Discriminating Powers of Partial Agreements of Names for Linking Personal Records

Abstract

Abstract:Machines have difficulty when using people’s names to link medical and other records pertaining to the same individuals because of nicknames, ethnic synonyms, truncations, misspellings and typographical errors. Present algorithms used to compute the discriminating powers (or ODDS) associated with partial agreements of names are based, inappropriately, on the degrees of outward similarity alone. They are particularly ineffective in dealing with names that look alike but are unrelated, and with related names that have little apparent similarity. A fundamentally different rationale is, therefore, proposed which, like the human mind, assesses the relatedness of two alternative forms of a name in terms of how often they are used, interchangeably in practice. This must be taken into account if the associated discriminating powers (ODDS) are to be correctly computed. A way of implementing this more precise approach is described and illustrated, using the given names on linked records from an earlier epidemiological study. This first study of two describes the logical basis for record linkage, a second one the empirical test.

Enhancing the Power of Record Linkage Involving Low Quality Personal Identifiers: Use of the Best Link Principle and Cause of Death Prior Likelihoods

“…The prior likelihoods for matches and nonmatches were recognized as early as 1969 by Fellegi and Sunter in their mathematical theory of record linkage (10), although the prior likelihoods have not been explicitly included in the calculation of scores until recent years (11,12). The specific inclusion of a term reflecting the prior likelihood is unnecessary where the underlying probability of a match being found is essentially equal to one, as in the Heartstart linkage.…”

Record Linkage Strategies

“…In other situations, the probability of a match between two data sets will be quite high, and relatively little error may be present in the variables used for linkage. We need guidelines for when simpler techniques can be applied; as noted in Newcombe, Fair, and Lalonde [9,10], accuracy versus simplicity will often be an important tradeoff.…”

Discriminating Powers of Partial Agreements of Names for Linking Personal Records

“…For the partial agreements that repeat, value specific ODDS may be calculated using either formula (1) or formula (2) from [1]:…”

“…In a preceding paper [1] a rationale was described whereby the discriminating powers, or ODDS, associated with partial agreements of people's names may be computed precisely for the purposes of record linkage, independent of their outward similarities or lack thereof (Newcombe, Fair and Lalonde [1]). This is a fundamentally new approach which, unlike any earlier algorithm, is exact inasmuch as the data to support it are available.…”