Suicide is the second leading cause of death for youth aged 10 to 24 years in the United States, with rates increasing significantly since 2007 due to factors including increased firearm access and worsening youth mental health. Franklin County, Ohio, reflects national trends, and a regional collaborative aims to reduce pediatric suicide rates by 25% by 2030.
Nationwide Children’s Hospital led a collaborative quality improvement initiative to address youth suicide by focusing on protective factors and mitigating suicide vulnerability through health care and community-based interventions, including the Zero Suicide framework and school-based prevention programs.
The Signs of Suicide prevention program, implemented in more than 275 Ohio schools and reaching more than 85 000 students, combines trusted adult and student training with universal depression and suicide screening to increase early identification, provide steps to access care, and reduce suicidal behaviors. The suicide safer care bundle for health care centered around screening, assessment, safety planning, and lethal means counseling. Efficient processes for supporting care transitions, like the Caring Contacts text messaging program, were implemented across multiple sites to bridge gaps in follow-up care. Educational platforms promote children’s mental health, break stigmas, and provide resources for suicide prevention to the community and the health care workforce.
Our collaborative extends the Zero Suicide initiative by integrating community prevention strategies across diverse health care settings to reduce youth suicide. Although early results are promising, challenges in data collection and the need for community engagement highlight the importance of continued adaptation and widespread commitment to addressing youth suicide.
Introduction
Suicide is the second leading cause of death in the United States for young people aged 10 to 14 and 15 to 24 years—only unintentional injury ranks higher in those age groups.1 An upsurge in youth suicide started in 2007, with a 69.5% increase in the mortality rate for youth suicide since. Increased firearm access and overall worsening of youth mental health are hypothesized to underlie this trend.2 Franklin County, Ohio, is representative of national trends. The youth suicide rate in Franklin County remained lower than the national rate during the 2011–2020 period, but it increased at nearly the same pace as the national youth rate.
Youth suicide is tragic, and the sequelae are far-reaching, impacting family and friends, community, schools, first responders, and health care providers. In addition, for every death by suicide in youth, there are approximately 200 to 300 attempts.3 Because of these stark numbers and growing importance as an indicator of youth mental health, pediatric suicide was chosen as one of 8 Pediatric Vital Signs for targeted attention by a regional collaborative in Central Ohio. With this knowledge we aimed to reverse the trend of increasing pediatric suicide rates in our community (Franklin County, Ohio) with the goal of a 25% reduction by the year 2030.
Methods
Partners
This quality improvement initiative was coordinated by Nationwide Children’s Hospital (NCH) with external partners including suicide-focused research scientists, school-based prevention teams, psychiatry and behavioral health and primary care providers, and community youth-serving agencies. The Model for Improvement framework used by the Institute for Healthcare Improvement (IHI) was used and supported by project management and quality improvement consultants.
Approach
The project was approached in multiple phases starting with the identification of key drivers believed to either enhance known protective factors or mitigate or decrease known suicide vulnerability factors, with consideration of social and environmental drivers of care in the local environment. Suicide prevention efforts in pediatric populations must consider developmental differences, family systems, school needs, and the unique risk and protective factors in this age group. In line with previous findings of health care contact in the year prior to suicide death,4 a review of Franklin County youth suicides from 2015 to 2017 found that NCH had provided lifetime care, defined as a visit to a health or mental health care provider, to 68% of the young people aged 5 to 19 who died by suicide. In the 12 months before death, 14 of 17 (82%) of these youth had received care in the outpatient (n = 12) or emergency (n = 2) setting, with most of those contacts occurring in primary care (n = 6) and behavioral health (n = 5). These results encouraged an approach to key drivers focused on care at the hospital and related health care institutions initially.
