Although OA is recognized as the most common joint disorder, its pathogenesis is unclear and little is known about its natural history and progression. Epidemiologic studies have identified age as the strongest and most consistent correlate of OA, with distinct patterns of distribution by sex, ethnicity, and geographic location. There is growing consensus that OA represents a heterogeneous set of diseases that are affected by the interaction of multiple risk factors, producing a common pathway of disease; and that it is important to study multiple risk factors in association with generalized OA and site-specific subsets of OA. Two basic types of generalized OA are recognized, although it is quite possible that there are additional types of generalized OA. Genetic susceptibility, which implies some systemic factor of cartilage vulnerability, is thought to be the major risk factor for generalized OA. Mechanical factors associated with wear and tear on the joint have been the most important factors found in association with site-specific OA, although it is possible that systemic factors are also involved. Most likely there are multiple factors, such as genetic susceptibility; anatomic, biomechanical, and biochemical changes associated with aging; and exposure to various types of joint trauma and injury that are implicated in the etiology of OA. There is little known about progression of OA and the extent of the outcomes involved with progression of OA. Crucial to future epidemiologic research is the identification of homogeneous subsets of OA that facilitate standardized case definition, and research designs that incorporate multiple joints in one study and investigate the independent and interactive effects of multiple hypothesized risk factors. Cohort studies and longitudinal follow-up of persons with OA are needed in order to have a better understanding of the natural history of OA and its related outcomes.