Introduction: In the past 20 years there has been a significant decline in the number of inpatient psychiatric beds in the United States, while the number of patients seeking psychiatric treatment in the emergency department (ED) has increased over the same time period. Given the increase in demand for psychiatric services and decrease in availability of inpatient treatment the ED is becoming the de facto place of treatment for the majority of psychiatric crises. Psychiatric patients experience longer lengths of stay (LOS) when compared to non-psychiatric patients, especially when transfer to another facility is required. Therefore, improvements in the efficiency of evaluation, treatment and disposition of psychiatric patients benefit both patients and the EDs that care for them.
Methods: To improve throughput and reduce wait times in the ED at our Level I trauma center located in the Upstate region of South Carolina, we implemented several improvements. We then tracked pre- and post-intervention metrics, including LOS and the time from ED consultation order to the completion of psychiatric consultant documentation. The intervention consisted of several protocols with various checkpoints for required documentation necessary for progression through overall mental health evaluation and treatment. In addition, structured psychiatrist and social worker evaluation-note templates were standardized to improve documentation accuracy, consistency, efficiency and overall patient safety. A separate tracking system is monitored by a dedicated psychiatric advanced practice provider to ensure compliance on note completion and order set utilization. The time from ED consult order to completion of psychiatric consultant documentation and mean LOS (in hours +/-standard deviation [SD]) were measured for six months before (10/2016 to 03/2017) and eight months after (4/2017 to 11/2017) institution of these protocols. We then compared pre- and post-intervention measures using Student’s t-test (p<0.05).
Results: The number of ED patients seen by a psychiatrist were 3,331 and 4,482 in the pre- and post-intervention time frames, respectively. Overall mean LOS significantly decreased from 38.2 (SD+57.5) to 24.9 (SD+37.6) hours after institution of these new protocols. In addition, mean LOS for patients discharged to home or to a psychiatric facility also significantly decreased from 36.9 (SD+53.7) to 21.8 (SD+30.7) and 42.8 (SD+66.5) to 31.8 (SD+49.1) hours, respectively. Time from consult order to completion of ED psychiatrist documentation significantly decreased from 11.3 (SD+9.8) to 6.2 (SD+6.9) hours. All four comparisons were significantly different with p-values ≤ 0.01.
Conclusion: The implementation of these protocols showed a rapid, sustained improvement in overall efficiency of evaluation and disposition of psychiatric patients in our ED. The decrease in time to evaluation for patients discharged home, as well as a decreased time to transfer to inpatient level of care for those requiring hospitalization made for greater throughput and decreased demand on ED resources. Of note, this improvement in efficiency was observed despite an increase in the volume of psychiatric patients seen by the ED over the course of the study. Our institution continues to track outcomes and has implemented further changes including hiring several dedicated ED psychiatrists, with a goal of providing 24/7 availability of in-house psychiatrists embedded in the ED in an effort to further decrease LOS and improve patient care.
Given the shortage of psychiatrists and declining numbers of psychiatric hospital beds, until an alternative solution for this difficulty of access to psychiatric services is implemented the demand for psychiatric services in the ED will remain high. While more study is needed to determine the generalizability of our findings, we believe that implementation of similar interventions would likely benefit other EDs struggling with delays in psychiatric evaluation and disposition.
