To assess the role of observational data bases in technology assessment, we examined 26 articles from the Framingham Heart Study that evaluated a technology, therapy, or predictive instrument. These assessments were grouped into four categories: (a) the study of a technology voluntarily in use by the cohort, (b) the application of an external technology to members of the cohort, (c) the use of the Framingham results to evaluate an unrelated assessment, and (d) the use of the results to validate predictive instruments from other studies. Factors that contribute to the ability of the study to assess voluntary and external technologies include long-term follow-up, a stable cohort, and storage of such nonnumeric data as cardiograms and blood samples. Framingham results have been used to determine outcome measures in later studies. Although the Framingham Study was not designed to assess a technology, we found that large-scale, observational data bases can and do contribute to technology assessment.