The degree of clinical agreement amongst different physicians on the presence or absence of generalized lymphadenopathy was assessed in 32 randomly selected participants from a prospective study of male sexual contacts of men with AIDS or an AIDS-related condition (ARC). Three physicians completed the same standard examination that was developed for the main project and conducted the examination of the anatomic regions in the same order on each person, at approximately the same time, and in random order. One physician (Doctor A) was the physician responsible for conducting examinations on the main cohort from which these participants were selected. Intra-observer agreement was assessed by comparing Doctor A's examinations on these participants with those he had recently conducted within a one and a half month period in the main study. Acceptable levels of intra-observer agreement (kappa = 0.72) and interobserver agreement (kappa = 0.66) were demonstrated for the presence or absence of generalized lymphadenopathy for Doctor A and Doctor B, a physician who periodically replaced Doctor A in the main project. Agreement between Doctors A, B, and C, was less satisfactory (kappa of 0.45 and 0.39, respectively). Doctor C was the least experienced with the standardized examination. However, during the progress of this study, agreement between the three doctors improved (kappa values for the latter 16 participants ranged from 0.60 to 0.86) suggesting that experience with the criteria and the standardization of the examination may enhance agreement.(ABSTRACT TRUNCATED AT 250 WORDS)