Information Sharing Across Institutions to Enhance Operations of Psychiatric Emergency Rooms in New York City

In the fall of 2013, the New York Presbyterian Comprehensive Psychiatric Emergency Program (Columbia CPEP) was renovated. Creating a larger space doubled the patient capacity from 12 to 24 and physically removed the CPEP from the general medical emergency room. Within a year, patient volume had significantly increased, which led to an evaluation of staffing models and processes to increase efficiency. This project began as a quality improvement project to learn more about processes used by other CPEPs and psychiatric emergency rooms. We surveyed other New York City hospitals with psychiatric emergency rooms to identify promising practices that might be implemented in the Columbia CPEP. Psychiatrists from the following psychiatric emergency rooms were interviewed: Columbia CPEP, Mount Sinai Beth Israel Medical Center CPEP, Mount Sinai St. Luke’s and Roosevelt Hospital CPEP, Kings County Hospital Center CPEP, Bellevue Hospital Center CPEP, and Maimonides Medical Center psychiatric emergency room.

All participants agreed to share information, with the understanding that it would be distributed to all group members. As such, each institution stood to gain from participation by learning about other institutions’ practices. Even though these institutions have similar missions and regulatory guidelines, substantial variability was noted in policies and procedures. Some CPEPs have explicit policies defining which patients can safely be sent directly to the psychiatric emergency service and which patients first require medical clearance in the general medical emergency room. Some institutions require all patients to change into hospital gowns and undergo a physical exam as well as basic labs tests. Other institutions have either a de facto or proceduralized “fast track” for patients who they believe are likely to be treated and released. These patients are examined in their street clothes without physical exams or lab tests. One institution reported hospitalizing about 75% of all patients, while the others hospitalized approximately 25%−45% of patients. Shift length also was found to vary across institutions—some requiring attending physicians to work eight-hour shifts and others requiring 12-hour shifts to minimize handoffs. One institution offered shifts ranging from four to eight hours. Despite the desire to have consistent full-time staff, the institutions were found to vary widely among full-time, part-time, and moonlighter attending physicians. One institution reported having no full-time emergency room attending physicians, and two reported that all emergency room attending physicians are full-time.

These initial interviews led to two immediate results. First, the observation that other PESs identified a primary physician provider for each patient at morning rounds prompted leadership at the Columbia CPEP to adopt this practice to enhance patient care. Second, perhaps the most robust finding of the survey was that interviewees were interested in joining a CPEP/psychiatric emergency room e-list to share ideas, ask questions, and address common challenges.

The e-list includes psychiatrists at each emergency room in the initial survey. In addition to being used for the initial quality improvement project at the Columbia CPEP, the e-list contributes a sense of cohesion and community and has quickly developed into an ongoing resource for obtaining information about specific practices at the participating institutions. One of the first questions posed to the e-list after the initial survey involved the role of social workers. A CPEP director reached out to the other institutions to determine the various clinical responsibilities filled by social workers in their emergency rooms. The responses indicated that social workers do not serve as primary clinicians but commonly perform psychosocial evaluations, plan for disposition, locate hospital beds, obtain insurance authorizations, arrange transportation, and obtain collateral information. One CPEP for children reported utilizing social workers as primary clinicians, who present their cases to attending physicians.

Another e-list topic, which was prompted by the following posting, involved the use of telemetry monitoring of patients in CPEPs or psychiatric emergency rooms. “I have a quick question for all: do any of your CPEPs or psych ERs use cardiac monitors on patients? We never heard of such a thing (for starters, if they need that they aren't medically stable enough for psych) since those machines/cords can be used as weapons or strangulation hazards, but our medical ED is asking us why not. Prompt replies appreciated.”

Three CPEP attending physicians responded on the same day, all stating that their emergency rooms prohibit the use of telemetry for safety reasons. This finding helped the CPEP attending physician enhance the safety discussion with his general medical emergency department colleagues by adding the information that allowing telemetry in a CPEP would make their institution an outlier compared with peers.

An emergency and public psychiatry fellow utilized the e-list to distribute an official survey querying leadership about the training practices for their clinical and nonclinical staff in verbal de-escalation, manual restraint, and trauma-informed care. All hospitals participated in the survey, and the results were shared with the group. The results reflected a wide variability in training in these areas, which served as a learning experience for leadership. Many members commented on how the survey questions and results sparked interest in augmenting their staff training and education in these areas.

One of the liveliest discussions involved a question posed to the group about what specific triage criteria various institutions require patients to meet in order to go directly to the CPEP or psychiatric emergency room without an evaluation in the medical emergency room. The specific criteria, such as vital sign parameters and comorbid general medical conditions, varied widely across institutions. One hospital used very specific criteria, which were embedded within its electronic medical record (EMR) in order to triage patients to either the general medical or psychiatric emergency room. Many group members shared their challenges with managing medically unstable patients, such as learning how to appropriately manage the flow of patients with the general medical emergency room. These discussions prompted one hospital to revise its triage criteria and incorporate a checklist system in its EMR.

In April 2015, the e-list was formalized into an actual Listserv and expanded to include leadership from all CPEPs and psychiatric emergency rooms in New York City. Even across institutions with a similar mandate and regulatory requirements in the same city, there is a wide variation in practices. Instead of risking reinventing the wheel when designing a quality improvement project at a specific site, an institution can gather data across various sites and obtain valuable information to help inform operational changes. People are interested in sharing information about their own sites to learn more about opportunities for improvement and to identify areas where their site may be an outlier. Via information gathering and sharing, a simple e-list has facilitated information integration across peer institutions, expanded the administrative team, and enhanced quality improvement.

The experience with the survey conducted by the emergency and public psychiatry fellow raises the possibility of using such e-lists both for informal data gathering and for research purposes. As health care reform, integrated care, and other environmental changes alter the health care landscape, collaboration is critical to facilitate sharing of promising practices to maximize quality improvement across institutions.

The authors thank Reza Amighi, M.D., Mirabela Bodic, M.D., Diana Dragatsi, M.D., Grant Mitchell, M.D., and Diana Samuel, M.D., for their help and support in the work that led to this publication.