The second step included a review of existing interventions. Lowering the rate of suicide among children and adolescents seen in health care settings requires a combined approach integrating detection of risk, crisis management, and suicide-specific treatment.5 These components of intervention may be best implemented in health care systems that use the Zero Suicide (ZS) framework. Comprising 7 core components, ZS emphasizes a system-wide approach to improve care and close gaps for individuals with suicidal ideation. Reductions in suicide attempts and deaths have been documented in outpatient mental health settings implementing a bundle of these suicide safer care best practices, particularly lethal means safety and quality improvement initiatives related to suicide prevention.6
In addition, community-based prevention approaches show promise. Specifically, school-based prevention and awareness programs decrease suicidal thoughts and behaviors in youth.7,8 Counseling to reduce or restrict access to lethal methods of suicide such as firearms and medications, both over the counter and prescription, and treatment (pharmacological or psychological) for mental health conditions like depression demonstrate efficacy in preventing suicidal behavior.9
Consistent with methods used by IHI, our team used this summary of prior evidence and an assessment of local practices to develop a key driver diagram. Specific key drivers were identified (Figure 1). Addressing youth resiliency and social emotional development are long-lasting protective interventions necessary to achieve the desired reduction in pediatric suicide.10 Suicide safer care is not only a philosophy but associated behaviors and strategies that promote informed, engaged providers, use best practices, and prioritize access to effective treatment. Lastly, suicide-related data that are timely and accessible are critical for understanding variation, relationships, and the impact of interventions and outcomes.
Key driver diagram for reducing pediatric suicide rates linking interventions to drivers.
Key driver diagram for reducing pediatric suicide rates linking interventions to drivers.
With these key drivers in mind, specific interventions were developed: (1) growth of school-based suicide prevention initiatives; (2) development and expansion of a standardized health care approach to safer suicide care; (3) design and dissemination of efficient and effective processes for transitions of care following behavioral health crisis; (4) ZS adoption in the community, including enhanced competence in safety planning and lethal means safety counseling; and (5) workforce development using technology (Figure 1).
Results
Growth of School-Based Suicide Prevention Initiatives
Schools are a critical setting for effective suicide prevention. The Signs of Suicide prevention program (SOS) is the only school-based suicide prevention program to show significant reductions in self-reported student suicide attempts over time through randomized controlled trials in the United States.11–13 SOS combines 3 best-practice features: (1) a trusted adult training for school staff and parents; (2) training for students on how to support peers displaying warning signs and how they themselves can access support; and (3) universal screening for depression and suicide. These elements increase awareness of depression and suicide, frame suicide as preventable, inform students of crisis resources, and teach action steps to respond to warning signs of suicide. Most importantly, SOS helps to identify a school’s most vulnerable youth through multiple pathways, including self-referral, standardized screening, and/or referral by a concerned peer or adult.
With NCH as a partner, more than 275 Ohio schools have implemented SOS to date. Schools are supported in ensuring that local implementation meets the needs of their communities. Readiness assessment, stakeholder identification, and training for school staff, classroom presenters, and parents/caregivers occur prior to classroom-based instruction. Students participate in 2 SOS sessions, each the length of a traditional class period, learning about depression, suicide risk factors, and how to seek help for oneself or a friend by accessing trusted adults. They complete a brief standardized depression/suicide screening (Brief Screening for Adolescent Depression) and can request counselor support via response cards for themselves or a peer. School staff review screenings, and when indicated, same-day risk assessments occur with parental consent and communication. Recommendations for follow-up and ongoing safety planning are discussed with parents.
Since the 2015–2016 academic year, more than 85 000 Ohio students have participated in the SOS program. Notably, even though 19% of students screen positive when asked about depression and suicidal thoughts, less than 3% required urgent mental health care or hospitalization, suggesting that with this type of training, assessment, and safety planning, students can typically get their needs met in the community.
