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

Probabilistic record linkage and an automated procedure to minimize the undecided-matched pair problem

“…The next step was to select best link(s), defined as the linked pair with the highest combined weight, achieved by each record 6 . One problem is the failure to match a death record with its corresponding birth record, which yields non-matched records.…”

Record Linkage in the Cancer Registry of Tyrol, Austria

“…Oneo ft he keyd ecisions during implementation wastochoosespecificvaluesfor the critical bounds p 1 andp 2 .I no rdert o evaluate thisd ecision, one must bear in mind the consequences of false-positiveand false-negative decisions [12][13][14][15][16][17].…”

The EpiLink Record Linkage Software

“…We have developed ar elativelys imple probabilistic data linkagep rogram,c alled EpiLink, using as traightforwardm athematical approach that can be understood by non-expert users. Ouraim wastoproducea program that is cost-effective in termsofcomputer andhuman resources, andthatcan be used as an alternativetomore complex probabilistic record linkage programs [12][13][14][15] Thec hoice of information itemsa nd statistical parameters to useiscritical to the success of probabilisticl inkage. Running the linkagew ithout ap reliminary analysis of the information in the databases is likely to result in alarge number of erroneous linkages.…”