Keywords

Youth suicide continues to be of significant concern, with the number of youth presenting for suicidal thoughts and behaviors to pediatric hospitals across the nation doubling between 2008 and 2015 (Plemmons et al., 2018). Many youth treated in emergency settings do not receive follow-up care, and even those who do fail to show improved outcomes (Asarnow et al., 2011). When risk for suicide is detected, the current standard of care is to hospitalize youth deemed to be at imminent risk. This remains the case despite an absence of randomized controlled trials demonstrating this approach alone saves lives (Kennard et al., 2019; Goldman-Mellor et al., 2021). Inpatient hospitalization temporarily protects suicidal patients from engaging in self-harm by restricting access to lethal means. However, most patients receive little or no suicide-specific treatment, and risk for suicidal behavior remains extremely high in the month after discharge (Chung et al., 2019). Furthermore, most hospitalized adolescents who attempt suicide receive limited follow-up care (Doupnik et al., 2020; Spirito et al., 2011). Rates of noncompliance with first appointments post-hospitalization have been reported to be as high as 42%, and in at least one study of adolescent suicide attempters, 25% never attended a single follow-up appointment (Burns et al., 2008; National Action Alliance for Suicide Prevention, 2017). This is concerning as follow-up mental healthcare within 7 days of discharge has been associated with a decreased risk of suicide (Fontanella et al., 2020). Furthermore, outpatient mental healthcare, when provided, is often nonspecific and inadequate to address suicidal thoughts and behaviors. Unfortunately, few professionals are trained in evidence-based suicide care (Schmitz et al., 2012), even though promising treatments for suicidal youth exist (see Zullo et al., Chap. 8, this volume).

Pediatric hospitals and academic medical settings are well-positioned to drive innovation by creating comprehensive evidence-driven approaches to reducing suicide risk. Such approaches should incorporate a continuum of upstream prevention programming as well as family-centered acute clinical services. Nationwide Children’s Hospital (NCH) is an example of a pediatric hospital with a large Behavioral Health (BH) service line that offers multiple levels of care from prevention through crisis stabilization and inpatient psychiatric care. This chapter will review integration of the Zero Suicide framework into NCH’s existing preventable harm reduction quality improvement (QI) initiative to improve clinical care.

Setting the Stage

Zero Suicide is an aspirational goal designed to catalyze transformational change in an organization by providing best practice tools and strategies designed to improve suicide care (www.zerosuicide.edc.org). The Zero Suicide framework is based on three critical factors:

  1. 1.

    Core values that reinforce the belief that suicide can be eliminated by improving service access and quality through continuous quality improvement.

  2. 2.

    Systems management to create a culture that no longer finds suicide to be an acceptable outcome, where aspirational but achievable goals are set to eliminate suicide attempts and deaths and service delivery and supports are organized accordingly.

  3. 3.

    Evidence-based clinical care practices that all staff are trained to provide, delivered consistently across the system of care.

More specifically, Zero Suicide consists of seven elements essential to full implementation:

  • Lead system-wide culture change committed to reducing suicides.

  • Train a competent, confident, and caring workforce.

  • Identify individuals with suicide risk via comprehensive screening and assessment.

  • Engage all individuals with suicide risk using a suicide care management plan.

  • Treat suicidal thoughts and behaviors using evidence-based treatments.

  • Transition individuals through care with warm handoffs and supportive contacts.

  • Improve policies and procedures through continuous quality improvement.

Healthcare systems that bundle core components of Zero Suicide show evidence of reduction in patient suicides and suicidal behaviors (Hampton, 2010; Layman et al., 2021; Stapelberg et al., 2021; Turner et al., 2021). Eliminating gaps in suicide safer care requires a comprehensive, system-wide approach, in our case specifically adapted to a pediatric population. The first step for any organization is the completion of the Zero Suicide Organization Self-Study (Zero Suicide, 2021a). This tool helps to identify organizational strengths and gaps and serves as a baseline as well as a needs assessment.

NCH Zero Suicide Organizational Self-Study summary

Strengths

Gaps

Teams engaged in crisis care had high levels of training

Routine suicide risk assessments administered during diagnostic assessments

Evidence-based suicide risk assessment for programs routinely managing psychiatric crises

Suicide-specific treatment programs (e.g., Dialectical Behavior Therapy)

Inconsistent processes for suicide risk screening, evidence-based risk assessment, and safety planning across services

Limited input from attempt and loss survivors

Inconsistent suicide-specific training across all programs and disciplines

Gaps in communication and documentation of prior treatment and safety plans

Limited follow-up during transitions in care

Inconsistent coordination with families and schools

Lead Element of Zero Suicide

As with most organizations that adopt the Zero Suicide framework, successful implementation and incorporation of core values into the culture of NCH’s large BH service line required the following:

