Relationships between patient, family, and significant other disposition preferences in psychiatric emergency services and the clinical symptom ratings and disposition decisions of psychiatrists.

A previous paper examined the relationship between 10 clinical ratings made by psychiatrists and 465 psychiatric emergency service (PES) dispositions. A logistic regression model explained 52% of the variance in the psychiatrist's decisions to admit to inpatient services or release. The 5 clinical ratings in the model, rated on 8-point scales, were severity of danger to self, psychosis, inability to care for self, impulse control, and depression. The current paper examines the relationships between patient and family/significant other dispositional preferences expressed in the PES, and the psychiatrist's disposition decision and the 10 clinical ratings. It found that the dispositional preferences of the patient, family, and other individuals (community clinician, police, or ambulance driver) were very consistent with the actual decision made by the PES psychiatrist. Further, it found that these preferences explained an additional 35% of the variability in the psychiatrist's decision to admit or release. Finally, the ten clinical ratings made by the PES psychiatrist explained only a small amount of the variance in the dispositional preferences, and some of the significant coefficients were in the opposite direction expected. Patient, family and other significant individuals' expressed preferences and expectations in the PES are important and should be routinely recorded in the charts. The small relationships between clinical ratings and preferences suggests that the reasons the family/significant other and the patient seeks hospital admission or release may differ with psychiatrists' reasons for disposition.