Making Trauma Less Traumatic: Implementing Inpatient Pediatric Screening of Acute Stress Symptoms Free

Our institution is a pediatric burn center and a level 1 pediatric trauma center in an embedded tertiary care children’s hospital with a statewide referral base. All patients who meet the ACS definition of trauma are admitted to our pediatric trauma service, averaging 450 patients a year. In a 2019 program review, the ACS cited the lack of validated screening for acute stress or posttraumatic stress disorder for children as a weakness in our trauma program. Although our hospital has an inpatient pediatric psychology and psychiatry consultation team, that service is not resourced to perform routine screening and consultation for pediatric trauma patients. At our hospital, social workers and case management have focused on identifying basic needs (food insecurity and lack of health insurance), and each individual covers 25 to 50 patients at a time, limiting their ability to focus on mental health or provide additional screening. The main drivers for improving consult frequency included the recent ACS requirement, the worsening pediatric mental health crisis, paucity of available mental health providers, the increased risk of acute stress symptoms in pediatric trauma patients, and a desire to improve our patient experience and satisfaction.

We included patients aged 3 to 15 years admitted to the pediatric trauma service at our institution during our intervention. A minimum of 3 years of age was considered necessary for a child to be developmentally able to engage in therapy. Patients over 15 years old are admitted to the adult trauma service at our institution. Patients with a preferred language other than English were included through use of an in-person or phone interpreter. Patients were not eligible for screening until transfer to the inpatient unit if admitted to intensive care unit. We excluded patients with prior mental illness, those with a prior established relationship with our institution’s psychology team, those with a concern of suicidal or homicidal ideation, end-of-life care discussions, or those needing new psychiatric medication(s) because of being outside of the scope of practice of the LCSW.

Using the model by Waynik et al 2016 to develop and implement a clinical pathway and QI methodology, we assembled a team of stakeholders and experts including representatives from pediatric psychology, psychiatry, trauma surgery, nursing, pediatric and surgery residents, and child life. We reviewed available literature to inform key components of pathway development, building an orderset that identifies measures and a process for data and progress review.²² 

Our main intervention was the bundle of changes associated with the creation of the MTLT team. The team consisted of a newly created position for a dedicated LCSW, a child abuse pediatrician, a pediatric hospitalist, a pediatric nurse, and a trauma team nurse coordinator. The team was strategically staffed with a medical director for outreach and educational efforts, a program coordinator to follow-up with families and assist with providing resources, and an LCSW to screen and provide ongoing therapeutic interventions. To identify a standard process and barriers to MTLT consultation for pediatric trauma patients, we completed a fishbone diagram (Supplemental Fig 1), a process map for our ideal state, and a key driver diagram (Fig 1).

In anticipation of the new ACS accreditation requirement, an LCSW was hired in early 2020 and the MTLT consult became available in July 2020. Consults were initially requested by the trauma service providers messaging the LCSW via the electronic health record (EHR) in-basket. We initially delayed the active phase of improvement interventions to avoid overload of learners and turnover that occurs with the beginning of the academic year in July and to align with timing and bandwidth for other hospital initiatives. Subsequent Plan-Do-Study-Act (PDSA) cycles beginning in November 2020 included interventions to streamline workflow and improve communication, including education of frontline staff and multidisciplinary services, formal electronic consultation order, a contact pager, and landline MTLT phone number (Table 1). EHR interventions occurred at the earliest possible date based on our institution’s Information Technology and registration capabilities. Education included the updated ACS requirement for psychological screening, inclusion and exclusion criteria, the purpose of the MTLT program, and how to place an MTLT consult. Quarterly educational presentations were made to pediatric and surgical residents, along with reminders in e-mail form in the weekly resident announcements. The MTLT team attended pediatric nursing meetings on all 4 pediatric floors as well as the pediatric intensive care unit to provide education. Educational flyers were posted on nursing units, in staff bathrooms, and break rooms highlighting reasons to consult the MTLT team, how to place the consult order, and how to contact the LCSW.