  • Visible support and direction from senior leadership

  • Clear expectations for staff, consistent communication, procedures, and workflows demonstrating organizational responsibility for suicide prevention

  • Ongoing training and education

  • Minimal increase in provider burden including documentation or workload requirements

  • Efforts to reduce provider anxiety related to engaging in suicide care and the potential for adverse events

Accordingly, staff across the BH service line were provided an overview of the framework with an overwhelmingly enthusiastic response. The Zero Suicide Workforce Survey (Zero Suicide, 2021b) was then provided to all staff, clinical and nonclinical, with an 80% response rate. The Survey provided a snapshot of staff’s self-perception of their competence, comfort, training needs, and perceived skill with regard to providing suicide care which was then used to create a training plan. Informed by the Workforce Survey results, our approach to training focused on the following:

  • Engage clinical leaders (e.g., managers and supervisors) across the service line.

  • Review the aims of Zero Suicide and its intent to be both an aspirational goal as well as a specific bundle of interventions.

  • Focus initially on the Lead, Train, Identify, and Engage elements, given that some suicide-specific treatments were already part of the continuum of care.

  • Standardize skills in screening, assessment, and safety planning in an interactive fashion.

  • Develop workflows and process maps flexible enough to tailor for each department.

  • Ensure the electronic medical record (EMR) supports decision-making and reduces clinician burden to enhance communication and eliminate duplicate efforts.

  • Ensure leaders are able to monitor compliance and provide timely feedback to support staff at different developmental levels.

A key innovation in setting the stage for implementation was the development of an NCH Zero Suicide “Toolkit” as a centralized location in the EMR to house all suicide-specific tools and processes across encounters and levels of care (e.g., automated prompts and workflows for screening, assessment, safety plans, and risk categorization). This information had previously been siloed and difficult to access as patients moved among programs and levels of care, disrupting communication of clinically significant information and leading to inconsistent processes. Workflows were informed by our lived experience champion who provided valuable feedback regarding acceptability to patients while pointing out the frustration of answering the same questions about suicide risk and safety to each new provider in the midst of a crisis. Availability of this toolkit contributed to progress in all seven of the Zero Suicide elements, including “Improve” as it facilitated easy access to process measures. Furthermore, it has been well-received by clinicians who report it does indeed improve access to suicide risk-related information and enhances efficiency.

Train and Engage Elements of Zero Suicide

Training of staff across disciplines and programs laid the foundation for all ongoing Zero Suicide efforts in BH which included a focus on clinician attitudes and biases, supervision strategies, and legal considerations. Clinical competencies addressed were evidence-based screening (Ask Suicide-Screening Questions; Horowitz et al., 2012), risk assessment (Columbia-Suicide Severity Risk Scale; Posner et al., 2011), risk and protective factors, safety planning (Stanley & Brown, 2012), and lethal means safety and risk categorization. Since the initial department-wide series of Zero Suicide trainings, training on suicide care competencies has been incorporated into onboarding processes of all BH teams. An aspect of the Engage element is to create a suicide care management plan based on level of risk. Our risk categorization now clearly identifies those higher-risk youth needing additional services and is visible to all departments across the hospital. A next step is to develop a consistent set of practices across departments for those high-risk youth.

Treat and Transition Elements of Zero Suicide

As reflected in the Self-Study and Workforce Survey summaries, BH services had some programs in which clinicians were highly skilled in treating youth presenting with suicidal thoughts and behaviors and others with low comfort and competence. Guided by the extant literature, NCH had developed a continuum of best practice targeted interventions and suicide-specific treatments. These included a Psychiatric Crisis Department with a strong emphasis on safety planning and lethal means safety, a crisis stabilization unit, inpatient psychiatry units, and intensive outpatient and partial hospitalization programs, in addition to an outpatient Dialectical Behavior Therapy (DBT) program. The Zero Suicide framework helped knit these and other elements into a more cohesive whole and served as a focus for training clinical providers on suicide-specific interventions.

Effectively addressing transitions in care has been difficult, and we have not yet fully implemented a transition care pathway for youth stepping down to a lower level of care. We have attempted to implement warm handoffs, but this can be a challenge in a large hospital system when referrals are made outside the hospital’s system of care. On a positive note, NCH has improved care transitions for high acuity youth by adding non-demand caring contacts in the form of validating text messages with images of hope and supportive language sent to adolescents in the year following discharge, regardless of whether they are transitioning within NCH or to an outside agency. We will continue to address the Transitions element by revisiting and formalizing protocols for a transition care pathway supported by review of data to determine linkage, follow-through, and dropout rates.