Suicide Safer Care Bundle
Several interventions were packaged as a suicide safer care bundle for implementation in the health care setting after an assessment of existing care in the specialty behavioral health setting. The bundle includes (1) screening for suicide using a suicide-specific evidence-based tool, Ask Suicide-Screening Questions,14 (2) further evaluation of positive screens using an evidence-based assessment tool, the Columbia–Suicide Severity Rating Scale,15 (3) collaborative development of individualized safety plans, and lethal means safety counseling, (4) determination of appropriate disposition based on risk categorization, and (5) safer care transitions through supportive contacts and connection to follow-up care. To support implementation and to monitor compliance and outcomes, a toolkit was developed in the electronic health record to centrally house all these elements.16 The suicide safer care bundle was first introduced in outpatient and inpatient mental and behavioral health (BH) settings with a focus on monitoring implementation efforts in various bundle elements.
In the NCH BH specialty setting, one key metric was determining the number of youths screened for suicide risk out of all eligible youths, with a goal of 85% screening compliance for both new patient and 30-day follow-up visits. Following implementation in July 2019, screening compliance for all eligible youths rose from 44% to an average of 79% over an 18-month period, which included the onset and evolution of the COVID-19 pandemic. During this timeframe, the implementation of suicide screening during initial visits was particularly successful, with compliance rates exceeding 90% by the third month of the initiative and remaining at this level since. With changes to the hospital’s suicide prevention policy and toolkit enhancements in subsequent improvement cycles, combined screening rates further increased to 87% and 89%, respectively (Figure 2A). In calendar year 2023, 69 957 suicide screens were completed in 79 054 eligible youths, achieving an 88% compliance rate.
(A) Laney P’ chart displaying completed NCH behavioral health suicide screening in eligible youth from July 2019 through December 2023. B, Laney P’ chart displaying completed NCH non–behavioral health suicide screening in eligible youth from July 2019 through December 2023. ASQ, Ask Suicide-Screening Questions; BH, behavioral health; NCH, Nationwide Children’s Hospital.
(A) Laney P’ chart displaying completed NCH behavioral health suicide screening in eligible youth from July 2019 through December 2023. B, Laney P’ chart displaying completed NCH non–behavioral health suicide screening in eligible youth from July 2019 through December 2023. ASQ, Ask Suicide-Screening Questions; BH, behavioral health; NCH, Nationwide Children’s Hospital.
Recognizing that many youths who die by suicide have health care visits in the year prior to their death,17 the collaborative extended the bundle to areas outside of BH. The NCH primary care network, one of the largest hospital-owned pediatric primary care networks in the nation, implemented the bundle in September 2020 with other pediatric primary and specialty care areas including Developmental and Behavioral Pediatrics, Adolescent Medicine, and School Health following suit. The number of youths screened for suicide risk out of all youth eligible was one of the initial accountability measures in these areas with a goal of 85% screening compliance. In these non-BH care settings, suicide screening has increased 500% over time from a preimplementation baseline of 12% to 72% (Figure 2B).
To achieve community-wide reduction in youth suicides, the suicide safer care bundle needed to be adapted for, and adopted by, health and mental health partners regionally. A 2-year learning collaborative, funded by the Ohio Department of Mental Health and Addiction Services, was launched to facilitate dissemination and implementation of bundle elements in community settings including pediatric primary care, community behavioral health organizations, and regional/community hospital systems. A collaborative toolkit was developed initially using paper forms because of the variability in partner health record platforms. There have been continuing challenges with tracking and accessing data in these real-world community settings; still, 2 pediatric practices focused on completion of suicide screening in eligible youth. At the start of the collaborative, neither practice was engaged in suicide screening, and both set a goal of screening 80% of eligible youth. Practice A increased screening from a baseline of 0% to 89%, whereas Practice B increased from 0% to 44% (Figure 3A, 3B). A community mental health agency that was already screening approximately 75% of eligible patients quarterly when joining the collaborative showed a trend toward an increase in screening compliance, achieving above 90% for 4 consecutive quarters. Another community health agency increased the number of individuals screened for suicide risk from 404 per quarter at the onset of the collaborative to 1049 by the conclusion, a 160% increase.
