CONSTRUCTING REPRODUCTIVE HISTORIES BY LINKING VITAL RECORDS

Certificates of 1,449,287 live births and fetal deaths filed in Georgi a from 1980 through 1992 were linked to create chronologies that, excl uding induced abortions and ectopic pregnancies, constituted the repro ductive experience of individual women. The authors initially used a d eterministic method (whereby linking rules were not based on probabili ty theory) to link as many records as possible, knowing that some of t he linkages would be incorrect. They subsequently used a probabilistic method (whereby evaluation of linkages was developed from probability theory) to evaluate each linkage, and they broke those that were judg ed to be incorrect. Of the 1.4 million records, 38% did not link to an other record. From the remaining records, 369,686 chains of two or mor e events were constructed. The longest chain included 12 events. Of th e chains, 69% included two events; 22% included three events. Longer c hains tended to have lower scores for probable validity. The probabili ty-based evaluation of chains affected 3.0% of the records that had be en in chains at the end of the deterministic linkage. A greater percen tage of records in longer chains were affected by the evaluation. Unfo rtunately, the small subset of records that were the most difficult to link tended to overrepresent groups with the greatest risk of adverse pregnancy outcomes. Researchers contemplating a similar linkage can a nticipate that, for the majority of records, linkage can be accomplish ed with a relatively straightforward, deterministic approach.