Related Experiment Videos
Record linkage strategies: Part II. Portable software and deterministic matching
A Wajda1, L L Roos, M Layefsky
1Manitoba Centre for Health Policy and Evaluation and Faculty of Medicine, University of Manitoba, Winnipeg, Canada.
Methods of Information in Medicine
|August 1, 1991
Summary
This study introduces LINKS, a flexible and cost-effective software for record linkage. It supports both deterministic and probabilistic methods, aiding researchers in data quality assessment and linkage strategy development.
Area of Science:
- Health Informatics
- Computer Science
- Biostatistics
Background:
- Effective record linkage is crucial for health research and data analysis.
- Existing software may lack portability, flexibility, or cost-efficiency.
- Need for adaptable tools to handle diverse data linkage challenges.
Purpose of the Study:
- To describe the LINKS software package for record linkage.
- To highlight its portability, flexibility, and low-cost features.
- To demonstrate its utility in real-world data projects.
Main Methods:
- Development of a modular software package (LINKS).
- Implementation of tools for deterministic and probabilistic linkage.
- Inclusion of data quality and structure assessment modules.
- User-defined weights and decision rules for customized strategies.
Main Results:
- LINKS demonstrated portability, flexibility, and cost-effectiveness.
- Successfully identified mortality in California AIDS data using deterministic approaches.
- Facilitated comparisons with alternative software and linkage strategies.
- Highlighted potential cost savings by avoiding manual data entry.
Conclusions:
- The modular design of LINKS enhances control and allows unique linkage strategies.
- Flexible software and deterministic approaches can optimize data linkage efficiency.
- LINKS provides a valuable tool for researchers performing complex record linkage tasks.