Project Details - Ongoing
- Grant Number:R01 HS023808
- Funding Mechanism:
- AHRQ Funded Amount:$1,701,413
- Principal Investigator:
- Project Dates:7/1/2017 to 4/30/2022
- Care Setting:
- Health Care Theme:
Integrated healthcare data from a broad set of sources is required for many purposes, including assuring high-quality care delivery and enabling patient-centered outcomes research. While each clinical encounter with the health system generates information, this information is stored in numerous independent clinical repositories with no single unique identifier to integrate the data sources. Fragmentation of clinical information prevents clinicians from having access to comprehensive patient health information, potentially risking patient safety. Therefore, effective evidence-based patient information matching methods are needed to maximize the accuracy and completeness of healthcare data.
The goals of this project are to implement and evaluate emerging consensus-based recommendations for best practices to match health information sources at the patient level in order to provide evidence to meaningfully inform next steps in the formulation of a national patient identity management strategy. Previously, the team embedded patient matching algorithms in the Nation’s largest health information exchange, which contains thousands of diverse operational and clinical data sources. In the current project, the team will evaluate the impact of the algorithms on the quality, standardization, and discriminating power of data collected in healthcare settings.
The specific aims of this project are as follows:
- To implement three general classes of recommended matching data enhancements and measure the resulting matching accuracy improvements.
- To implement four novel matching algorithm enhancements and assess the resulting matching accuracy improvements.
- To measure the matching accuracy improvements resulting from using combinations of (1) three best practice matching policy recommendations and (2) four novel matching algorithm enhancements.
The team will use randomly sampled, manually reviewed clinical data sets to evaluate the matching accuracy of the current algorithm, and the relative improvements resulting from the algorithm enhancements. Evaluative measures will include the following algorithm calibration statistics: sensitivity, specificity, positive predictive value, and area under the receiver operating characteristic curve. This project has the potential to advance strategies for matching patient data to generate comprehensive clinical information to support patient care and patient-centered outcomes research.