OBJECTIVE: The study examined factors associated with admission to public and private hospitals from a mental health care emergency screening system operating under a longstanding mandate to maximize use of private inpatient treatment. METHODS: For 206 patients evaluated at the mental health emergency screening site over a two-and-a-half-month period, data were collected on demographic and clinical characteristics, admission history, services received during the emergency encounter, system variables such as time the patient spent at the emergency screening site, number of admission sites asked to accept the patient, and all reasons cited by providers for refusing to admit the patient. Logistic regression was used to develop a model of factors most likely and least likely to be associated with private hospitalization. RESULTS: Overall, 60 percent of the sample was refused admission by one or more providers, and 55 percent, who were not accepted by and private hospital, became public patients. Private hospital admission was associated with patient or family involvement in referral and disposition, private or Medicaid insurance, a presenting problem of depression or suicidality, and longer time spent at the emergency screening site. Public admission was associated with no insurance, a past history of major public hospitalization, current or past history of assaultiveness, a presenting problem of aggression, and lack of any discharge site. CONCLUSIONS: Private providers are reluctant to admit patients who have characteristics associated with public hospitalization. In the restructuring of health care, a more fully privatized system will likely be called on to absorb such patients. Their care, treatment, and impact on the system should be carefully monitored and evaluated.