Effects of Record Linkage Errors on Disease Registration

Abstract:

Reliable record linkage is a prerequisite for high-quality population-based disease registration. Rapid developments in computer processing have made record linkage both more efficient and more reliable in recent years. At the same time, concerns about confidentiality increasingly hinder record linkage in many disease registries. This paper provides basic algebraic models describing the effects of record linkage errors on monitoring disease incidence. Homonym errors, that is, erroneous linkage of records that pertain to distinct individuals, lead to underestimation of incidence in the registry population. The degree of underestimation strongly depends on the discriminating power of personal identifiers and the record linkage procedure on the one hand, and the number of registered cases on the other hand. Synonym errors, that is, failure to link notifications on the same individual, lead to overestimation of incidence in the population base. The combined effects of record linkage errors are illustrated with empirical examples. We conclude that it is the largest and most informative disease registries that are potentially affected most by impediments of record linkage procedures due to unduly restrictive confidentiality rules.

Keywords

Disease Registries - Epidemiology - Incidence - Record Linkage

• REFERENCES

• 1 Newcombe HB. Handbook of Record Linkage. Methods for Health and Statistical Studies, Administration, and Business. Oxford – New York – Tokyo: Oxford University Press; 1988
• 2 Fellegi IP, Sunter AB. A theory for record linkage. J Am Stat Assoc 1969; 64: 1183-210.
• 3 Jaro M. Advances in record linkage methodology as applied to matching the 1985 census of Tampa, Florida. J Am Stat Assoc 1989; 84: 414-20.
• 4 Gill L, Goldacre M, Simmons H, Bettley G, Griffith M. Computerised linking of medical records: methodological guidelines. J Epidemiol Comm Health 1993; 47: 316-9.
• 5 Howe GR, Lindsay J. A generalized record linkage computer system for use in medical follow-up studies. Comput Biomed Res 1981; 14: 327-40.
• 6 Arellano MG, Petersen GR, Petitti DB, Smith RE. The California Automated Mortality System (CAMLIS). Am J Publ Health 1984; 74: 1324-30.
• 7 Wajda A, Roos LL. Simplifying record linkage: Software and strategy. Comp Biol Med 1987; 17: 239-48.
• 8 Langley JD, Botha JL. Use of record linkage techniques to maintain the Leicestershire Diabetes Register. Comp Meth Progr Biomed 1994; 41: 287-95.
• 9 Carpenter M, Fair ME. The November 15 1991 workshop on record linkage methodology. Health Reports 1992; 4: 84-8.
• 10 Match ware Technologies, Inc. AUTOMATCH Generalized Record Linkage System. Silver Spring, MD; 1992
• 11 Jaro MA. Probabilistic linkage of large public health data files. Stat Med 1995; 14: 491-8.
• 12 Clarke EA, Marrett LD, Kreiger N. Appendix 3 (c) Cancer registration in Ontario: a computer approach. In: Jensen OM, Parkin DM, MacLennan R, Muir CS, Skeet RG. eds. Cancer Registration. Principles and Methods. LARC Scientific Publications No. 96. Lyon: International Agency for Research on Cancer; 1991
• 13 Häfner H, Pfeifer-Kurda M. The impact of data protection laws on the Mannheim case register. In: Psychiatric Case Registers in Public Health: a Worldwide Inventory 19601985. Von Horn GHMM, Geil R, Gulbinat WH, Henderson JH. eds. Amsterdam: Elsevier; 1986: 366-71.
• 14 Knox EG. Confidential medical records and epidemiological research. BMJ 1992; 304: 727-8.
• 15 Westrin C-G, Nilstun T. The ethics of data utilization: a comparison between epidemiology and journalism. BMJ 1994; 308: 522-3.
• 16 Schmidtmann I, Michaelis J. Untersuchungen zum Record Linkage für das Krebsregister Rheinland-Pfalz (Technical Report, in German). Tumorzentrum Rheinland-Pfalz; Mainz: 1994
• 17 Fett MJ. The development of matching criteria for epidemiological studies using record linkage techniques. Int J Epidemiol 1984; 13: 351-5.
• 18 Storm HH. Completeness of cancer registration in Denmark 1943-1966 and efficacy of record linkage procedures. Int J Epidemiol 1988; 17: 44-9.
• 19 van den Brandt PA, Schouten LJ, Goldbohm RA, Dorant E, Hunen PMH. Development of a record linkage protocol for use in the Dutch cancer registry for epidemiological research. Int J Epidemiol 1990; 19: 553-8.
• 20 Calle EE, Terrell DD. Utility of the National Death Index for ascertainment of mortality among cancer prevention study II participants. Am J Epidemiol 1993; 137: 235-41.
• 21 Coleman M, Estève J, Damiecki P, Arslan A, Renard H. Trends in Cancer Incidence and Mortality. IARC Scientific Publications No. 121 Lyon: International Agency for Research on Cancer; 1993
• 22 Parkin DM, Muir CS, Whelan SL, Gao Y-T, Ferlay J, Powell J. eds. Cancer Incidence in Five Continents, Volume VI. IARC Scientific Publications No 120 Lyon: International Agency for Research on Cancer; 1992
• 23 Brenner H, Schmidtmann I. Determinants of homonym and synonym rates of record linkage in disease registration. Meth Inform Med 1996; 35: 19-24.
• 24 Roos LL, Wajda A, Sharp SM, Nicol JP. Software for health care analysts: a modular approach. J Med Syst 1987; 11: 445-64.
• 25 Newcombe HB, Fair ME, Lalonde P. Discriminating powers of partial agreements of names for linking personal records. Part I: The logical basis. Meth Inform Med 1989; 28: 86-91.
• 26 Newcombe HB, Fair ME, Lalonde P. Discriminating powers of partial agreements of names for linking personal records. Part II: The empirical test. Meth Inform Med 1989; 28: 92-6.