Additional interventions included improvements to the EHR. The MTLT consult order was created in lieu of in-basket messaging and added as a default order to the pediatric trauma admission order set, and discharge smart phrases were created with anticipatory guidance regarding the MTLT team, the importance of screening for acute stress symptoms after a trauma, and contact information for the MTLT team (Table 1). Contact information was provided to all patients discharged from the pediatric trauma service so that families were able to contact the LCSW after discharge if the patient developed new or worsening symptoms days to weeks after the trauma. The LCSW used validated screening tools including the Child Stress Disorder Checklist – Short Form, Mood and Feelings Questionnaire, and the Child PTSD Symptom Scale.^23–25  Screening for acute stress symptoms did not need to occur before discharge, given that symptoms may develop well after discharge, and the LCSW could provide outpatient education, resources, and therapy. Any trauma patient not screened in the hospital (because of rapid turnover or weekend admission) was followed up by the LCSW via telephone call with the primary guardian. If concerns were identified (either current symptoms, elevated screening scores, or other areas of concern) requiring additional follow-up at any point, the LCSW introduced psychoeducation and coping skills and continued providing therapy virtually or in person in the outpatient setting until the patient could be connected to local resources.

Data for eligible admitted trauma patients, presence of consultation order, successful consult completion, and findings of screening examinations were analyzed from a deidentified weekly report generated in the EHR. We monitored data during a baseline period after MTLT team creation without active improvement interventions (September 2020–November 2020), our implementation phase with active PDSA cycles (November 2020–April 2021), and a sustainment period after meeting our goals for the primary measure (April 2021–November 2021).

The primary outcome measure was the proportion of eligible pediatric trauma admissions with an MTLT consult. Secondary measures included percent of patients with positive screens (symptomatic patients), the frequency of endorsed symptoms, and the proportion of patients lost to follow-up. Positive screens were defined as patients or families identifying new symptoms after a trauma, including but not limited to avoidance, hypervigilance, mood changes, increased anxiety, sleep disturbance, nightmares or intrusive thoughts, irritability, anger, regression, and appetite changes. Positive screens were classified by symptom type and trauma type. We defined lost to follow-up as having a consult order placed, but not being seen in person or reached by phone despite 2 attempts with voicemails left at every number listed in our electronic medical record within the subsequent 45 days after admission. As a balancing measure, the number of families declining intervention, including screening or therapy and resources, was also tracked.

Weekly consultation frequencies for our primary outcome were analyzed as a time series outcome variable in a statistical process control P chart (QI Macros 2022, Scoville Associates, Raleigh-Durham, NC). Standard Shewhart rules were used for chart interpretation and special cause variation identification.

This study was reviewed by the Institutional Review Board and was determined to be exempt from Institutional Review Board approval (#920-3453).

From September 2020 through November 2021, a total of 304 patients were admitted to the pediatric trauma service between the ages of 3 and 15 years. The average age was 8.3 years old, and the majority were white, with Medicaid insurance, and male. The top 2 mechanisms of trauma were fall (40.8%, n = 124) and motor vehicle accident (16.1%, n = 49) (Table 2). Before any intervention, 60 eligible pediatric trauma patients were admitted, and the mean baseline MTLT consult weekly frequency was 4.1%. Special cause variation resulting in a shift of the mean occurred 5 weeks into the intervention after provider education and establishment of a MTLT pager with an improvement in the mean weekly frequency to 47.1%. Subsequent ongoing provider education, reminders, and creation of a formal EHR consult order resulted in reaching our goal consultation rate with a subsequent increase of the mean weekly frequency to 83.3% of eligible patients receiving consultation orders. We continued scheduled provider education and ultimately demonstrated 100% of eligible patients having order for MTLT consultation. We sustained that value over the next 25 weeks despite the start of a new academic year with new frontline providers (Fig 2).

In total, there were 213 MTLT consults placed, 3 of which were excluded because of prior psychiatry involvement. Of the 210 remaining consults, 147 (70.0%) patients were screened by the MTLT team. Because of relatively short length of stay and lack of weekend staffing, 63 of the 147 (42.9%) were screened in person and 84 of the 147 (57.1%) were screened via phone, with 62 (29.5%) lost to follow-up. One patient with a consult died before screening could occur.

