Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 24.
Published in final edited form as: School Ment Health. 2023 Feb 24;15(2):583–599. doi: 10.1007/s12310-023-09572-3

Leader and Provider Perspectives on Implementing Safe Alternatives for Teens and Youth – Acute (SAFETY-A) in Public School Districts Serving Racial/ethnic Minoritized Youth

Stephanie H Yu 1, Tamar Kodish 1, Laurel Bear 2, J Conor O’Neill 3, Joan R Asarnow 4, David Goldston 3, Karli K Cheng 5, Xinran Wang 1,6, Sylvanna M Vargas 1,4, Anna S Lau 1
PMCID: PMC10449380  NIHMSID: NIHMS1896431  PMID: 37622166

Abstract

Racial/ethnic minoritized (REM) youth represent a high-risk group for suicide, yet there are striking disparities in their use of mental health services (MHS) even after risk is identified in schools. Prior research suggests that school-based risk assessments and hospitalization encounters can be negatively experienced by REM youth and families, thus deterring likelihood of seeking follow-up care. The Safe Alternatives for Teens and Youth-Acute (SAFETY-A) is a brief, strengths-based, cognitive-behavioral family intervention demonstrated to increase linkage to MHS when implemented in emergency departments. With its focus on strengths and family engagement, SAFETY-A may cultivate a positive therapeutic encounter suited to addressing disparities in MHS by enhancing trust and family collaboration, if appropriately adapted for schools. Thirty-seven school district leaders and frontline school MHS providers from districts serving primarily socioeconomically disadvantaged REM communities participated in key informant interviews and focus groups. First, interviews were conducted to understand usual care processes for responding to students with suicidal thoughts and behaviors, and perspectives on the strengths and disadvantages of current practices. An as-is process analysis was used to describe current practices spanning risk assessment, crisis intervention, and follow-up. Second, focus groups were conducted to solicit perceptions of the fit of SAFETY-A for these school contexts. Thematic analysis of the interviews and focus groups was used to identify multilevel facilitators and barriers to SAFETY-A implementation, and potential tailoring variables for implementation strategies across school districts.

Keywords: youth suicide prevention, school-based implementation, racial/ethnic disparities

As the second leading cause of death among U.S. youth ages 10 – 24, adolescent suicide is a critical public health concern (CDC, 2021). Rates of mental health service (MHS) use for youth with suicidal thoughts and behaviors (STBs) are alarmingly low and estimated at 28.3% (Hom et al., 2015). Furthermore, racial/ethnic minoritized (REM) youth – or those belonging to racial/ethnic groups “distanced from access to power and resources” and “excluded from mainstream social, economic, cultural, or political life” (Racial Equity Tools, n.d.) – are at elevated risk for STBs but are more likely than their White peers to have unmet treatment needs (Ivey-Stephenson et al., 2020; SPRC, 2021). In nationally representative samples of youth with STBs, Latinx youth were 60% less likely to receive outpatient MHS than White youth (Wu et al., 2010), while the odds of receiving outpatient treatment were 63%, 81%, and 97% less for Black, Indigenous, and Asian American/Hawaiian and Pacific Islander youth, respectively, than for White youth (Nestor et al., 2016). Extant research suggests structural and logistical barriers to MHS for REM families, such as transportation, cost, insurance, and lack of linguistically appropriate care, as well as cultural misalignment of MHS provision related to stigma, differing cultural beliefs about mental health, and poor therapeutic alliance (Barnett et al., 2020; Buckingham et al., 2016; Guo et al., 2017; Hom et al., 2015; Lu et al., 2021).

Access to Care within School-based Mental Health Services

Schools often provide first response to youth with suicide risk (Chen et al., 2022) and school-based MHS have been suggested to enhance access to care by serving youth where they are, thereby reducing logistical barriers such as cost and transportation (Bear et al., 2014). Schools are the most common setting in which youth access and use MHS (Duong et al., 2021) and REM youth are more likely to access MHS in schools than in clinics (Cummings et al., 2010; Kataoka et al., 2007). State legislatures are increasingly requiring school districts to adopt suicide prevention policies and practices (Smith-Millman & Flaspohler, 2019). California required local education agencies to adopt training and policies for suicide prevention, intervention, and postvention in middle and high schools by 2017–18 (Assembly Bill [AB] 2246; PSPP, 2016) and in elementary schools by 2020–21 (AB 1767; PSPP, 2019). By 2020, 23 states enacted similar policies, while 33 states mandated at least some form of training (AFSP, 2020).

However, disparities in care continuity after detection of suicide risk persist for REM youth even within school-based MHS. In the Los Angeles Unified School District, only 43% of Latinx students compared to 73% of White students received community-based MHS after being identified with suicide risk (Kataoka et al., 2007). In a medium-sized California school district with a relatively robust school-based MHS system, only 57.1% of Latinx and 43.4% of Asian American students received care following risk assessments (Kim et al., 2018). Ruptures in care continuity for youth with STBs are a priority for suicide prevention efforts and have been suggested to disproportionately impact REM youth (Brent et al., 2020; Fontanella et al., 2020). Thus, it is imperative to identify and address drivers of disparities in care continuity to advance equity in school-based MHS for REM youth with STBs.

Racial/Ethnic Disparities in Follow-up Care Following Risk Detection for Youth in Schools

In introducing the Socio-Cultural Framework for Health Services Disparities, Alegría et al. (2010) asserted that “larger social context raises opportunities and places limits on access, quality, and outcomes of health care” (p. 366) for REM groups. The framework identifies mechanisms that maintain health disparities within interactions between community and care systems at societal, organizational, and individual levels, positing that “each point of interaction between the two major systems – community and treatment – represents a key site of understanding and potential for change” (p. 367). Relevant to the current study, one key point of interaction is the clinical encounter between clients and providers, wherein negative experiences may create and maintain disparities (Alegría et al., 2010).

Indeed, initial encounters with MHS following identification of STBs can be negatively experienced by students and families, thereby deterring pursuit of follow-up care. Prior qualitative work in one school district indicated that families felt excluded from decision-making processes in suicide risk assessments, and students described the experience of emergency transport and hospitalization as traumatic (Kodish et al., 2019). These types of crisis responses can often result in loss of trust in MHS (Jones et al., 2021; Xanthopoulou et al., 2022), which may be especially the case for REM groups with a history of discriminatory treatment in health care (Benjamins & Whitman, 2014). For example, Latina adolescents presenting to the emergency department (ED) with STBs described feeling fearful, stigmatized, alienated, and disempowered (Hausmann-Stabile et al., 2018). Follow-up care rates post-hospitalization are low (43% to 56%), with even lower rates among Latinx and Black youth (Bridge et al., 2012; Fontanella et al., 2020). Racial/ethnic disparities in MHS initiation for students with STBs in schools are partially explained by different rates of caregiver consent to services (Kim et al., 2018), which are linked to delays in MHS initiation (Planey et al., 2019; Yeh et al., 2005). This mirrors disparities in caregiver engagement in treatment processes for Latinx, Black, and Asian American families overall (Barnett et al., 2020; Kapke & Gerdes, 2016), and observations of minimal collaboration with caregivers in routine school-based risk assessments specifically (Kodish et al., 2020). This may relate to previously described structural barriers and cultural misalignment of MHS that impact caregiver opportunity and willingness to engage (Barnett et al., 2020). Addressing these barriers in school-based risk assessment is important since a recent regional analysis found that schools were the most common initiators of youth psychiatric mobile response team and emergency transport to hospitals (Chen et al., 2022).

