OBJECTIVE: The purpose of this article is to characterize the use of psychotherapy based on episode duration. METHOD: Data were analyzed from the household section of the 1987 National Medical Expenditure Survey. The authors determined the demographic characteristics, provider and reason for visit distribution, mental and physical health status, and expenditures associated with very short-term (one to two sessions), short-term (three to 10 sessions), intermediate-term (11 to 20 sessions), and long-term (> 20 sessions) psychotherapy. RESULTS: Long-term psychotherapy accounted for 15.7% of psychotherapy users and 62.9% of total psychotherapy expenditures. Age above 65 years, black race, and less than 12 years of education decreased the likelihood of receiving long-term psychotherapy. Whereas long-term psychotherapy episodes tended to be provided by the specialty sector (65.7%) for specific mental conditions (53.8%), very short-term episodes were predominantly provided by the general medical sector (72.2%) for general medical or unspecified conditions (68.3%). Psychotropic medication use and, to less extent, psychiatric hospitalization tended to be more common among longer- as opposed to shorter-term users. CONCLUSIONS: Long-term and short-term psychotherapy tend to be provided by different health care professionals for the treatment of different types of health conditions. To help ensure the future of third-party payment for long-term psychotherapy, research is needed to better define the conditions under which long-term psychotherapy achieves benefits that equal or surpass those of other medical services or procedures of similar cost.