Security and Privacy Standards for Patient Matching, Linking and Aggregation

Linking and aggregating patient data to improve patient safety and privacy.
Agency
  • Office of the National Coordinator for Health Information Technology (ONC) 
Start Date
  • 6/16/2015

 

Functionality
  • Linking of Clinical and Other Data for Research
  • Use of Clinical Data for Research

 

STATUS: Active Project

BACKGROUND

Linking and aggregating patient data from disparate sources cannot be done without “matching” these data in a secure manner that protects patent privacy. Effective patient matching allows users of the data (whether patients, providers, or researchers) to draw correct inferences using the data that have been linked together. Without accurate patient matching, stakeholders may inadvertently draw inaccurate conclusions that could significantly impair patient safety and privacy.

PROJECT PURPOSE & GOALS

Under the Office of the National Coordinator for Health Information Technology (ONC), this project will identify the best patient attributes to address the challenge of linking patients’ data across research, clinical, and claims data sets in order to support the patient-centered outcomes research (PCOR) data infrastructure that enables standardization and sharing of patient data across organizations. This project includes work along five distinct tracks.

The project objectives are to:

  • Improve data quality by standardizing patient attributes and algorithms that can be used to reliably perform patient matching across clinical and claims data sets to improve algorithm match rates.

  • Create an open source visual tool for patient matching and aggregation.

  • Create a privacy and security application programming interface (API) or PCOR infrastructure security “layer”.

  • Include clinical data research networks in the piloting and testing of the proposed standards and services.

  • Integrate the National Plan and Provider Enumeration System (NPPES) provider identification as an additional attribute to improve patient matching across sources.