Promoting Collaborative, Family-Centered Approaches to School-based Risk Assessment

To promote care continuity, collaborative family-centered approaches to school-based risk assessment are needed that can cultivate trust and instill hope within these encounters (Xanthopoulou et al., 2022). The Safe Alternatives for Teens and Youth – Acute (SAFETY-A) is a brief cognitive-behavioral, family-centered therapeutic assessment and suicide prevention intervention shown to be effective for improving rates of follow-up MHS use when delivered in the ED (Asarnow et al., 2009). Through a strengths-based approach, the SAFETY-A consists of individual sessions delivered to the youth, then caregiver, and finally the family together (Asarnow et al., 2011). SAFETY-A also engages trauma informed principles, such as centering youth physical and psychological safety, trust-building, collaboration, and individual empowerment and agency (O’Neill et al., 2021). One essential aspect of SAFETY-A, which departs from suicide risk assessment as usual, is starting by eliciting youth and family strengths to identify reasons for living and sources of protective support. Placing these components early in the intervention sets a tone for a collaborative and empowering interactions that can defuse feelings of stigma, de-escalate unsafe urges, and inform development of a personal safety plan.

By cultivating a collaborative encounter focused on strengths, SAFETY-A may be suited to enhancing trust and fostering positive views of MHS among youth and families to facilitate later care linkage (Hom et al., 2015; Buckingham et al., 2016), which may interrupt the drivers of racial/ethnic disparities in care related to lack of trust and stigma (Alegría et al., 2010). With its focus on caregiver psychoeducation and family engagement, SAFETY-A has potential to reduce racial/ethnic disparities in family acceptance of MHS (Kodish et al., 2019). A collaborative stance can address stigma-related concerns, differing cultural beliefs about mental health, and prior negative experiences with services. In a randomized trial, 92% of youth who received SAFETY-A in the ED accessed follow-up care compared to 76% who received enhanced usual care (Asarnow et al., 2011). Later subgroup analyses revealed that White youth appeared equally likely to receive follow-up care regardless of condition, but SAFETY-A resulted in increased care linkage among REM youth (Kodish et al., 2022). Delivering SAFETY-A in schools may also divert some students from traumatic ED visits or emergency transport to hospitals when the family safety net can be strengthened (Kodish et al., 2019). Thus, implementing SAFETY-A in schools holds promise as a trauma-informed approach to reducing disparities in care continuity and MHS linkage in schools for REM youth (O’Neill et al., 2021).

Implementation Research to Assess the Fit of SAFETY-A for Schools

When transporting an evidence-based practice from one care context (EDs) to another (public schools), it is essential to assess the fit of the innovation with the new service context. The Exploration, Preparation, Implementation, Sustainment (EPIS) framework offers guidance on factors within the outer context (i.e., system), inner context (i.e., organization), and factors related to the innovation being implemented (i.e., fit of the innovation for the new care context), that impact implementation and sustainment (Aarons et al., 2011). Implementing an evidence-based suicide prevention protocol in schools requires strategic planning to attend to resource constraints, workforce capacity, liability concerns, and state and local policies that relate to the outer context (Frank et al., 2021; Stein et al., 2010). Understanding and attending to the inner context of the organization (e.g., climate, culture), leadership, and provider needs is also necessary (Aarons et al., 2011). Given that school districts vary in terms of their practices, policies, and readiness for implementation, implementation strategies may require tailoring across local contexts. Thus, a critical first step is the Exploration phase of EPIS to understand the fit of the evidence-based practice with aspects of the inner and outer contexts of the new practice setting. In this case, we examined the fit of SAFETY-A with the resources and constraints of school districts, workforce readiness for and perceptions of the intervention, and its compatibility with existing policies and workflows in suicide risk assessment (Lyon et al., 2014).

The Current Study

The current study was situated in the initial stages of engaging potential school district partners in the implementation of SAFETY-A in schools primarily serving REM families in Southern California. Toward this goal, we employed a qualitative approach with two primary aims. Aim 1 sought to generate an as-is analysis of the landscape of schools’ current suicide prevention practices and resources based on key informant interviews with school district leaders. Leaders were recruited for interviews given their (1) knowledge of their districts’ suicide prevention protocols, (2) insights into their districts’ resources and readiness for implementing a new evidence-based intervention, and (3) capacity to make decisions about implementing new programs in their districts. Leader interviews also sought to assess their perspectives on their districts’ capacity and readiness for implementation. Aim 2 sought to elicit perceptions of SAFETY-A and its fit for schools to identify potential implementation determinants from focus groups with both district leaders and frontline school MHS providers who conduct school-based risk assessments. Qualitative analyses were conducted to achieve two aims. First, we compared an as-is analysis of routine suicide risk assessment practice with a to-be analysis anticipating how SAFETY-A could be embedded within current district workflows and resources, mapping the processes that diverged between routine practice and the practice innovation. Second, we identified likely determinants of SAFETY-A implementation success based on participants’ perspectives of SAFETY-A’s fit with their school contexts.

Method

Participants

All study procedures were approved by the University of California, Los Angeles Institutional Review Board. Participants were recruited through purposive sampling from Southern California school districts that attended state-mandated AB 2246 suicide prevention trainings. Recruitment proceeded in two stages. First, district leaders – including administrators and directors of mental health, counseling, and student services – were invited via email to conduct key informant interviews. Inclusion criteria included employment within a school district serving primarily REM families, and involvement in their districts’ suicide risk protocols. Leaders from nine out of 10 districts approached were successfully recruited. See Table 1 for district demographics.

Table 1.

School district demographics and characteristics

Demographics District 1 District 2 District 3 District 4 District 5 District 6 District 7 District 8 District 9
Number of schools 19 33 20 11 30 20 42 32 8
Number of students 16,531 25,789 13,802 5,442 25,409 14,875 23,185 23,394 11,404
Students receiving free or reduced lunch, n (%) 10,124 (61.2) 12,956 (50.2) 12,096 (87.6) 3,037 (55.8) 18,386 (72.4) 14,119 (94.9) 19,592 (84.5) 6,765 (28.9) 8,100 (71.0)
English language learners, n (%) 4,040 (24.4) 5,942 (23.0) 3,594 (26.0) 554 (10.2) 8,443 (33.2) 4,624 (31.1) 5,388 (23.2) 3,584 (15.3) 1,069 (9.4)
Student race/ethnicity, %
 Hispanic/Latinx 40.9 20.0 94.0 64.5 95.3 88.3 86.5 32.0 89.2
 Asian American 52.5 15.8 0.2 8.9 2.4 1.2 4.6 33.8 2.6
 Pacific Islander 0.1 0.1 0.2 0.1 0.1 0.6 0.2 0.4 0.2
 White 2.1 59.3 0.6 17.4 1.1 1.0 2.8 20.1 6.1
 Black 0.6 1.1 4.7 4.7 0.3 7.7 2.9 4.1 1.3
 Indigenous or Alaska 0.1 0.1 0.1 0.2 0.2 0.1 0.2 0.1 0.2
 Native
 Multiracial 2.7 3.5 0.1 4.3 0.2 0.9 0.9 8.7 0.2
 Not reported 1.1 0.0 0.0 0.0 0.5 0.1 0.7 0.6 0.1
Open in a new tab

Second, district leaders who participated in interviews were then asked to nominate frontline MHS providers and key members of their districts’ crisis response teams to participate with them in focus groups. District leaders were included given their knowledge of current resources, personnel capacity, and policies. Frontline providers were included to learn from their direct experiences using current suicide prevention protocols with students and caregivers. Based on guidance that 9–17 interviews and 4–8 focus groups provide adequate data saturation (Hennink & Kaiser, 2021), the total sample included 37 unique participants: 15 district leaders who participated in key informant interviews and 33 individuals (11 of the 15 interviewed leaders and 22 frontline providers) who participated in five focus groups. See Table 2.

Table 2.