Of those screened, 48 of 147 (32.7%) were symptomatic. The most common symptom reported in symptomatic patients was avoidance (n = 16), followed by hypervigilance or increased startle (n = 15), mood changes (n = 14), and nightmares or intrusive thoughts (n = 12). Appetite changes (n = 2), physical complaints such as nausea or headache (n = 2), and developmental regression (n = 2) were the least common observed symptoms (Table 3). All symptomatic patients received at least 1 therapy session and 14 of 48 (29.2%) were provided with additional mental health resources in their community. No families refused screening, resources, or therapy. The families of 3 asymptomatic patients asked for additional resources: 2 had parents who were symptomatic after their child’s trauma and 1 had a symptomatic sibling.

This study demonstrates that a successful QI intervention via utilization of LCSWs to provide therapy and resources can improve screening for psychiatric symptoms after physical trauma and allow pediatric trauma programs to meet the new ACS regulations. Given the ACS mandate and the national paucity of psychiatric providers, we feel that this pathway is of significant value to other pediatric trauma programs around the country.

Utilization of a LCSW for screening and therapy rather than a psychologist or psychiatrist was intentional in our model. This approach is both more cost effective for hospitals and more feasible given the greater availability of social workers compared with other mental health providers. Unlike psychiatrists and psychologists, LCSW training focuses on family empowerment and knowledge of local resources and systems. Not only can they initiate evidence-based psychoeducation and coping skills, but they also can connect patients to continued care in the outpatient setting. Additionally, our program recognizes that many patients may not develop symptoms for days to weeks after a trauma. Families can contact the LCSW to seek counseling after discharge should symptoms develop, but our LCSW also attempted to contact families after discharge to ensure no new symptom development in the days to weeks after discharge.

Successful interventions throughout implementation were associated with special cause variation in our primary outcome measure. These key components included the development of the pathway utilizing an LCSW, and implementation of validated approaches to successful pathway implementation, including order set utilization, development of standard work for consultation, broad education, and implementation of multiple modalities to increase ease of communication with the MTLT. Our program employs multiple novel features that may make it more generalizable to other hospitals. First, our findings regarding the overall rate of symptomatic trauma patients (33.5%) are consistent with prior literature. In Mazo et al,²⁶  pediatric trauma patients were assessed after a traumatic event: 41% of patients met criteria for ASD and 31% met criteria for PTSD. A similar study by Ridings and colleagues²⁷  found that 35.5% of pediatric trauma patients experienced clinically significant symptoms of PTSD 30 days after their trauma. The fact that no families refused screening, resources, or therapy is likely indicative of the changing stigma of mental health in the United States.

As a single center QI intervention, there are inherit limitations to generalizability to other settings. Our MTLT team creation was based on creation and funding of a distinct LCSW position, and thus we recognize that the implementation of this clinical pathway is subject to local resource availability, including funding and an available, experienced LCSW with posttrauma therapy training. Our data are limited by nearly one-third of patients being unable to be reached for screening, which could introduce bias in the proportion of positive screens or the constellation of symptoms presented. Having a single therapist responsible for contact resulted in limitations over weekends and holidays for screening in the hospital for patients with shorter lengths of stay, placing an increased reliance on phone contact for those patients, which is challenging postdischarge. Additionally, many patients and families seen by the psychiatry or psychology services may have benefitted from the ongoing resources and outpatient follow-up provided by our LCSW but were ineligible for the program.

Hospitalization after trauma is an often-missed opportunity for screening pediatric patients for ASD or PTSD and providing interventions. Clinical pathway implementation, provider education, EHR improvements, and utilization of LCSWs to provide therapy and resources can improve screening for psychiatric symptoms after a physical trauma and allow pediatric trauma programs to meet the new ACS regulations by adapting the pathway and resources provided here. Future quality improvement measures will focus on continued education efforts to promote sustainability and address limitations caused by single LCSW availability. We plan to present this information to hospital leadership to request additional LCSWs to ameliorate pediatric mental health workforce limitations and address the surging pediatric mental health crisis. Additionally, as the literature suggests that symptoms may not begin for weeks to months after a trauma, future iterations of this project should focus on improving our follow-up process, potentially using additional LCSWs in clinic to screen our patient population further out from their initial trauma.

The authors thank Dr Ashley Sutton for her review of the manuscript, and the Children’s Hospital at the University of North Carolina School of Medicine for their ongoing support of this project.