Results Summary
PCORI funded the development of PCORnet®, the National Patient-Centered Clinical Research Network, to make research faster, easier, and less costly to conduct. PCORnet is made up of Partner Networks of healthcare systems, patients and communities, and health plans that harness the power of large amounts of health data.
PCORI supports projects, designed in partnership with federal funding organizations, to improve the quality of data used in clinical research. This project is one of two projects designed to link data from PCORnet and the U.S. Food and Drug Administration’s (FDA) Sentinel medical product monitoring system.
What was the project about?
Antimicrobials are medicines that help fight infections from bacteria or fungi. Doctors and hospitals often prescribe them when they aren’t needed. This overuse leads to problems, such as the medicines not working, because the bacteria they are meant to kill become resistant.
Tracking how often hospitalized patients receive antimicrobials is an important part of preventing resistance to a medicine. The Centers for Disease Control and Prevention, or CDC, works with hospitals to track and understand data on antimicrobial use.
One way to help hospitals track antimicrobial use is through databases that combine information from sources like patients’ health records or insurance claims. PCORnet created a shared database system that includes data from patients’ electronic health records, or EHRs. Through its Sentinel Initiative, the U.S. Food and Drug Administration, or FDA, created a shared database system with data from insurance claims. This system is called the Sentinel System. The FDA uses this system to monitor the safety of medical products such as medicines and vaccines. But neither of these systems had a way to help hospitals track antimicrobial use to share with the CDC. The project team wanted to create and test a new method for using the PCORnet and the FDA shared database systems to track antimicrobial use in the format used by the CDC.
What were the results?
Were data in the format needed by CDC? The project team found that the databases did not have information in the format needed by CDC. The databases needed a new format for information on the antimicrobial medicine prescribed and how patients took the medicine. The databases also needed a new format for information about patients’ locations as they moved through the hospital during their stays, such as from the emergency room to a hospital bed.
Could the team develop a new method to track antimicrobial use? The project team compiled the information from the databases on antimicrobial use into the format needed by CDC. The team created a free toolkit to help healthcare systems that are part of PCORnet and the FDA’s Sentinel Initiative track antimicrobial use and share these data with CDC.
Who was in the project?
The project included EHR data from five healthcare systems in Illinois, Louisiana, and Tennessee. The healthcare systems belong to three clinical data research networks that are part of PCORnet. Healthcare systems provided up to 12 months of data on patients’ antimicrobial use.
What did the project team do?
The project team looked at
- What data each database had to track antimicrobial use
- What data the project team needed to find in EHRs
- Whether a new method could track antimicrobial use by combining data from the databases
What were the limits of the project?
Some healthcare systems had restrictions that limited the data that they could share with PCORnet. This may limit the number of PCORnet partners that can use the toolkit.
Future projects can use the toolkit to track antimicrobial use for health conditions for which medicine misuse is common.
How can people use the results?
Research teams can use these methods when planning studies that use the PCORnet database to track hospitalized patients’ antimicrobial use.
Professional Abstract
PCORI funded the development of PCORnet®, the National Patient-Centered Clinical Research Network, to make research faster, easier, and less costly to conduct. PCORnet is made up of Partner Networks of healthcare systems, patients and communities, and health plans that harness the power of large amounts of health data.
PCORI supports projects, designed in partnership with federal funding organizations, to improve the quality of data used in clinical research. This project is one of two projects designed to link data from PCORnet and the U.S. Food and Drug Administration’s (FDA) Sentinel medical product monitoring system.
Objective
To expand the capabilities of PCORnet and the FDA’s Sentinel System to measure antimicrobial use by hospitalized patients at the patient care unit level
Project Design
Design Elements | Description |
---|---|
Design | Developing and testing a system to measure and aggregate antimicrobial use by hospitalized patients at the patient care unit level |
Data Sources and Data Sets | Records from five healthcare systems from PCORnet Partner Networks |
Analytic Approach | Data characterization process to standardize information needed to capture hospitalized patients’ time in different hospital units and medication administration dates and times |
Developed an open-source toolkit that can be integrated into the PCORnet and FDA CDMs. The toolkit includes a final data model, a data dictionary, queries to help local institutions assess the quality of their data, queries to generate antimicrobial use data, and a table structure to enable the queries to work on an extended data model. |
The FDA’s Sentinel System is an active surveillance system that identifies and assesses medical product safety issues. Both PCORnet and the Sentinel System have common data models (CDMs) that organize data into a standard structure for use by researchers. The project team identified antimicrobial use for hospitalized patients as an important domain for improving the compatibility of data from the PCORnet and Sentinel CDMs.
The Centers for Disease Control and Prevention (CDC) supports national reporting of antimicrobial use through its National Healthcare Safety Network (NHSN), which works with hospitals to track and understand data on antimicrobial use. Hospitals can use these data to support appropriate antimicrobial use, prevent antimicrobial resistance, and improve patient care. However, the PCORnet and Sentinel CDMs did not have the technical infrastructure needed to help hospitals and healthcare systems in their networks track and submit antimicrobial use data to CDC’s NHSN.
The project team worked with five healthcare systems from PCORnet Partner Networks in Illinois, Louisiana, and Tennessee to develop and test a system to enable hospitals and health systems to prepare antimicrobial use data for submission to CDC’s NHSN. Each healthcare system provided the team with up to 1 year of data on patients’ antimicrobial use.
CDC’s NHSN requires hospitals to report patients’ locations in a standard format called patient care units, which include the emergency department, ambulatory care, inpatient care, or observation unit. The project team used the CDC’s NHSN guidelines to determine the existing data elements from each CDM. The team found that the CDMs did not have standardized data elements on within-hospital patient movement and medication administration to allow hospitals and health systems to track and submit antimicrobial use data to CDC’s NHSN.
The project team developed a system to standardize data on patients’ antimicrobial use during hospital stays. These data included patients’ locations as they moved through the hospital, such as between inpatient unit types, and medication administration, which when joined to patient location, allowed the team to quantify antimicrobial use by location. After collecting the data, the team used queries to check that the data appeared correct and to aggregate the data in the correct format for submission to CDC’s NHSN.
Results
The system produced the metrics required for reporting patient antimicrobial use data to CDC’s NHSN. As a result, the project team developed an open-source toolkit that can be integrated into the PCORnet and Sentinel CDMs. The toolkit includes a final data model, a data dictionary, queries to help organizations assess the quality of their data, queries to generate the antimicrobial use data, and a table structure to enable the queries to work on an extended data model. The team has incorporated a medication administration table essential for the reporting into the PCORnet CDM.
Limitations
Individual hospitals and health systems need to map their unit types to the codes required by CDC; however, most hospitals in the United States have these mappings through their infection control departments. Patient movement has not yet been incorporated into the PCORnet CDM. Due to variability between and within vendors’ proprietary systems, local code mappings are not usually interoperable; thus, local mapping and confirmation of codes is required.
Conclusions and Relevance
The open-source toolkit can increase the technical capacity of additional PCORnet and FDA Sentinel partner networks to report data to CDC’s NHSN and track antimicrobial use.
Future Needs
Future projects can demonstrate the application of the open-source toolkit for antimicrobial use monitoring in other settings or for other research goals.