School leader and provider participant demographics

Demographics Key Informant Interview Participants
(n = 15)		 Focus Group Participants
(n = 33, includes 11 key informant interview participants)		 All Participants
(n = 37)
Age, M (SD) 50.82 (7.23) 47.93 (7.81) 48.03 (7.92)
Gender, n (%)
 Female 12 (80.00) 31 (93.94) 34 (91.89)
 Male 3 (20.00) 2 (6.06) 3 (0.08)
Race/Ethnicity, n (%)
 White 11 (73.33) 19 (57.58) 22 (59.46)
 Hispanic/Latinx (including those also identifying as White) 2 (13.33) 12 (36.36) 12 (32.43)
 Asian American 2 (13.33) 2 (6.06) 3 (0.08)
Staff Role, n (%)
 District administrator 9 (60.00) 12 (36.36) 14 (37.84)
 Community schools program specialist 3 (20.00) 2 (6.06) 3 (0.08)
 Mental health provider
  Clinical supervisor -- 2 (6.06) 2 (0.05)
  Licensed clinician/therapist -- 3 (9.09) 3 (0.08)
  Unlicensed trainee (e.g., social work intern) -- 2 (6.06) 2 (0.05)
 School counselor/psychologist/social worker 3 (20.00) 8 (24.24) 9 (24.32)
 Student well-being liaison -- 2 (6.06) 2 (0.05)
 Principal -- 1 (3.03) 1 (0.03)
 Community mental health agency partner -- 1 (3.03) 1 (0.03)
Additional Demographics Provided by Key Informant Interview Participants Only
Approximate years in current position, M (SD) 4.53 (4.91) -- --
Approximate years in current school district, M (SD) 15.93 (12.09) -- --
Educational background, n (%)
 Master’s Degree
  Administration (e.g., Education Administration) 6 (40.00) -- --
  Counseling/School Counseling 6 (40.00) -- --
  Marriage and Family Therapy 4 (26.67) -- --
  Educational/School Psychology 4 (26.67) -- --
  Social Work 1 (6.67) -- --
  Teaching/Education (e.g., Curriculum and Instruction, Special Education) 5 (33.33) -- --
 Doctorate Degree (e.g., PhD, EdD, PsyD) 4 (26.67) -- --
Professional licenses/credentials, n (%)
 California Teaching Credential 5 (33.33) -- --
 California Administrative Services Credential 10 (66.67) -- --
 Special Education Credential 3 (20.00) -- --
 Pupil Personnel Services (e.g., school counseling) 10 (66.67) -- --
Open in a new tab

The Intervention: Safe Alternatives for Teens and Youth-Acute (SAFETY-A)

The SAFETY-A consists of individual session components delivered to the youth, caregiver, and family together (Asarnow et al., 2011). In addition to behavioral risk assessment and safety planning, SAFETY-A aims to build hope and reasons for living. Session components begin with the youth and caregiver identifying strengths in the youth and family/environment. The youth session includes the use of an emotional thermometer to clarify situations and reactions that increase the risk of suicidal urges and behavior. This information is used to develop a safety plan that specifies situational triggers, warning signs of unsafe feelings, and coping behaviors to down-regulate emotions and suicidal urges. The caregiver session includes counseling in restricting access to potentially fatal methods and coaching on protective and supportive monitoring. In the family session, the youth and caregiver review and practice the safety plan, and commitment to using/supporting the safety plan is assessed. These components are followed by “Care Linkage Contacts” to monitor and motivate linkage to care.

Procedures

Key Informant Interviews

Interviews (n = 15) were semi-structured, conducted via phone, and audio-recorded with oral consent obtained at the outset of the interview by the first author (SHY). Participants received a $20 gift card. The interview was approximately 1 hour and organized into two parts. In Part 1, hypothetical vignettes were presented to obtain a walk-through of the district’s protocol for responding to suicide risk. Specific prompts were used to elicit information on how students are identified with suicide risk, risk assessment tools used, whether/how student outcomes were tracked, rate of transport to hospital or ED, training backgrounds of risk assessors, suicide prevention training provided, implementation support (i.e., funding, supervision, consultation), and types of follow-up MHS provided. In Part 2, participants were asked about perceived strengths and needs of their district’s current protocol, provided a brief description of SAFETY-A, and asked about inner context factors, such as perceptions of district readiness for implementation, workforce capacity, prior district experiences with evidence-based interventions, and barriers that might prevent adoption of SAFETY-A.

Focus Groups

Focus groups were designed to elicit participant perspectives on the fit of SAFETY-A for their districts and to gauge potential for involvement in a partnered implementation effort. Focus groups (n = 5) were conducted, recorded, and facilitated by multiple authors (SHY, TK, LB, JO, ASL) via a HIPAA-compliant video conferencing platform, with oral consent obtained at the outset. Participants received a $50 gift card. Focus groups were two hours long and ranged in size from 6 – 9 participants, with participants representing at least two districts. The focus group began with an orientation to SAFETY-A as delivered in EDs, with components illustrated in videos. Next, focus group participants were asked about SAFETY-A’s fit with inner and outer contexts of their schools in terms of: (1) existing district resources, personnel capacity, and policies, (2) existing risk assessment and crisis intervention procedures, and (3) the needs of students and families served in their school communities.

Data Analysis

Interviews and focus groups were transcribed, and coding was conducted in Dedoose qualitative analysis software (Version 9.0.17). The coding team consisted of SHY and TK. Coding, consensus, and comparison methodology (Willms et al., 1990) was used to develop the coding taxonomies from part 1 of the interview guide for Aim 1 (as-is process analysis of current suicide risk assessment practices), and part 2 of the interview guide for Aim 2 (SAFETY-A implementation determinants). Coders applied an open coding process to all interview transcripts to refine the a priori codes developed from the interview guide and generate additional codes. Coders double coded four interview transcripts using the initial coding taxonomies and met to resolve discrepancies and refine the taxonomies. Coders then independently coded an equal number of the remaining interview transcripts, meeting regularly to ensure consistency and reduce coder drift. An open coding process of the focus group transcripts by both coders yielded no new codes, thus the focus groups were analyzed together with part 2 of the interviews to investigate Aim 2 implementation determinants.

The final coding taxonomy for Aim 1 included 63 codes characterizing current practices for risk assessment, crisis intervention, follow-up procedures, and caregiver involvement. As-is process analysis, primarily described in the business literature, can facilitate understanding of how to “re-engineer” an existing process and develop a plan for redesign to better achieve a goal (Cheung & Bal, 1998; Datta, 1998). Current practice can be visualized in an as-is process model (Cheung & Bal, 1998). Codes indicating specific suicide prevention practices (e.g., safety plan) were enumerated and categorized as they were identified across district leader descriptions and depicted in a flowchart. Frequency of codes was computed at the district-level to illustrate the penetration of current practices across districts. The as-is process model was then juxtaposed with a to-be process model (Okrent & Vovurka, 2004) that mapped the SAFETY-A care process.

The final coding taxonomy for Aim 2 included 37 codes concerning factors that were relevant to potential implementation of SAFETY-A. Nodes denoting implementation facilitators/strengths versus barriers/needs were added to the coding taxonomy. We first employed thematic analysis to integrate codes identified from part 2 of the interviews into thematic patterns (Braun & Clark, 2006). Coders independently reviewed text excerpts for the initial codes, identified key themes, and met to establish consensus. Notably, facilitators/strengths and barriers/needs nodes were often applied for the same codes, where facilitators in one district were identified as barriers in another district. Thus, themes were indicated as being described as barriers, facilitators, or both. Next, the coders independently reviewed field notes from the focus groups and extracted preliminary themes, then reviewed full transcripts for thematic patterns, and finally mapped themes identified from the focus groups onto the themes identified from the interviews. Coders then met to establish consensus and consolidate the final thematic patterns concerning Aim 2 implementation determinants.

