Locating Members of a Cohort Identified Retrospectively From Limited Data in 50-Year-Old Records: Successful Approaches Employed by the Hanford Thyroid Disease Study

Purpose Long-term follow-up studies sometimes rely on retrospective identification of the cohort to be followed. Few such studies have been conducted, however, that rely on very old source records, perhaps because of concern that identification and location of a cohort based on very old records would not be feasible. Methods The Hanford Thyroid Disease Study (HTDS), conducted in the 1990s, identified 5199 cohort members using Washington state birth records from 1940–1946. The limited information contained therein was used to trace cohort members to the present day, nearly 50 years later. Results We found the best strategy to locate individuals efficiently is to use a combination of methods and resources in a phased approach, beginning with readily available and easy-to-use sources of information before employing more time-intensive strategies that are costly and rely on information that is difficult to obtain. Motor vehicle licensing records and directories were the most useful individual sources of information. Using this approach, the HTDS successfully located 94% of the 5199 cohort members identified. Conclusions It is feasible to successfully trace and locate a very high proportion of individuals identified from very old records that contain little information.