A number of studies have suggested that there is considerable variation in the frequency with which physicians order particular tests. This paper reviews the relationship of variations in ambulatory testing to characteristics of physicians' training and practice; and organizational factors such as physician group size, method of reimbursement, and ease of access to technical services. Results from previous investigations suggest that specialty training, more recent physician graduation, and large group practice settings are associated with significantly higher test use. Information available about other physician characteristics and organizational factors is equivocal. The extent of variations in practice has important implications for the cost and quality of medical care. Greater effort should be made to document the effects of nonmedical factors and to clarify how those effects are mediated and how they might be modified to improve the use of medical resources.