Results

Aim 1. Current Suicide Prevention Practices and Resources

As-Is and To-Be Process Models

Figure 1 presents the as-is process model of suicide prevention practices described by leaders in the key informant interviews. These data represent practices described from interview prompts concerning: 1) risk assessment, 2) crisis intervention, 3) follow-up and re-entry procedures, and 4) involvement of caregivers. Shading in the figure depicts the penetration of processes across the districts recruited. Figure 2 presents a to-be process model of how SAFETY-A may be implemented in schools (O’Neill et al., 2021). Shading in the figure depicts SAFETY-A processes that were not frequently described as routine procedures by participants.

Figure 1.

Figure 1

Open in a new tab

As-is process model of suicide prevention practices described by school district leaders representing nine school districts of Southern California in key informant interviews

Figure 2.

Figure 2

Open in a new tab

To-be process model applying Safe Alternatives for Teens and Youth-Acute (SAFETY-A) in schools from the school addendum of the SAFETY-A manual (O’Neill et al., 2021)

Current Resources and Supports for Suicide Prevention Protocols

Training.

Participants from most districts described annual trainings for school-based suicide prevention focused on district protocols, though some reported inconsistent training.

Funding.

Participants from about half of districts described having no dedicated funding to support suicide prevention while others described reliance on external funding (e.g., grants) to maintain or expand current protocols. Participants from multiple districts described allocating funding for suicide prevention from within their Local Control and Accountability Plan (LCAP), which is used to set goals, plan services, and leverage resources to support student outcomes.

Risk Assessment and Referral Tracking.

Participants from most districts described centralized, district-wide tracking of student risk assessments and MHS referrals. Participants in a few districts described tracking only at specific school sites. Participants also varied in their ability to provide exact rates of hospital transport following student risk assessments.

Aim 2. SAFETY-A Implementation Determinants

Seventeen potential determinants of SAFETY-A implementation were identified, with exemplar quotes in Table 3. Themes were identified as (1) exclusively facilitators, (2) exclusively barriers, or (3) both, depending on the attributes of the school district. We describe these latter themes as tailoring variables given their need for tailored implementation strategies.

Table 3.

SAFETY-A implementation determinants: Interview and focus group themes and exemplar quotes

Themes Exemplar Quotes
Facilitators
1. Legislation/policy (AB 2246) driving districtwide action “We are a preschool through transition school district, so when we did our board policy and all that, we were not specific to just 7th to 12th. It was district wide. So, I know they’ve now passed the other one for elementary, while we were already doing it with the whole school district with AB 2246.” [I-13]
2. Collaborative and supportive risk assessment teams “That collaboration and team effort, it’s definitely very, you know, all of us collaborating that means, we are, we donť feel we are alone in the risk assessment, which is definitely very comforting.” [FG-4]
3. Staff dedication to student mental health “The strengths are the caring adults and whatever it takes attitude, right? It’s the people. The strengths are the people and the care.” [I-14]
4. Perceived benefits of SAFETY-A “So this plan, or this program, has in place that opportunity for the parent and the school counselor or the school personnel to really start building that relationship so that hopefully that comfort level then increases and then hopefully there’s more opportunity then for follow through on the parents’ part because they’re building that trust, that trusting relationship. So, I really like that part of the program.” [FG-1]
Barriers
5. Lack of funding “Again can we improve? Yeah, if we had more funding that would help, but you know. So you have to be able to do everything, even though you don’t have funding right?” [I-14]
6. School and individual liability “I just can’t help but worry about the liability of saying okay, I’m working with the student, I’m working with the parent, we have a plan, everything seems to be going great, and then they go home, and then something happens. And then I have to go back to my plan, thinking where did I go wrong?” [FG-1]
7. Challenges with getting everyone “on the same page” “Well we do, but we are also reviewing to make sure, because you know it’s, it’s a constant, the constant challenge to have everyone on the same page, you know?” [I-7]
Subtheme: Variability in provider confidence with risk assessment “I think that some of our academic counselors are more comfortable with the suicide assessment than others. And so, I think there are some that would not be comfortable doing this, as we are already not comfortable of necessarily even with training, doing a suicide risk assessment.” [FG-5]
Subtheme: “Different hats”: Competing staff priorities “I would say the most difficult part about all of this is the fact that school psychologists are in charge and they also wear different hats in special education and are in charge of assessments and timelines and holding [Individualized Education Program] meetings.” [I-10]
8. Long wait times to access community-based MHS “It’s whether or not there’s anybody on the other side who has space in their calendar to take them on and there can be a delay that turns the families off because they get tired of trying to work within the system.” [FG-5]
Tailoring Variables
9. School district culture valuation of student mental health Facilitator: “Because the district is forward thinking, they’re very collaborative and open to new ideas, and it’s a district that really value the mental health of their kids.” [I-9]
Barrier: “But there really isn’t more of a universal approach to really help the entire school community know that mental health is important. It is a need. Our kids are crying out for help...We need to shift our mindset and really put this as a priority.” [I-6]
10. Leadership support for mental health initiatives Facilitator: “We have a new Superintendent who is very, very much invested in our schools’ safety and students’ mental health. And so it’s very easy for me to bring in, you know, a program forward and to say that it’s going to benefit our students.” [I-3]
Barrier: “And so there has to be approval from the district-level in order for changes to truly happen and because there isn’t really much advocacy for mental health at that level, it makes it challenging for the change to happen across the board.” [I-6]
11. Reliance on external agencies for risk assessment and deploying emergency transport Facilitator: “We hardly ever use PMRT. We do everything locally. We train our staff, we train our psychs, our interns…the intern academy, and we do it as a team. And we would like to keep it that way, internally, because we know the students and their families.” [FG-4]
Barrier: “We provide social and emotional support and we do safety assessments. But for hospitalization purposes, PMRT would assess the students. So, we’re not the final determiners of whether a student goes to the hospital or not.” [FG-1]
12. Workforce capacity and provider burden Facilitator: “When there is something like this particular protocol or program or training, it is put throughout the school and we can tap into different individuals because you don’t need to be necessarily, have particular credentials to offer this.” [FG-1]
Barrier: “I'm wondering about the capacity. At the high school we’ve got four counselors and they have a caseload of about 400 each, and it seems like their days are just chock full of academic careers and social-emotional learning, not any deeper extended therapy.” [FG-4]
Subtheme: Presence of licensed mental health clinicians at school sites Facilitator: “Right now is fortunate that in the past five years we’ve had some mental health therapists within our department…we also have a very strong intern Academy.” [FG-4]
Barrier: “We don’t have clinical therapists at our schools. Our school counselors do safety assessments for suicide intervention. And we do rely on PMRT when it’s a high risk.” [FG-1]
13. Provider attitudes towards adopting new practices Facilitator: “I’m so excited about this program that you’re talking about because we absolutely want to have the most up-to-date, evidence-based, cutting-edge practices and procedures and protocols in place to be able to support our programs and families.” [I-7]
Barrier: “Absolutely. And you're always going to have a few people that, been there done that, ‘Wait, another new program?’ You have some of those.” [I-7]
14. Interpretation support for school community threshold languages Facilitator: “And we provide everything in Spanish and English, and now we’re also starting to do it in Mandarin.” [FG-3]
Barrier: “The one concern with some populations might be, as one of our staff mentioned, we think, was the language, right? So, we needed translators and all of that.” [FG-5]
15. Perceptions of stigma and knowledge about mental health in families Facilitator: “Last year we had a foundation, it’s called the With Hope Foundation, that came out from Orange County and did a suicide, a suicide awareness workshop for our parents, and it was well attended, and the parents, they want the information.” [FG-1]
Barrier: “There is, you know, much more stigma in certain populations with mental health and having somebody else be involved and how much you’ll share with outsiders.” [FG-5]
Subtheme: Perceptions of caregiver engagement in risk assessment Facilitator: “I think when I have made calls, they have been supportive. They have been able to kind of follow, kind of, the recommendations that we as a team have provided.” [FG-4]
Barrier: “There are often times where students and parents can’t be in the same room and can’t get to that strengths-based place. And it’s really difficult when you have a student and a parent in the room and the parent is not listening to that student or not validating.” [FG-1]
16. Resources to mitigate cost and insurance barriers Facilitator: “…‘We don’t have the $100,’ and that’s the reason why a child is not being able to service, then we pick up the tab. So, I have a therapist in our department, that we will serve the family and the student at no charge for the family.” [FG-4]
Barrier: “I think everybody else’s concern is when we have this student who’s super high risk, who has frequent suicidality, but they have insurance. They can’t do full-service partnership or come into an agency that’s going to meet them where they’re at. They are kind of stuck.”
[FG-4]
17. Relationship-building and outreach efforts to engage families Facilitator: “I think one of the things that will be key to this is developing that level of trust with the school and the families beforehand…So, I think in addition to our counseling services, and I think the fact that we started with the foundation of [Positive Behavioral Interventions and Supports] at all our school sites will help to make this a little more successful.” [FG-5]
Barrier: “And so the resistance I could see from our newly immigrated families, who are really kind of scared of the whole mental health talk and understanding it, their knowledge is very limited. There is trust issues there with the majority culture.” [FG-4]
“We have families who are living, you know, in our school district who live in fear every day because of immigration statuses and blended family statuses when it comes to immigration, and so there is a resistance, and fear to seek services.” [FG-3]
Open in a new tab

