Project Summary

One of PCORI’s goals is to improve the methods that researchers use for patient-centered outcomes research. PCORI funds methods projects like this one to better understand and advance the use of research methods that improve the strength and quality of comparative effectiveness research.  

This research project is in progress. PCORI will post the research findings on this page within 90 days after the results are final.

What is the project about?

Patient health data are often scattered and incomplete. Linking records from multiple sources can help researchers collect complete information on patients and improve the quality of data for research.

Record linkage is the process of finding pieces of information on the same person across different large sets of data, such as disease registries and other medical records. But existing methods pose security and privacy risks. They may also be hard to implement and time consuming. Also, using existing methods, researchers can measure the accuracy of their process only after records have been linked.

In this study, the research team is developing and testing new methods to check if two data sets can be linked together. The team is also looking to securely link new data to existing patient data without identifying individual patients.

How can this project help improve research methods?

Results may help researchers link new data to existing patient data more easily while protecting private information.

What is the research team doing?

To develop the new methods, the research team is using two data sets of patient records. First, the team is creating measures that describe the data to see if the two data sets can be linked. The team is also developing new methods to link only new data to existing patient data, rather than linking full data sets, which can take a long time.

To measure the accuracy of the new methods, the research team first is using the new methods with real patient data to create a linked data set in which patients can’t be identified. Then the team is creating a linked data set by hand-matching data in which patients can be identified. The team is then comparing the linked data set that doesn’t contain private information with the one that does. Also, the team is comparing the quality of the data before and after linkage to measure the impact of the methods on data quality.

Research methods at a glance

Design Element Description
Goal
  • Develop methods to assess linkability of a data set
  • Develop and evaluate an incremental privacy-preserving record linkage method
  • Evaluate the performance of the record linkage method using human verified linkages
  • Measure changes in data quality of pre-linkage and post-linkage data sets to understand the impact of record linkage
Approach Modeling, intrinsic and distributional data characterization, manual records review

Project Information

Toan Ong, PhD
Anschutz Medical Campus - University of Colorado Denver
$928,262
Incremental Privacy-Preserving Record Linkage (iPPRL) to Reduce Barriers to Data Sharing and Improve Data Quality^

Key Dates

November 2018
August 2022
2018

Study Registration Information

^This project was previously titled: Incremental Privacy-Preserving Record Linkage to Improve Data Quality

Tags

Award Type
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 15, 2022