Emergency medicine practitioners (EMPs) often provide ‘medical clearance’ before evaluation by a psychiatry practitioner (PP). We set out to determine the level of agreement between EMP impression and disposition as determined by PPs.
Patients and methods
This was a prospective observational study in an urban tertiary teaching hospital emergency department. We collected data from February to April 2011. We used a convenience sample of patient encounters evaluated by EMPs and subsequent referral for psychiatric evaluation. We asked EMPs whether they thought the patients would be admitted or discharged following psychiatric evaluation, and if discharged, whether to outpatient psychiatric follow-up or to no follow-up. EMPs were asked to base their opinion upon their general impression following their brief medical evaluation. They were not given guidelines on which to base their decision. The EMPs were blind to PP decisions. The κ-statistic was used to calculate agreement between the EMP’s impression and disposition decision by the PP. We excluded patients who were acutely intoxicated, in police custody, or lived in an extended care facility.
We included 156 patient encounters over the study period and had complete data for 152 encounters. Of these, 86 (55%) were admitted, 46 (30%) were discharged with no specific psychiatric follow-up, and 20 (13%) were discharged with a follow-up plan. EMPs predicted the exact disposition in 77/152 (51%) cases (κ=0.264, 95% confidence interval 0.77–0.333). Agreement was higher for admitted patients, with EMPs predicting inpatient admission for 57/86 (66%) of these patients. Other factors associated with higher agreement scores were years in emergency medicine practice by the EMP and suicidal ideation by the patient.
EMPs did not reliably predict psychiatric disposition decisions based on clinical ‘gestalt’. Future research will focus on clinical guidelines to help EMPs better independently assess need for emergency psychiatric services.