Abstract

Block et al. (I) reported a method to estimate year of birth using the Social Security number in cohort studies in which the age of a significant portion of cases was unknown. Social Security numbers have proven useful to us in a different way in a long-term follow-up study of childhood lead poisoning. We have been following cases diagnosed up to 68 years ago, and for many of these cases it has been decades since their vital status was estab-lished. We were able to obtain a Social Security number for most of these individuals but had difficulty ascertaining their vital status at any point after they were lost to medical follow-up. By using the tables and extrapolations provided by Block et al., we identified year and state of issue of each Social Security number. We knew, therefore, cases to be alive up to the date of issue, and we knew in which state to concentrate follow-up efforts. This method increased follow-up for 60 (13 percent) of the cases. On average, 14 years of follow-up were gained for each of the 60 cases, and total years of follow-up were increased 6 percent for the study group overall. The greatest gains in years of follow-up were for female cases who did not obtain Social Security numbers until middle age. This application of Social Security numbers may be of use to other investigators involved in retrospective follow-up studies, par-ticularly studies of childhood diseases where age of onset occurs prior to the age of obtaining a