Note: “I” indicates that the quote was stated in one of the 15 interviews; “FG” indicates that the quote was stated in one of the five focus groups.

Facilitators

Four facilitators of SAFETY-A implementation were identified. Participants noted the benefit of legislation/policy (i.e., AB 2246) as a driver of comprehensive, district-wide implementation of suicide prevention training and practices. Participants described their risk assessment teams as collaborative, communicative, and supportive of one another, and frequently described staff dedication and willingness to do “whatever it takes” to support student mental health. Participants reported key perceived benefits of SAFETY-A given its collaborative approach towards engaging families during risk assessment, and potential for reducing traumatic experiences with hospitalization and wait times for emergency transport.

Barriers

Four themes were identified as implementation barriers. Participants described lack of funding to maintain and expand their suicide prevention practices and workforce. Participants also expressed concerns about incurring liability by conducting risk assessment and safety planning themselves rather than mobilizing Psychiatric Mobile Response Teams (PMRT) or transport to the hospital for evaluation. Participants cited challenges with getting all providers “on the same page” with suicide prevention protocols, often stemming from variability in provider confidence with procedures and the “different hats” they wear with respect to educational and mental health priorities. Participants were concerned that long wait times for community-based MHS might deter families and limit effectiveness of Care Linkage Contacts.

Tailoring Variables

Nine themes were identified as potential facilitators or barriers to SAFETY-A, indicating the need to tailor implementation strategies to variable conditions across districts.

School district valuation of student mental health.

While most participants described high valuation of student mental health within their districts, there was some variability. One participant described their district as one that “really value[s] the mental health of their kids” while another participant noted the “need to shift our mindset and really put this as a priority” given lack of focused attention on mental health in their school district.

Leadership support for mental health initiatives.

Most participants described strong leadership support for mental health initiatives at the School Board or district administration levels; however, there was again some variability. One participant indicated leadership support for implementing innovations due to “a new Superintendent who is very, very much invested in our schools’ safety and students’ mental health.” Another participant described more difficulty “because there isn’t really much advocacy for mental health at that level.”

Reliance on external agencies for risk assessment and emergency transport deployment.

Participants described different degrees of reliance on external community-contracted agencies for risk assessment and emergency transport deployment (e.g., PMRT). One participant stated, “we hardly ever use PMRT” as the current norm is to “do everything locally.” In contrast, another participant indicated reliance on PMRT because school district providers are “not the final determiners of whether a student goes to the hospital or not.”

Workforce capacity and provider burden.

Participants varied in their concerns about provider capacity to deliver SAFETY-A components, including consideration of whether districts had licensed mental health clinicians at their school sites. One participant felt that SAFETY-A could flexibly increase local capacity “because you don’t need to have particular credentials” to be trained to deliver SAFETY-A. Yet, another participant shared uncertainty about workforce capacity to implement SAFETY-A, given that providers already “have a caseload of about 400 each” (ratio of counseling staff to students) in their district.

Provider attitudes towards adopting new practices.

Most participants described positive staff attitudes towards adopting new, evidence-based practices, especially if perceived as helpful to students. One participant expressed their own enthusiasm about “absolutely want[ing] to have the most up-to-date, evidence-based, cutting-edge practices and procedures and protocols in place” to support their families. However, the same participant noted the possibility of “a few people” expressing hesitance towards adopting “another new program.”

Interpretation support for school community threshold languages.

Participants described variable resources for delivering SAFETY-A with linguistically minoritized families, such as interpreters or multilingual providers. One participant described already “provid[ing] everything in Spanish and English” and also “starting to do it in Mandarin.” Another participant expressed lack of existing supports within their district and “need[ing] translators.”

Perceptions of stigma and mental health knowledge in families.

Some participants perceived stigma and differing cultural beliefs and knowledge about mental health as barriers to being able to implement SAFETY-A with REM and specifically immigrant families. However, others perceived widespread student and caregiver willingness to learn and engage in workshops, clubs, and assemblies about mental health and suicide prevention. One participant described a suicide awareness workshop as “well attended, and the parents, they want the information.” Another described “more stigma in certain populations with mental health.”

Resources to mitigate cost and insurance barriers.

Participants described variability in resources available to support families with cost and insurance barriers to accessing MHS. One participant described being able to “pick up the tab” and “serve the family and the student at no charge for the family” if they cannot cover the costs of community MHS. Another participant described families being “kind of stuck” when experiencing insurance-related barriers to MHS.

Relationship-building and outreach efforts to engage families.

Participants described differing levels of relationship-building and trust between their districts and school communities. One participant stated that “developing that level of trust with the school and the families beforehand” would be key to SAFETY-A implementation, and that “the foundation of [Positive Behavioral Interventions and Supports] at all our school sites will help to make this a little more successful.” Another participant described potential engagement challenges with “our newly immigrated families” due to “trust issues there with the majority culture.”

Discussion

Implementing a strengths-based, trauma-informed family intervention like SAFETY-A in schools can address potential drivers of racial/ethnic disparities in MHS use for youth with STBs. A community network of school collaborators was assembled to examine SAFETY-A’s fit with current practices, identify implementation determinants, and suggest considerations for tailoring implementation to meet the needs of REM families in a range of school districts.

Given that limited research has explored routine suicide risk assessment practices in schools (Crepeau-Hobson, 2013), we first sought to understand the landscape of suicide prevention practices and resources in local school districts, which varied in the comprehensiveness of their protocols and use of specific risk assessment practices. Overall, the as-is analysis reflected a rather impressive level of existing infrastructure despite limited resources, and considerable alignment with the to-be process model. Participants described multiple practices in their risk assessment protocol that were aligned with aspects of SAFETY-A (e.g., safety planning), as well as best practice recommendations for suicide prevention overall (Brodsky et al., 2018), with somewhat more variable report of other practices (e.g., means restriction). SAFETY-A can bolster existing processes. For example, participants reported using standard risk assessment questions (e.g., Columbia-Suicide Severity Rating Scale; Posner et al., 2011), which can be augmented by behavioral assessment through SAFETY-A components (e.g., can the youth identify strengths, warning signs, and coping resources to build a safety plan?).

Among the SAFETY-A processes not reflected in the as-is analysis were features of the strengths-based approach, with discussion of youth and family strengths to instill hope and build rapport. To address mechanisms implicated in MHS disparity frameworks, enhanced efforts to cultivate a positive therapeutic encounter may be especially important for earning trust among immigrant and REM communities with a longstanding history of discrimination experiences in care systems (Alegría et al., 2010; Benjamins & Whitman, 2014). Research in the ED suggests that suicide risk assessments instilled hope when they involved unscripted conversation, warmth, and a collaborative safety plan, rather than a sole focus on gathering information to determine imminence of risk, which engendered hopelessness and loss of trust (Xanthopoulou et al., 2022).

