Issues related to notifying subjects in epidemiological studies that they may be at risk of adverse effects from the agents being investigated were discussed. The discussion was based on presentations made at the Methodology of Notification workshop sponsored by NIOSH in August 1990 and focused on the NIOSH notification policy. It was emphasized that the use of individual notification for NIOSH epidemiological studies has a relatively long history. In 1981, NIOSH conducted a pilot study to notify workers exposed to 2-naphthylamine (91598) of their risk of bladder cancer. This study which was based on the philosophy that the workers had a right to know risk information pertaining to them that is held by the group performing the research has become the model upon which criteria and guidelines for notification were eventually developed. The approach can also be used as a primary and secondary prevention technique by allowing those at risk to take appropriate steps to reduce their risk. NIOSH developed a technique to address both older record linkage studies and current or future studies. Once a study was judged to warrant notification, a procedure for selecting the method of notification based on an algorithm for relative and attributable risks was developed. Workers in studies which showed more serious risks such as a standardized mortality ratio greater than 200 or a lifetime absolute attributable risk of 0.001 or greater were to receive individual letters of notification. Participants in studies showing lesser degrees of risk were to receive notifications directed at groups rather than at individuals. For each study, NIOSH developed a notification implementation plan that involved collaboration between NIOSH and companies and unions to draft the notification materials and to address related issues such as medical screening and support services. For studies completed after 1988, the Industrywide Studies Branch of NIOSH provides worker notification as the final component of each study. To date, NIOSH has performed 14 notifications to workers exposed to a variety of agents. NIOSH has contracted with academic research institutions to evaluate long term medical, socioeconomic, and behavioral impacts of the pilot notification performed in 1981 and to develop a technique to evaluate short term impacts of notification projects. To date, no evidence has come to the attention of NIOSH that the notifications have resulted in increased levels of anxiety or pyschological sequelae or an increase in the number of frivolous lawsuits.