Admission to a close ward was analyzed at the Department of Psychiatry, University of Oulu using 888 patients and their 1861 assessment and treatment episodes. Of all referrals for involuntary assessment (n = 237, 12.7% of all episodes) a total of 44 (2.4%) used "questionable" juridical criteria: the final diagnosis was not psychosis. In the follow-up, the admission of the questionable patients was mainly considered a clinical necessity, and at least one third of them were diagnosed as being psychotic and 2 committed suicide. An elevated probability of belonging to the questionable group was seen among patients in their first treatment episode, with minimal professional education, female sex, short treatment time, or residence in a rural area. The result suggests that some inequality existed between women and men, less and more educated and residents of urban and rural areas. The results also reflect conflict between the ethics and clinical practice of involuntary commitment, and the phrasing of the law, especially its diagnostic limitation to psychotic states only.