(A) Laney P’ chart displaying completed suicide screening in a community primary care pediatric practice in youth aged 12 to 19 years presenting for annual well-child visits from October 2022 through February 2024. (B) Laney P’ chart displaying completed suicide screening in a community primary care pediatric practice in youth aged 12 to 20 years presenting for annual well-child visits from March 2022 through February 2024. ASQ, Ask Suicide-Screening Questions; BH, behavioral health; C-SSRS, Columbia–Suicide Severity Rating Scale; EHR, electronic health record; NCH, Nationwide Children’s Hospital; PHQ-9, Patient Health Questionnaire-9.
(A) Laney P’ chart displaying completed suicide screening in a community primary care pediatric practice in youth aged 12 to 19 years presenting for annual well-child visits from October 2022 through February 2024. (B) Laney P’ chart displaying completed suicide screening in a community primary care pediatric practice in youth aged 12 to 20 years presenting for annual well-child visits from March 2022 through February 2024. ASQ, Ask Suicide-Screening Questions; BH, behavioral health; C-SSRS, Columbia–Suicide Severity Rating Scale; EHR, electronic health record; NCH, Nationwide Children’s Hospital; PHQ-9, Patient Health Questionnaire-9.
Transitions of Care
Efficient and effective processes for support during transitions of care were developed to bridge known gaps, especially delayed and nonexistent follow-up care. Caring Contacts, a simple, low-effort intervention consisting of contacting a patient with nondemand supportive messaging, has shown protective effects in suicidal outcomes like ideation, attempts, hospitalizations, and deaths.18,19 Text messaging was chosen as the method of contact for our youth population based on widespread use of technology in this age group and with the belief that messages in this format would normalize the experiences of a system intervention for youth and reduce the stigma often associated with seeking mental health care. At NCH, youth aged 13 years or older who present with suicidal ideation or behavior receive one-way automated caring contact text messages consisting of 19 texts sent over the course of 1 year (first day after enrollment, weekly for 8 weeks, then monthly for 10 months). Since 2019, more than 132 000 separate text messages have been sent to NCH BH patients with a history of suicidal ideation or behavior. Feasibility and acceptability data suggest that most recipients felt supported and hopeful and that messages would help others like them.20 Through collaborations, 11 participating community sites also successfully implemented a Caring Contacts program with more than 1100 unique patients enrolled and more than 13 000 messages sent by partner sites who have sustained this effort.
To increase provider hand-off communication and primary care follow-up after an emergency department crisis visit, a letter detailing essential elements like suicide risk level, recommended timeframe for follow-up, and suggested treatment was created for use by the discharging provider. These letters were offered to specialists for their communication with the primary care provider of record in real time. Within the NCH primary care network, these letters go to a work queue where team members from care coordination, social work, and psychology triage and initiate follow-up contact based on identified need. Ongoing quality improvement work is centered on monitoring and improving the completion percentage of discharge letters following an emergency department crisis visit and expanding the scope to additional primary care sites.
Zero Suicide Adoption in the Community
Teams representing 14 Ohio counites, largely in rural Appalachia, including primary care pediatric practices, behavioral health organizations, and community/regional hospitals, participated in the Zero Suicide Academy sponsored by NCH. The Zero Suicide Academy, offered by the Zero Suicide Institute, is a 2-day didactic suicide prevention and care training initiative for health and behavioral health care organizations committed to reducing suicide among individuals in their care. Teams participated in interactive and small group sessions, providing an opportunity to learn about the Zero Suicide framework and its 7 elements. Site-specific action plans were developed and implemented, guided by results of an Organizational Self-Study designed to help organizations evaluate their current level of suicide care (https://zerosuicide.edc.org/resources/key-resources/organizational-self-study) and a Workforce Survey designed to assess the readiness and confidence of health care providers and staff in providing care to patients at risk of suicide (https://zerosuicide.edc.org/resources/key-resources/workforce-survey).
