This study describes the epidemiology and risk/prognostic factors for back pain episodes of care in a population representing the nonelderly in the United States. Claims forms from the RAND Health Insurance Experiment, a randomized controlled trial of the use of health services, were analyzed. Claims forms were selected if one of the patient-designated reasons for the visit was back pain. Visits were grouped into episodes of care. Descriptive statistics were calculated for episodes. Multivariate logistic regression was used to calculate adjusted odds ratios for independent explanatory sociodemographic and health status variables associated with back pain episodes of care. The study concludes that back pain episodes of care occur commonly in the adult U.S. population, but usually are brief and recur infrequently.