Linking Health Care and Vital Statistics Databases
Author Information
Author(s): Li Bing, Quan Hude, Fong Andrew, Lu Mingshan
Primary Institution: University of Calgary
Hypothesis
What is the best way to achieve a high linkage rate in a cost-effective manner?
Conclusion
The combination of surname, sex, and date of birth is the optimal choice for deterministic record linkage.
Supporting Evidence
- The highest linkage rate was 88.0% using surname, sex, and date of birth.
- The correct linkage rate was 96.9% for the same combination.
- Linkage rates varied by age but not by sex.
Takeaway
This study looked at how to connect different health records to find out who died and why. They found that using a person's last name, gender, and birthday works best.
Methodology
Deterministic record linkage was used with combinations of personal identifiers to assess linkage rates.
Potential Biases
Potential selection bias due to missing personal health numbers, especially in younger age groups.
Limitations
The study's results may not be generalizable to other regions or countries due to variations in data quality.
Participant Demographics
Residents of Calgary Health Region, Alberta, Canada, during fiscal years 1998/99 through 2001/02.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website