To enhance safety planning and lethal means safety counseling, a template was adapted from the Stanley and Brown Safety Planning Intervention,21 resulting in a prioritized written list of warning signs, coping strategies, and resources for support during a crisis. Based on this, a safety resource was developed for community use, particularly in pediatric primary care. In partnership with Columbus Public Health, mental health providers at NCH trained in Counseling on Access to Lethal Means (https://zerosuicide.edc.org/resources/trainings-courses/CALM-course) distributed lockboxes to caregivers of youth experiencing an acute suicidal crisis, to limit access to lethal means, particularly firearms and medications, and prevent suicide deaths. Over a 3-year period, 3740 lockboxes with associated counseling were provided to families in Franklin County. In addition, 674 lockboxes were provided to 11 community partners for distribution across central and southern Ohio.
Workforce and Community Education Using Technology
In order to increase access and scale training efforts, NCH has created platforms and content to educate health care providers, families, and advocates, promote mental health for children, and break stigmas. The Behavioral Health Learning Library is a venue for health and behavioral health professionals to access evidence-based/best practices and earn continuing education credits at no cost. Seven courses covering suicide prevention topics from screening and assessment to quality improvement are available (https://www.nationwidechildrens.org/specialties/behavioral-health/for-providers/learning-library).
On Our Sleeves, a program of the Kids Mental Health Foundation, which provides free educational content to communities across the United States focused on normalizing discussions related to child mental health and encouraging help-seeking, has reached more than 6 million people since 2018. Resources for suicide prevention include spotting warning signs, reducing risk in youth of color, and helping a child who has lost someone to suicide (https://kidsmentalhealthfoundation.org/).
Discussion
Our collaborative to prevent youth suicide extends previous work of the Zero Suicide initiative focused on health care system prevention by engaging diverse primary care and behavioral health settings while integrating community prevention strategies and using technology to provide a broader range of tools at various touch points. Zero Suicide offers a common language and set of tools for suicide prevention efforts in large health care settings. Because we are taking a population approach to extend such work beyond the hospital setting into the community, we have introduced additional interventions found to be successful in other settings.
Several lessons have emerged in the process. First, we acknowledge the early stage of our work with an aim of continuous quality improvement. Our goal is the reduction of suicide deaths among youth in Franklin County. Screening rates and other early targets such as the implementation of care bundles are process steps en route to the bigger goal. Time and data will tell if our results are successful, but we will continue to modify interventions and accelerate implementation; current trends look promising given recent data across Franklin and 4 other Ohio counties where interventions have been implemented (Figure 4).
Ohio 5-county 5-year rolling average suicide rate per 100 000 children and youth aged 5 to 19 years.
Ohio 5-county 5-year rolling average suicide rate per 100 000 children and youth aged 5 to 19 years.
Second, the state of data collection, quality, and analysis in nonhospital settings is concerning. Many community providers lack electronic health records, and the ability to implement process tracking and monitor patient outcomes is limited. Collaboratives aiming for broadscale implementation will have to allow more time and resources for data collection in the field.
Third, the engagement of community residents, school representatives, and clinicians has been impressive. Communities are hungry for suicide prevention activities among youth and willing to do the hard work of reviewing the evidence, assessing the fit of interventions, and implementing new protocols. It behooves the field to tap into this energy and earn the trust of the community with careful planning and commitment to address the epidemic of youth suicide.
Dr Chapman conceptualized and designed the project, collected and analyzed the data, drafted and reviewed the manuscript, had full access to all the data, and takes responsibility for the integrity and accuracy of the data. Drs Ackerman and Thomas contributed to the concept and design of the project, reviewed and contributed to the initial draft of the manuscript, and reviewed the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
CONFLICT OF INTEREST DISCLOSURES: The authors have no conflicts of interest to disclose.
FUNDING: The suicide prevention learning collaborative was supported by a grant from the Ohio Department of Mental Health and Addiction Services. The Ohio Department of Mental Health and Addiction Services had no role in the design and conduct of the project.
Comments