347 - Improving Documentation of Behavioral Crisis Events: A Quality Improvement Initiative

Publication Number: 4340.347

Ankita T. Gupta, Vanderbilt University Medical Center, Nashville, TN, United States; Katherine S. Spencer, Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN, United States; Delana Vallery, Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN, United States; Heather Kreth, Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN, United States; Laura Parks, Monroe Carell Jr. Children's Hospital at Vanderbilt, Hendersonville, TN, United States; Lisa Rhinehart, Vanderbilt University Medical Center, Cross Plains, TN, United States; Alison Herndon, Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN, United States; David P. Johnson, Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN, United States

Background: Pediatric acute care hospitalizations for mental health crises have risen over the last decade. In 2023, our hospital established a behavioral response team, called the Behavioral Event Support Team (BEST) to implement a standardized approach to identifying and managing behavioral crises in emergency and inpatient settings. No structured method for documenting these events in our electronic health record (EHR) exists.

Objective: To increase the rate of structured documentation of BEST codes in the EHR from 0% to 75% by January 2026.

Design/Methods: We formed a multidisciplinary team in January 2024 intending to improve interdisciplinary communication of behavioral events for children with mental health conditions. We developed a key driver diagram to guide improvement efforts (Figure 1). We created reports to capture all patients with structured documentation of a BEST code in the EHR, and those for whom a BEST code was initiated. We obtained retrospective baseline data from November 2024 to March 2025 prior to the first intervention. The primary measure was the weekly proportion of BEST codes with corresponding structured EHR documentation, calculated as the number of documented BEST events in the EHR divided by the total number of BEST activations. The balancing measure was the number of BEST activations per week to ensure our efforts did not deter BEST code activations. We used annotated statistical process control charts to monitor and analyze performance over time.

Results: Baseline data included 145 BEST code activations. None had structured documentation in the EHR. Following the development of a structured documentation method, targeted education for the patient safety team, and the initiation of documentation efforts, we observed special cause variation, prompting a centerline shift to 49%. After six months of continued education and initiation of patient safety team leadership check-ins, we introduced a dedicated nursing documentation flowsheet in September 2025 after which we have observed six consecutive points above the mean and special cause variation with a point outside the control limits (Figure 2). BEST code broadcast volume remained stable after the first intervention (Figure 3).

Conclusion(s): QI methods and targeted interventions were associated with sustained increases in structured EHR documentation of behavioral events without reducing overall event reporting due to staff burden. This work enables interdisciplinary communication of behavioral events in the hospital setting in order to provide better care for this vulnerable patient population.

Figure 1: Key Driver Diagram


Figure 2: BEST Code EHR Documentation
P-chart representing percent of BEST Codes with EHR documentation on the y-axis, and weeks on the x-axis. The three labeled interventions were implementation of EHR structured documentation, initiation of patient safety leadership check ins, and initiation of nursing documentation.

Figure 3: Total Number of BEST Broadcasts By Week
I-Chart representing the total number of BEST Code Broadcasts over time with baseline period representing the period prior to first intervention.