In the as-is process, students identified with low risk for suicide are referred to follow-up care depending on school resources, community partnerships, and the families’ insurance. SAFETY-A can augment MHS linkage through Care Linkage Contacts to structure consistent outreach and follow-up. For students whose risk is deemed as imminent or unclear, assessors in the as-is process often seek consultation, primarily internally with a district administrator but sometimes with external community partners and crisis services (e.g., PMRT) for determination on transport/hospitalization. Thus, in some districts, emergency transport may be mobilized for students when determination of risk is outsourced. If SAFETY-A were to be adopted in schools, emergency transport would be deployed only when students were assessed as with imminent risk following the SAFETY-A care process designed to build hope, de-escalate suicidal urges, wrap the youth in family support, and build a safety plan. One potential advantage could be to divert youth from hospitalization if adequate supports are observed through family-centered safety planning, which may reduce REM youth experiences of traumatic hospitalizations that can deter follow-up MHS use (Kodish et al., 2019; Jones et al., 2021). In both as-is and to-be processes, there was similar representation of school re-entry support and MHS linkage.

Perhaps the largest gap between the as-is and to-be process models was in how caregivers are involved in the risk assessment process. While participants described some caregiver involvement, these practices appeared to be the least structured aspect of usual care protocols. Participants often described notifying caregivers after emergency transport had been mobilized for the student, or keeping students and caregivers separate during the risk assessment out of concern that caregiver reactions might impact the nature of the information shared by students. These findings are aligned with prior qualitative reports suggesting that caregivers are routinely notified in the wake of suicide risk being detected, but that caregiver outreach and engagement post-crisis is less systematic (Nadeem et al., 2013). In SAFETY-A, caregivers are integrated into risk assessment procedures, and there is a focus on enhancing caregiver engagement and collaboration to support student safety. Given observed racial/ethnic disparities in caregiver consent for youth MHS following school-based risk assessment (Kim et al., 2018), as well as treatment engagement overall, collaborative, family-centered approaches may be critical for increasing MHS linkage among REM families (Marraccini et al., 2020).

The second aim was to identify likely determinants of SAFETY-A implementation success based on participants’ perspectives of SAFETY-A’s fit with their school contexts. Tailoring implementation strategies and interventions for low-resource contexts is vital to prevent the possibility of widening disparities since barriers to implementation success tend to be concentrated where minoritized families are served. In the outer context, participants described the effectiveness of state policy like AB 2246 for driving action to improve school-based suicide prevention. However, participants reported that current funding and resources are generally inadequate to support protocols to comply with unfunded mandates. System-level changes, such as increased funding for districts to allocate towards suicide prevention, are crucial for reducing school and provider burden and boosting reach and effectiveness of protocols. Likewise, access barriers to MHS result in long wait times that can deter families and limit effectiveness of patient navigation and motivational interventions such as SAFETY-A’s Care Linkage Contacts. In these situations, structural care access barriers must be addressed first (Alvarez et al., 2022).

Within the inner context, participants consistently described their risk assessment teams as collaborative and supportive, indicating positive organizational climate. Many participants also described their district’s organizational culture as having high valuation of student mental health and strong leadership support that would facilitate SAFETY-A implementation. Positive organizational climate, proficient organizational culture, and leadership support for initiatives are positively associated with school implementation outcomes (Frank et al., 2021; Stein et al., 2010). Districts where participants described less conducive culture and leadership support may require relationship building and motivational enhancement prior to implementation.

Another inner context variable is absorptive capacity, or the ability to incorporate new knowledge and skills based on existing knowledge and skills (Aarons et al., 2011). Participants described different levels of reliance on outside agencies for suicide risk assessment. Although having community partnerships is a strength, reliance on outside expertise may reduce school opportunities to build internal capacity for risk assessment, as well as relationships with the school community. Thus, implementation of SAFETY-A across districts will require tailoring to their local infrastructure. For example, school districts may be able to build on current partnerships to receive consultation as a part of their protocol while still strengthening internal capacity to conduct risk assessments (Kodish et al., 2019). As another example, districts varied in terms of multilingual provider capacity and language interpreter support. Lack of linguistically appropriate care is a commonly cited structural barrier contributing to racial/ethnic disparities (Barnett et al., 2020). Such resources may need to be installed for SAFETY-A implementation in schools serving multilingual communities with non-English speaking caregivers.

For provider factors, participants emphasized providers’ dedication and willingness to go above and beyond their duties to support student mental health. However, evident strengths of their workforce are tempered by capacity and resource constraints related to the realities of their day-to-day workload. These themes are aligned with prior provider reports of resource limitations as barriers to implementing EBPs in school settings (Lyon et al., 2013). Furthermore, determinations about imminent risk may require clinical decisions that can feel beyond the scope of their training and role (Walsh et al., 2013). Participants cited concerns with liability, getting all providers “on the same page,” and provider confidence with risk assessment. Providing accessible training (e.g., online, alternating times) and ongoing consultation can reduce provider discomfort and build self-efficacy with suicide prevention protocols (Walsh et al., 2013). Enhanced educational implementation strategies (e.g., hands-on training with active teaching strategies for skill-building, practice, performance feedback) may be indicated in districts with few clinically trained staff (Walsh et al., 2013). While implementing best practices in suicide risk assessment may be the best safeguard against liability concerns, school districts are likely to experience tension when weighing the benefits of adopting SAFETY-A with the risk of incurring liability. Some school districts may ultimately choose not to assume liability and continue to outsource risk assessment and safety planning (e.g., to PMRT). However, disadvantages of this choice include current pressures on the MHS system (illustrated by long wait times for PMRT; Kodish et al. 2019), missed opportunities for relationship-building with the school community, and having students undergo ED visits or hospitalization experiences that are potentially avoidable when the family safety net can be strengthened. These cost-benefit analyses should be discussed with school partners in the exploration phase of implementation.

Within the school community, participants perceived potential barriers to SAFETY-A among REM families, augmented by fear of MHS-seeking specifically in immigrant families with undocumented and mixed immigration status. Yet, participants also recognized families’ willingness to learn and engage in suicide awareness when opportunities were offered. Intervention adaptations may be needed to alter the language of programs and materials to reduce potentially stigmatizing language, as well as identify strategies for enhancing emotional communication in students and families who may be less accustomed to conversations about mental health. Furthermore, resources may need to be identified, such as no-cost MHS or partnerships with community legal aid organizations, to assist undocumented families without insurance. Exclusionary immigration policies deter help-seeking among families with undocumented and mixed immigration status and exacerbate disparities at a structural level (Perreira & Pedroza, 2019). Proactively providing psychoeducation to immigrant families can allow opportunities for them to learn about available safe and accessible MHS resources.

Finally, participants had positive impressions of SAFETY-A’s collaborative approach to engaging the family to facilitate MHS linkage. Intervention adaptations and relationship-building efforts, as well as tailoring of as-is school policies and procedures, may be needed to foster safety and trust between the school and REM families, particularly in districts with more fragmented relationships with their school communities. Examples include eliminating stigmatizing aspects of as-is procedures, such as intrusive personal searches, involvement of school police, and physical restraint in emergency transport, whenever possible (Kodish et al., 2019). For situations in which means restriction is indicated, taking a collaborative approach to personal searches may reduce perceptions of this practice as punitive.

