|
Activity Number:
|
239
- The 2020 FCSM Data Quality Framework: Future Directions
|
|
Type:
|
Topic-Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #317346
|
|
|
Title:
|
Expanding Data Linkage Opportunities and Measuring Quality
|
|
Author(s):
|
Lisa B. Mirel* and Dean Resnick and Christine Cox and Jonathan Aram
|
|
Companies:
|
CDC/NCHS/DAE and NORC and NORC and CDC/NCHS/DAE
|
|
Keywords:
|
National Center for Health Statstics;
Data Linkage Program;
privacy preserving record linkage
|
|
Abstract:
|
The Data Linkage Program at the National Center for Health Statistics is exploring ways to expand data linkage opportunities, including assessing the ability to conduct privacy preserving record linkages (PPRL). PPRL methods do not require the exchange of personally identifiable information but rather rely on hashing algorithms that increase privacy protections. In order to assess the quality of a PPRL based linkage, comparisons were made between the linked data sets produced using the standard approach with unencrypted identifiers against the PPRL approach. This methodological assessment included an evaluation of how the use of hashing algorithms may affect the quality of the linked data and further affect the inference in secondary analysis based on those data. By assessing the hashing algorithms with a “truth” source, this uncertainty was quantified by measuring linkage errors. The presentation will conclude with next steps and potential uses for PPRL in future studies.
|
Authors who are presenting talks have a * after their name.
