The purpose of this study was to estimate the incidence of clinically identified (i.e., ascertained either because of symptoms or an incidental chest radiograph) sarcoidosis in a geographically and ethnically defined United States population. We employed a research-quality database of a large health maintenance organization (HMO), Kaiser Permanente, Northwest Region (KPNW), which collects morbidity and utilization data for a random sample of members sociodemographically representative of the service area of the HMO, to identify persons with suspected or confirmed sarcoidosis. Nine clinically identified incident cases of sarcoidosis were verified within the 1.87*10(5) person-year sample over a 21-year span, 1967-1987, an estimated annual all-ethnicity incidence rate of 4.8/10(5) (95 percent confidence interval ((CI)): 1.7, 7.9). Among persons of Caucasian ancestry, the estimated annual incidence was 2.8/10(5) (95 percent CI: 0.4, 5.2). If the incidence among Caucasians and the relative risk of developing sarcoidosis among African-Americans in the KPNW population is representative of the U.S. population, we estimate the incidence of clinically identified sarcoidosis in the latter to be 7.3. If, in addition, clinically identified cases constitute 42 percent of those that are clinically ascertainable (i.e., identified by mass population screening in addition to cases detected because of symptoms or incidental chest radiographs), as observed in the Swedish population, our estimate of the U.S. incidence of clinically ascertainable sarcoidosis would be 17.4.