Study limitations must be considered. First, the study took place in a relatively progressive region with state mandated policies on school suicide prevention. Findings may not be generalizable to districts with different state and local mandates, as well as different resource supports and school communities. Second, although our approach enabled us to solicit rich qualitative data, this also reduced the uniformity of the data collected. Participants may not have described all district practices, perhaps due to varied experiences with and knowledge of their districts’ protocols depending on their roles. Third, respondents may have also been susceptible to biases, such as presenting their districts’ practices more positively. A systematic review of school risk assessment and suicide prevention protocols could validate the current study’s qualitative work on districts’ as-is procedures. Fourth, the study focuses on leader and provider perspectives and does not include essential perspectives from students and families themselves on the acceptability and helpfulness of SAFETY-A. To be able to approach students and families who might receive SAFETY-A, it was necessary to first partner with district leaders and providers in a position to make decisions about implementing new programs in their districts. However, district leaders and providers may not be positioned to accurately report on the fit of SAFETY-A for families. In the next stage of our work, now funded by an exploratory intervention development grant (R34MH126670), we are gathering student and caregiver perceptions of SAFETY-A to inform potential adaptations that may enhance its cultural responsiveness.

The current study suggested several facilitators of SAFETY-A implementation, including a dedicated workforce and leadership who care immensely about their students’ mental health. These strengths are tempered by systemic barriers that constrain resources and contribute to provider burden and hesitation. Discussions also revealed considerable district-level variability in perceived integrity and uniformity of current suicide prevention practices, workforce capacity and coordination, and perceived readiness for SAFETY-A implementation, indicating the need to tailor implementation strategies to address district variables across multiple levels for systems of care serving REM communities. Future work will build on the current study’s findings, while gathering perspectives from youth and caregivers on the acceptability of SAFETY-A. It is critical to partner with and incorporate REM youth and caregiver voices given that they are in the best position to provide feedback on the relevance of SAFETY-A for their communities. School community perspectives will inform adaptations to SAFETY-A intervention components and implementation strategies, which will then be iteratively refined and tested with school districts to address the critical youth suicide crisis for REM youth and families.

Acknowledgements:

This research was supported by funds from the University of California, Los Angeles (UCLA) Immigrant Youth Task Force and UCLA Asian American Studies Center’s Patrick & Lily Okura Research Grant on Asian Pacific American Mental Health awarded to the first author (Stephanie H. Yu). Additional funding was provided by: U79SM080041 awarded by the Department of Health and Human Services, Substance Abuse and Mental Health Services Administration (SAMHSA; Asarnow and Goldston, Co-PIs) and R34MH126670 awarded by the National Institute of Mental Health (NIMH; Lau, PI). We also thank Dylan Aguirre and Ana Aguirre for their assistance in transcribing several interview and focus group recordings.

Footnotes

Declaration of Conflicting Interests:

The Author(s) declare that there is no conflict of interest.

