Evaluating the quality of anonymous record linkage using deterministic procedures with the New York State AIDS registry and a hospital discharge file.

Linkage of same-person records across multiple databases relies on high-quality, uniformly available identifying information. These data quality issues become increasingly important when personal names are not available for record linkage. Using deterministic decision criteria, we linked records from two population-based files in the absence of personal names. The sensitivity of anonymous record linkage procedures ranged from 32 to 85 per cent for the two years studied, and the positive predictive value (PPV) ranged from 14 to 99 per cent. Decreasing sensitivity and PPV were primarily attributed to (1) errors in computerized identifying information and (2) the deterministic decision criteria specified for record linkage. An evaluation of the contribution of personal names to the quality of record linkage found no measurable impact.

[1]  A Wajda,et al.  The art and science of record linkage: methods that work with few identifiers. , 1986, Computers in biology and medicine.

[2]  H. Storm,et al.  Completeness of cancer registration in Denmark 1943-1966 and efficacy of record linkage procedures. , 1988, International journal of epidemiology.

[3]  J. Kobayashi,et al.  Hospital charges for people with AIDS in Washington State: utilization of a statewide hospital discharge data base. , 1988, American journal of public health.

[4]  H B NEWCOMBE,et al.  Automatic linkage of vital records. , 1959, Science.

[5]  M E Fair,et al.  Application of exact ODDS for partial agreements of names in record linkage. , 1991, Computers and biomedical research, an international journal.

[6]  Howard B. Newcombe,et al.  Handbook of record linkage: methods for health and statistical studies, administration, and business , 1988 .

[7]  P. Meis,et al.  A comparison of women in and out of a prematurity prevention project in a North Carolina perinatal care region. , 1988, American journal of public health.

[8]  M E Fair,et al.  Discriminating powers of partial agreements of names for linking personal records. Part I: The logical basis. , 1989, Methods of information in medicine.

[9]  H B Newcombe,et al.  Use of the Canadian Mortality Data Base for epidemiological follow-up. , 1982, Canadian journal of public health = Revue canadienne de sante publique.

[10]  A Wajda,et al.  Record Linkage Strategies: Part II. Portable Software and Deterministic Matching , 1991, Methods of Information in Medicine.

[11]  J. Neaton,et al.  An evaluation of the Social Security Administration master beneficiary record file and the National Death Index in the ascertainment of vital status. , 1983, American journal of public health.

[12]  A Wajda,et al.  Record Linkage Strategies , 1991, Methods of Information in Medicine.

[13]  P. Diehr,et al.  The use of large data bases in health care studies. , 1987, Annual review of public health.

[14]  R E Smith,et al.  The California Automated Mortality Linkage System (CAMLIS). , 1984, American journal of public health.

[15]  M E Fair,et al.  Discriminating powers of partial agreements of names for linking personal records. Part II: The empirical test. , 1989, Methods of information in medicine.

[16]  H. Shannon,et al.  Comparison of individual follow-up and computerized record linkage using the Canadian Mortality Data Base. , 1989, Canadian journal of public health = Revue canadienne de sante publique.