• Print

released: 23 Feb 2011 author: AIHW

In 2005, the Institute was funded to create a linked aged care database to enable analysis of pathways through aged care services. The linkage strategy for the project involved using a Statistical Linkage Key (SLK) because of the lack of either a name or a common person identifier on the data sets being linked. This paper validates the results obtained using the SLK linkage strategy by comparing it directly with a name-based linkage strategy.

ISSN 1833-1238; ISBN 978-1-74249-124-0; Cat. no. CSI 11; 54pp.; INTERNET ONLY

Full publication

Publication table of contents

  • Preliminary material
    • Title and verso page
    • Contents
    • Acknowledgments
    • Abbreviations
    • Summary
      • Methods
      • Results
     
  • Body section
    1. Introduction
      • 1.1 Data
        • ACCMIS data
        • National Death Index data
         
      • 1.2 Report structure
       
    2. Linkage Strategies
      • 2.1 Types of data linkage
      • 2.2 SLK-based linkage
        • The strategy
          • Step 1: Defining the structure of the match keys
          • Step 2: Incorporating cross-program use
          • Step 3: Identification of keys to use in matching
          • Step 4: Stepwise matching using selected match keys
           
        • Results
         
      • 2.3 Name-based linkage strategy
        • The strategy
          • Step 1: Run pass
          • Step 2: Output pairs of matches with weights
          • Step 3: Clerical review
          • Step 4: Enact clerical review decisions
           
        • Results
       
    3. A comparison of two linkage strategies
      • 3.1 Type of link comparisons
        • Perspective
         
      • 3.2 Comparing links
        • NDI perspective
        • ACCMIS perspective
        • Two-way ACCMIS–NDI link comparison
         
      • 3.3 Analysis of SLK-based contested, missed and false links
        • Contested (mixed) links
          • ACCMIS perspective
          • NDI perspective
           
        • SLK missed links
          • Identifying additional keys
            • Possible additional keys to use
            • Keys that would incorporate too many false links
             
          • Data quality – missing postcode on the NDI
           
        • SLK false links
         
      • 3.4 Overall link quality
        • Direct estimates
        • Refined estimates
         
      • 3.5 Summary
       
    4. Analysis comparisons
      • 4.1 Age and sex
      • 4.2 Cause of death
      • 4.3 State and territory at death
      • 4.4 Summary
     
  • End matter
    • Appendix 1: Data
      • A1.1 Aged and Community Care Management Information System
        • Data structure
        • Identifying ACCMIS clients
        • Data for linkage
          • Data quality
         
      • A1.2 National Death Index
        • Data structure
        • Data for linkage
          • Duplicate records
          • Data quality
       
    • References
    • List of tables
    • List of figures

Recommended citation

AIHW 2011. Comparing an SLK-based and a name-based data linkage strategy: an investigation into the PIAC linkage. Data linkage series no. 11. Cat. no. CSI 11. Canberra: AIHW. Viewed 1 September 2014 <http://www.aihw.gov.au/publication-detail/?id=10737418279>.