References

  1. Aarons GA, Hurlburt M, & McCue Horwitz S (2011). Advancing a conceptual model of EBP implementation in public service sectors. Adm Policy Ment Health, 38(1), 4–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. AFSP. (2020). State laws: Suicide prevention in schools (K-12). https://www.datocms-assets.com/12810/1602535612-k-12-schools-issue-brief-10-12-20.pdf
  3. Alegría M, Pescosolido BA, Williams S, & Canino G (2011). Culture, race/ethnicity and disparities: Fleshing out the Socio-Cultural Framework for Health Services Disparities. In Pescosolido BA, Martin J, McLeod J, & Rogers A (Eds.), Handbook of the sociology of health, illness, and healing (pp. 363–382). Springer, New York, NY. [Google Scholar]
  4. Alvarez K, Polanco-Roman L, Breslow AS, & Molock S (2022). Structural racism and suicide prevention for ethnoracially minoritized youth: A conceptual framework and illustration across systems. The American Journal of Psychiatry, 179, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Asarnow JR, Baraff LJ, Berk M, Grob CS, Devich-Navarro M, Suddath R, Piacentini JC, Rotheram-Borus M, Cohen D & Tang L (2011). An ED intervention for linking pediatric suicidal patients to follow-up MH treatment. Psychiatr Serv, 62(11), 1303–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Asarnow JR, Berk MS, & Baraf LJ (2009). Family intervention for suicide prevention: A specialized ED intervention for suicidal youths. Prof Psychol Res Pr, 40(2), 118–125. [Google Scholar]
  7. Barnett ML, Lau AS, Lind T, Wright B, Stadnick NA, Innes-Gomberg D, Pesanti K & Brookman-Frazee L (2020). Caregiver attendance as a quality indicator in the implementation of multiple evidence-based practices for children. JCCAP, 49(6), 868–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bear L, Finer R, Guo S, & Lau AS (2014). Building the gateway to success: An appraisal of progress in reaching underserved families and reducing racial disparities in school-based mental health. Psychol Serv, 11(4), 388–397. 10.1037/a0037969 [DOI] [PubMed] [Google Scholar]
  9. Benjamins MR, Whitman S (2014). Relationships between discrimination in health care and health care outcomes among four race/ethnic groups. J. Behav. Med, 37, 402–413. [DOI] [PubMed] [Google Scholar]
  10. Braun V & Clarke V (2006). Using thematic analysis in psych. Qual Res Psychol, 3(2), 77–101. [Google Scholar]
  11. Brent DA, Goldstein TR, & Benton TD (2020). Bridging gaps in follow-up appts after hospitalization and youth suicide: MH care disparities matter. Pediatrics, 3(8), e2013100. [DOI] [PubMed] [Google Scholar]
  12. Bridge JA, Marcus SC, & Olfson M (2012). Outpatient care of young people after emergency treatment of deliberate self-harm. J Am Acad Child Adolesc Psychiatry, 51(2), 213–222. [DOI] [PubMed] [Google Scholar]
  13. Brodsky BS, Spruch-Feiner A, & Stanley B (2018). The zero suicide model: Applying evidence-based suicide prevention practices to clinical care. Front. Psychiatry, 9, 33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Buckingham SL, Brandt NE, Becker KD, Gordon D, & Cammack N (2016). Collaboration, empowerment, and advocacy: Consumer perspectives about treatment engagement. Journal of Child and Family Studies, 25, 3702–3715. [Google Scholar]
  15. CA Dep’t of Education. (n.d.). CA school directory. https://www.cde.ca.gov/SchoolDirectory/
  16. Chen BC, Lui JHL, Liu V, Wright BA, Benson LA, Yen-Jui LR, & Lau AS (2022). Mapping racial/ethnic disparities in youth psych emergencies & associations with neighborhood racial composition, child opportunity, & inequality. Clin Psychol Sci. [Google Scholar]
  17. Cheung Y & Bal J (1998). Process analysis techniques and tools for business improvements. Bus. Process Manag, 4(4), 274–290. 10.1108/14637159810238174 [DOI] [Google Scholar]
  18. Crepeau-Hobson F (2013). An exploratory study of suicide risk assessment practices in the school setting. Psychology in the Schools, 50(8), 810–822. [Google Scholar]
  19. Cummings JR, Ponce NA, & Mays VM (2010). Comparing racial/ethnic differences in MHS use among high-need subpopulations across clinical and school-based settings. J Adolesc Health, 46, 603–6. 10.1016/j.jadohealth.2009.11.221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Centers for Disease Control and Prevention. (2021. December 2). Web-based Injury Statistics Query and Reporting System. https://www.cdc.gov/injury/wisqars/index.html
  21. Datta A (1998). Automating the discovery of as-is business process models: Probabilistic and algorithmic approaches. Information Systems Research, 9(3), 275–301. [Google Scholar]
  22. Duong MT, Bruns EJ, Lee K, Cox S, Coifman J, Mayworm A, & Lyon AR (2021). Rates of MH service utilization by children & adolescents in schools & other common service settings: A systematic review & meta-analysis. Adm Policy Ment Health, 48, 420–39. [DOI] [PubMed] [Google Scholar]
  23. Fontanella CA, Warner LA, Steelesmith DL, Brock G, Bridge JA, & Campo JV (2020). Association of timely outpatient MH services for youths after psychiatric hospitalization with risk of death by suicide. JAMA Netw. Open, 3(8), e2012887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Frank HE, Saldana L, Kendall PC, Schaper HA & Norris LA (2021). Bringing evidence-based interventions into the schools: An examination of organizational factors and implementation outcomes. Child & Youth Services. Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Guo S, Kim JJ, Bear L, & Lau AS (2017). Does depression screening in schools reduce adolescent racial/ethnic disparities in accessing treatment? JCCAP, 46(4), 523–536. [DOI] [PubMed] [Google Scholar]
  26. Hausmann-Stabile C, Gulbas L, & Zayas LH (2018). Treatment narratives of suicidal Latina teens. Arch Suicide Res, 22(1), 165–72. 10.1080/13811118.2017.1304305 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hennink M & Kaiser BN (2021). Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Soc Sci Med. Advance online publication. [DOI] [PubMed] [Google Scholar]
  28. Hom MA, Stanley IH, & Joiner TE Jr. (2015). Evaluating factors and interventions that influence help-seeking and mental health service utilization among suicidal individuals: A review of the literature. Clinical Psychology Review, 40, 28–39. [DOI] [PubMed] [Google Scholar]
  29. Ivey-Stephenson AZ, Demissie Z, Crosby AE, Stone DM, Gaylor E, Wilkins N, Lowry R, & Brown M (2020). Suicidal ideation and behaviors among high school students — Youth Risk Behavior Survey, United States, 2019. MMWR, 69(Suppl-1), 47–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jones N, Gius BK, Shields M Collings S, Rosen C & Munson M. (2021). Investigating the impact of involuntary psychiatric hospitalization on youth and young adult trust and help-seeking in pathways to care. Soc Psychiatry Psychiatr Epidemiol, 56, 2017–2027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kapke TL, & Gerdes AC (2016). Latino family participation in youth mental health services: Treatment retention, engagement, and response. Clin Child Fam Psychol Rev, 19,329–51. [DOI] [PubMed] [Google Scholar]
  32. Kataoka S, Stein BD, Nadeem E, & Wong M (2007). Who gets care? MHS use following a school-based suicide prevention program. J Am Acad Child Adolesc Psych, 46, 1341–48. [DOI] [PubMed] [Google Scholar]
  33. Kim JJ, Kodish T, Bear L, El-Hendi T, Duong J, & Lau AS (2018). Disparities in follow-up care for AA youth assessed for suicide risk in schools. AAJP, 9(4), 308–317. [Google Scholar]
  34. Kodish T, Kim JJ, Le K, Yu SH, Bear L, & Lau AS (2019). Multiple stakeholder perspectives on school-based responses to student suicide risk in a diverse public school district. School Mental Health, 12, 336–52. 10.1007/s12310-019-09354-w [DOI] [Google Scholar]
  35. Kodish T, Lau AS, Belin T, Berk M, & Asarnow JA (2022). Improving care linkage for racial-ethnic minority youths receiving emergency department treatment for suicidality: SAFETY-A. Psychiatric Services. Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Okrent MD & Vokurka RJ (2004). Process mapping in successful ERP implementations. IMDS, 104(8), 637–643. 10.1108/02635570410561618 [DOI] [Google Scholar]
  37. Lu W, Tod-hunter-Reid A, Mitsdarffer M, Muñoz-Laboy M, Yoon AS, & Xu L (2021). Barriers and facilitators for mental health service use among racial/ethnic minority adolescents: A systematic review of literature. Frontiers in Public Health, 9, 641605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lyon AR, Ludwig K, Romano E, Leonard S, Vander Stoep A & McCauley E (2013). “If it’s worth my time, I will make the time”: School-based providers’ decision-making about participating in an EBP consultation program. Adm Policy Ment Health, 40, 467–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lyon AR, Ludwig K, Romano E, Koltracht J, Vander Stoep A, & McCauley E (2014). Using modular psychotherapy in school mental health: Provider perspectives on intervention-setting fit. JCCAP, 43(6), 890–901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Marraccini ME, Griffin D, O’Neill JC, Martinez R, Chin A, Toole E, Grapin SL, & Naser SC (2021). School risk and protective factors of suicide: A cultural model of suicide risk and protective factors in schools. School Psychology Review, 51, 266–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Nadeem E, Santiago CD, Kataoka SH, Chang VY, & Stein BD (2016). school personnel experiences in notifying parents about their child’s risk for suicide: Lessons learned. The Journal of School Health, 86(1), 3–10. 10.1111/josh.12346 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Nestor BA, Cheek SM, & Liu RT (2016). Ethnic and racial differences in mental health service utilization for suicidal ideation and behavior in a nationally representative sample of adolescents. J Affect Disord, 202, 197–202. 10.1016/j.jad.2016.05.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. O’Neill JC, Goldston DB, Kodish T, Yu SH, Lau AS, & Asarnow JR (2021). Implementing trauma informed suicide prevention care in schools: Responding to acute suicide risk. EBPCAMH, 6(3), 379–392. 10.1080/23794925.2021.1917019 [DOI] [Google Scholar]
  44. Perreira KM, & Pedroza JM (2019). Policies of exclusion: Implications for the health of immigrants and their children. Annual Review of Public Health, 40, 147–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Planey AM, McNeil Smith S, Moore S, & Walker TD (2019). Barriers and facilitators to mental health help-seeking among African American youth and their families: A systematic review study. Children and Youth Services Review, 101, 190–200. [Google Scholar]
  46. Pupil Suicide Prevention Policies, Cal. Assemb. B. 2246 (2015–2016), Ch. 694 (Cal. Stat. 2016). https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201520160AB2246
  47. Pupil Suicide Prevention Policies, Cal. Assemb. B. 1767 (2019–2020), Ch. 642 (Cal. Stat. 2019). https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201920200AB1767
  48. Racial Equity Tools. (n.d.). Racial Equity Tools Glossary. https://racialequitytools.org/glossary
  49. Smith-Millman S & Flaspohler D (2019). School-based suicide prevention laws in action: A nationwide investigation of principals’ knowledge of and adherence to state school-based suicide prevention laws. School Mental Health, 11(6), 321–334. [Google Scholar]
  50. SPRC. (2021). Racial and ethnic disparities. https://sprc.org/scope/racial-ethnic-disparities
  51. Stein BD, Kataoka SH, Hamilton AB, Schultz D, Ryan G, Vona P & Wong M 2010). School personnel perspectives on their school’s implementation of a school-based suicide prevention program. J Behav Health Serv Res, 37, 338–349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Walsh E, Hooven C, & Kronick B (2013). School-wide staff and faculty training in suicide risk awareness: Successes and challenges. J Child Adolesc Psychiatr Nurs, 26(1), 53–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Willms DG, Best JA, Taylor DW, Gilbert JR, Wilson DM, Lindsay EA, et al. (1990). A systematic approach for using qualitative methods in primary prevention research. Medical Anthropology Quarterly, 4(4), 391–409. [Google Scholar]
  54. Wu P, Katic BJ, Liu X, & Fan B (2010). Mental health service use among suicidal adolescents: Findings from a US national community survey. Psychiatr Serv,61(1),17–24. [DOI] [PubMed] [Google Scholar]
  55. Xanthopoulou P, Ryan M, Lomas M, & McCabe R (2022). Psychosocial assessment in the ED: Experiences of people presenting with self-harm & suicidality. Crisis, 43, 299–306. [DOI] [PubMed] [Google Scholar]
  56. Yeh M, McCabe K, Hough RL, Lau A, Fakhry F, & Garland A (2005). Why bother with beliefs? Examining relationships between race/ethnicity, parental beliefs about causes of child problems, and mental health service use. JCCP, 73, 800–807. [DOI] [PubMed] [Google Scholar]

RESOURCES