Improve Element of Zero Suicide

Healthcare organizations implementing Zero Suicide engage in continuous quality improvement (QI) and are able to engage in high-risk work while minimizing serious harm or adverse events. Viewing suicide as a “never event” forces organizations to use best practices, apply continuous QI, and emphasize reducing errors while holding the system accountable, rather than blaming individuals. NCH had an existing commitment to reducing preventable harm via a “Zero Hero” initiative with significant improvements in serious safety events, such as central line catheter infections, using a QI methodology (Miller et al., 2011). Incorporating Zero Suicide was seen as a natural extension of the hospital’s QI efforts.

The development of an internal Zero Suicide Toolkit with discrete data elements made it possible to evaluate compliance and set realistic goals for improvement. Initial goals focused on adherence to suicide risk screening with the Ask Suicide-Screening Questions (ASQ) at first visit and every 30 days thereafter, suicide risk assessment with the Columbia-Suicide Severity Risk Scale (C-SSRS), and completion of a safety plan for youth identified as at elevated risk. Within 3 months of implementation, the initial goal of screening >90% of all new patients for suicide risk was met. Focus then shifted to ensuring that all patients with an acute positive screen received a same-day risk assessment and safety plan. After addressing reporting inconsistencies in the EMR, compliance of >93% was quickly established for same-day risk assessment and safety planning. The robustness of a systemic approach to suicide care was demonstrated when screening compliance at first appointment remained consistently in the target range (>90%) after the shift to telehealth at the onset of the pandemic.

A dynamic Zero Suicide dashboard is available to clinical leaders to monitor progress in real time, evaluate the performance of specific programs, and provide feedback to individual providers. In the future, patient care and outcome data will be prioritized, while process measures will continue to be collected to assess fidelity. QI methodology offers the ability to monitor process and treatment outcomes and, using an iterative approach, modify processes to improve patient outcomes (Valleru et al., 2019).

Gaps and Next Steps

The Zero Suicide framework was adopted at NCH in response to a large increase in high acuity patients and a desire to maximize the effectiveness of growing crisis and inpatient services. Over the past 5 years, NCH BH has successfully implemented many best practice elements. What was previously fragmented suicide care is now more intentional, offering staff greater connection to the work and youth and their families more hope for the best possible outcomes. Consistency of implementation and increased staff competence and confidence across the BH workforce effectively means that no matter where youth enter the BH system, they will receive the evidence-based suicide screening and, as appropriate, assessment, safety planning, and intervention. The model has now been adopted by the hospital broadly even outside of BH, and next steps include expansion to other departments starting with those seeing significant numbers of youth with elevated risk, such as primary care. This expansion requires:

  • Development of referral pathways that enhance each department’s internal capacity to screen and assess for suicide risk, thereby avoiding unnecessary emergent referrals for BH assessments.

  • Development of a suicide care pathway wrapping elements of care around patients at highest risk (e.g., immediate follow-up when patients fail to attend an appointment).

  • Ongoing monitoring to ensure effective transition of patients among services.

  • Ongoing monitoring of communication among providers and departments.

Successful adoption and sustainability of Zero Suicide requires careful planning, supportive infrastructure, and routine review of data aimed at enhancing practice and training. The Zero Suicide model is feasible both in pediatric hospitals and many other healthcare settings, and the release of suicide-specific standards by accrediting bodies has signaled that suicide prevention is a core responsibility of healthcare. However, these standards are still open to interpretation (e.g., frequency of screening) and require further investigation. The use of peer supports or other therapeutic interventions as well as greater specificity regarding risk stratification may help defray the burden on monitoring and environmental controls that currently exists.

Ideally, alternative payment models that support Zero Suicide and suicide-specific care practices would be offered by payers to incentivize this work. It is arguably in the best interests of managed care entities and accountable care organizations to invest in high-quality and effective suicide prevention. Decreased suicide behaviors and improved treatment outcomes should result in patients being supported in less restrictive care settings, decreased healthcare costs, and better experiences of care. Decreased flow to acute care services would, in turn, decrease pressure on the BH system and improve waitlists. This would likely be most effective if providers across communities collaborated to align suicide care practices.

Access to timely care continues to be a challenge, and, as indicated above, even when care is provided, utilization of follow-up referrals tends to be poor. Further investigation into engaging families and high-risk adolescents in treatment is warranted. This includes exploring technology to enhance engagement, increase access, improve lethal means safety and safety planning, and enhance a sense of connectedness. Additionally, using technology helps to increase fidelity to best practices that directly reduce suicide thoughts and behaviors, particularly in the face of a workforce that has variable training and comfort with using suicide-specific interventions.

In summary, Zero Suicide provides a flexible framework for combining best practice suicide care interventions into a comprehensive and systematic approach using real-time data to improve processes and outcomes. There is still a need for more research on the effectiveness of the model and for more systems to share lessons learned as well as how they have managed gaps in care to ease the implementation burden for organizations new to this model. By adopting and striving for the aspirational goal of Zero Suicide and embedding this model’s bundle of specific suicide care practices, healthcare systems expressly support their workforce, alter the quality of care provided, and increase patient safety.