Download PDF
Methods Inf Med 1998; 37(01): 64-68
DOI: 10.1055/s-0038-1634494
Original Article
Schattauer GmbH

Best-Link Matching of Scottish Health Data Sets

S. W. Kendrick
1   Information and Statistics Division, NHS in Scotland, Scotland
,
M. M. Douglas
2   National Health Service Central Register for Scotland, Scotland
,
D. Gardner
1   Information and Statistics Division, NHS in Scotland, Scotland
,
D. Hucker
3   3CSC Scotland Healthcare, Scotland
› Author Affiliations
Further Information

Publication History

Publication Date:
07 February 2018 (online)

    Permissions and Reprints

Abstract:

Methods are described used to link the Community Health Index and the National Health Service Central Register (NHSCR) in Scotland to provide a basis for a national patient index. The linkage used a combination of deterministic and probability matching techniques. A best-link principle was used by which each Community Health Index record was allowed to link only to the NHSCR record with which it achieved the highest match weight. This strategy, applied in the context of two files which each covered virtually the entire population of Scotland, increased the accuracy of linkage approximately a thousand-fold compared with the likely results of a less structured probability matching approach. By this means, 98.8% of linkable records were linked automatically with a sufficient degree of confidence for administrative purposes.

Keywords

Medical Record Linkage - Community Health Planning - Probability Theory - Registries
 

  • REFERENCES

  • 1 Sayers RC. Strategy to achieve a national-unique patient identifier. CI News No 5 July/August 1994 (National Health Service in Scotland Management Executive Health Systems Directorate).
    PubMedGoogle Scholar
  • 2 Womersley J. The public health uses of the Scottish Community Health Index (CHI). J Public Health Med 1996; 18: 465-72.
    CrossrefPubMedGoogle Scholar
  • 3 Hedley AJ, and McMaster W. Use of the National Health Service Central Register for Medical Research Purposes. Health Bulletin (Edinburgh) 1988; 46: 63-8.
    PubMedGoogle Scholar
  • 4 Kendrick SW, and Clarke JA. The Scottish Record Linkage System. Health Bulletin (Edinburgh) 1993; 51: 72-9.
    PubMedGoogle Scholar
  • 5 Kendrick S, and Mcllroy R. One-pass linkage – rapid creation of patient-based data. Current Perspectives in Healthcare Computing: Healthcare Computing 1996 Conference Proceedings..
    PubMedGoogle Scholar
  • 6 Newcombe HB. Handbook of Record Linkage. New York: Oxford University Press; 1988
    Google Scholar
  • 7 Fair ME. An overview of record linkage in Canada. Proceedings of the Social Statistics Section of the American Statistical association. 1995
    PubMedGoogle Scholar
  • 8 Baldwin JA, Acheson ED, Graham WJ eds. Textbook of Record Linkage. Oxford: Oxford University Press; 1988
    Google Scholar
  • 9 Heasman MA, and Clarke JA. Medical record linkage in Scotland. Health Bulletin (Edinburgh) 1979; 37: 97-103.
    PubMedGoogle Scholar
  • 10 Kendrick S. The development of record linkage in Scotland: the responsive application of probability matching. Proceedings of the 1997 Record Linkage Workshop. Washington D.C.: March 20-21st 1997. (In press).
    Google Scholar
  • 11 Newcombe HB. Age-related bias in probabilistic death searches due to neglect of the “prior likelihoods”. Computers and Biomedical Research 1995; 28: 87-99.
    CrossrefPubMedGoogle Scholar
  • 12 Newcombe HB, Fair ME, Lalonde P. Discriminating powers of partial agreements of names for linking personal records. Meth Inform Med 1989; 28: 86-96.
    PubMedGoogle Scholar
  • 13 Newcombe HB, Fair ME, and Lalonde P. The use of names for linking personal records. Journal of the American Statistical Association 1992; 87: 1193-208.
    CrossrefPubMedGoogle Scholar