In fact, the infrastructure has already been leveraged to pilot a system to monitor dangerously low blood sugar events among hospitalized patients with diabetes.
“We are proud of our outcomes with this work and see it as a stepping stone for future researchers,” said Trick. “We are already working on a new system to monitor antibiotic resistant organisms within healthcare settings using the lessons learned from this study.”
The Road Ahead for Data Linkage Infrastructure
Trick and Bill Hogan, the principal investigator for the Zika pilot, both stressed that developing an infrastructure that allows data linkage of two distributed networks like Sentinel and PCORnet was no small feat. Data is complex, and many challenges can make the linkage process time consuming and tedious. However, both feel the merits far outweigh these burdens.
“Once we take the effort to build these linked infrastructures, health institutions can benefit from national assessments that will tell us much richer detail about the health of our country,” said Hogan. “PCORnet and Sentinel are both valuable tools individually, but together they augment each other, painting a more detailed picture of what we seek to understand.”
These detailed pictures of our pressing health challenges are key because with greater detail, we don’t have to theorize with gaps in information. At the same time, we can get to the bottom of questions that need answers with greater speed and confidence. With that in mind, it is a pretty sure bet that we will see more data linkage efforts in the future.
This article is second in a two-part series on PCORnet’s data linkage projects. Learn more in Part 1.
PCORnet, the National Patient-Centered Clinical Research Network, is an innovative initiative of the Patient-Centered Outcomes Research Institute (PCORI). The goal of PCORnet is to improve the nation’s capacity to conduct clinical research by creating a large, highly representative network that directly involves patients in the development and execution of research. More information is available at www.pcornet.org.
The Patient-Centered Outcomes Research Institute (PCORI) is an independent nonprofit organization 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. PCORI is committed to continuously seeking input from a broad range of stakeholders to guide its work. More information is available at www.pcori.org.