This study considers the denial of access to inpatient care to those seeking hospitalization following psychiatric emergency service (PES) evaluation. It evaluates how civil commitment criteria, functional status, institutional constraints, social bias, and procedural justice indicators are likely to impact denial of care decisions, and considers 12 month outcomes. PES evaluations of 583 patients in 9 California county general hospitals were examined via logit modeling to determine those factors contributing to the decision to deny access to inpatient care. Differences in the importance of influences on the decision making process and outcomes at 12 months are examined in two contrasts: first, admitted and released patients seeking care, then, the latter group versus all other patients. Outcome measures include numbers of deaths, violent crimes, and involuntary readmissions to the PES. Of the patients evaluated, 8.4 % were denied access to inpatient care despite their avowed wish to be hospitalized. When compared to admitted patients seeking hospitalization or to all other patients, analyses show that clinicians relied on civil commitment admission criteria and the availability of a less restrictive alternative to the hospital in making decisions on patient retention. When compared with all other patients, the probability of unwanted release was greater for individuals evaluated in difficult circumstances, for those without insurance, and for those with higher functional status. Fewer deaths were observed in the group denied admission, though no other significant outcome differences were observed. Dangerousness and mental disorder in the absence of a less restrictive alternative to hospitalization, along with an overall assessment of the patient's functional status, are effectively employed as triage criteria in determining who is denied access to inpatient care following PES evaluation. While some higher functioning individuals are subjected to a variant standard of access to inpatient care because of a lack of insurance, and endure the misfortune of being evaluated under difficult clinical circumstances, outcomes seem contingent on clinicians' ability to distinguish between groups on the aforementioned triage criteria.