Suicide Postvention in Schools: What Evidence Supports Our Current National Recommendations?

Abstract

Suicide is a leading cause of death for school-aged preteens and adolescents and a growing risk for younger children. Schools are the ubiquitous institutional context serving this age group. These trends suggest a need for knowledge and guidance related to school postvention efforts, yet the available research is limited. Focusing on postvention, or the period after a peer suicide occurs, is critical to youth suicide prevention because this is a time of elevated suicide risk for youth. Targeted postvention interventions in schools can mitigate youth suicide risk and limit contagion within a school’s student body. This article explores the scientific literature related to school-based suicide postvention, describing the strength and limits of research supporting common recommendations for suicide postvention in schools. It identifies widespread recommendations for school postvention that have only preliminary supportive evidence and notes several areas in need of additional research. With clearer postvention best practices to guide their suicide crisis preparedness plans and postvention procedures, schools can better support students, families, and the community as a whole in order to prevent further tragedies.

Suicide postvention is an understudied area of vital importance to youth suicide prevention. Defined by Andriessen (2009), suicide postvention encompasses “those activities developed by, with, or for suicide survivors … to facilitate recovery after suicide, and to prevent adverse outcomes including suicidal behavior” (p. 43), specifically for suicide loss survivors. Schools are a critical context for youth suicide prevention (Singer et al., 2019). There are, however, some discrepancies between research and practice related to suicide postvention planning and best practices implementation (DeHart, 2019).

Putting this research into practice is especially important because youth suicide rates have increased over the past several decades (Capps et al., 2019; O’Neill, 2017; Owens, 2014). These substantial increases have occurred across gender, race, and ethnic groups, especially among Black youth (Gordon, 2020) and girls (Shain, 2019). Currently, suicide is the second leading cause of death among young people aged fifteen to twenty-four years and the third leading cause of death for children aged ten to fourteen years, with no difference in youth suicide rates by socioeconomic status (Braudt et al., 2019). This youngest age group has experienced significant increases in rates of suicide deaths since 1999 (Hedegaard et al., 2018).

Research shows that adolescents and young adults are more vulnerable to suicide after exposure to a suicide loss or a suicide attempt (Abrutyn & Mueller, 2014; Randall et al., 2015). This effect is sometimes called suicide contagion with reference to the more general social network theories (Kadushin, 2012), although this terminology has its critics (Abrutyn et al., 2019). Youth are also more likely than older age groups to die by suicide as part of a suicide cluster—a disproportionate rate of suicides occurring in a particular place or a limited period of time (O’Neill, 2017; O’Neill et al., 2020). Although suicide is a rare occurrence, each year there are an estimated one hundred to two hundred potentially preventable deaths due to suicide contagion and clusters.

Exposure to a peer’s suicide can result in suicidal ideation or behavior in adolescents and last for years, especially if these adolescents possess other risk factors (O’Neill, 2017; Poland et al., 2019). Research by Abrutyn and Mueller (2014) showed negative effects from exposure to a peer’s suicide up to four years later, exclusive of other risk factors, although this largely operates through lingering emotional distress. Suicide loss survivors, or people who have lost a significant other (or a loved one) to suicide, are also known to be at an elevated risk for suicide (Andriessen, 2009; Andriessen, Krysinska, Kõlves, et al., 2019). Finally, incidents of suicide clustering and suicide contagion have been repeatedly documented in settings with clear boundaries between group members and outsiders, settings like high schools (Haw et al., 2013). Other research has shown that schools vary significantly in their rates of suicidality (Benbenishty et al., 2018).

Having discernable best practices for postvention is especially important to prevent youth suicide. Because young people spend a large portion of their institutional lives in schools, this institution is key to postvention efforts (Wyman, 2014). The ability of schools to mobilize mental health professionals and other vital community stakeholders is a preventative response that can reduce risk (O’Neill, 2017; O’Neill et al., 2020). Schools possess an organizational structure beneficial to implementing evidence-based postvention that can have significant impact and prevent youth suicide deaths.

The purpose of this article is to examine the limitations and strengths of evidence supporting best practice recommendations related to postvention in general and school-based suicide postvention specifically. This article clarifies the kinds of research that produced postvention recommendations and synthesizes what we scientifically know (and its limitations) about school-based suicide postvention. By outlining the evidence supporting widespread recommendations, we point to the gaps in our knowledge base and identify the recommendations with stronger scientific support. We seek to clarify the specific procedures that may be important for schools to develop well before postvention is needed. We organize the recommendations using a public health framework from universal to selective and targeted prevention in terms of suicide crisis preparedness and active suicide postvention. This information can assist schools and school social workers in developing thorough and evidence-informed crisis plans and strategies to support peer suicide survivors and to mitigate suicide contagion. The results will not only help school social workers and administrators develop evidence-based protocols for school crisis planning, but they will also highlight the need for more research in particular areas within school-based suicide postvention.

Approach

To provide a clear perspective about the state of the science of postvention in general and as it specifically relates to schools, we reviewed relevant literature published between 1985 and 2019, which is the basis of postvention recommendations of leading suicide prevention organizations. We started with several Google Scholar searches associated with postvention using various combinations of the terms suicide, postvention, schools, education, program evaluation, systematic, reviews, and effectiveness. We scanned the titles of the resulting articles to assess the relevance of postvention interventions, practices, and related theories or inquiries to inform school-based methods of reducing youth suicide risk after a suicide death. We excluded books and gray literature such as conference abstracts, presentations, and reports and included articles focusing on postvention planning, school-based interventions, staff trainings, and mitigation of the increased risk to individuals bereaved by suicide (suicide contagion). We omitted articles focused on describing grief and bereavement in general and retained those focused on suicide grief and bereavement.

Because many of the articles we found cited the publication After Suicide: A Toolkit for Schools (American Foundation for Suicide Prevention & Suicide Prevention Resource Center, 2018) as a primary reference for schools to use when developing their response to a suicide death, we also explored the research cited in the suicide contagion section of the toolkit. This resource included three articles not previously identified in our literature search (Insel & Gould, 2008; Lake & Gould, 2013; Zenere, 2009). These additional articles included three more references related to school postvention (Bailley et al., 1999; Begley & Quayle, 2007; Johansson et al., 2006). Lastly, an advanced Google Search for articles to date whose titles included the three terms suicide, postvention, and school yielded fifty-three search results. Of these results all were previously identified, not peer reviewed, or not in English except one reference (Mirick et al., 2018), which we included in our study.

Findings

Much of the literature in this review was used to develop the postvention recommendations made by the American Foundation for Suicide Prevention (AFSP) and the Suicide Prevention Resource Center (SPRC), the leading research dissemination platforms for suicide prevention in the United States. These entities present evidence-based best practices for suicide postvention, disseminating information nationally that guides social workers, researchers, other mental health practitioners, organizations, schools, and communities. In 2018, the AFSP and SPRC collaborated to synthesize best practices, publishing a toolkit that serves as a research-based guide to address needs in the aftermath of a suicide.

In this article, we consider the toolkit’s recommendations within the context of the research that produced them, specifically discussing the scientific limitations or gaps in knowledge about school-based suicide postvention. This review clarifies the kinds of research that produced particular school recommendations and highlights the need for more specific school-based suicide postvention research. The findings are organized across the prevention spectrum, from universal health promotion to suicide crisis preparedness, active suicide postvention, and ongoing grief and bereavement interventions. Studies were organized into three main categories: empirical, theoretical, and synthesizing.

Table 1 summarizes the articles published from 1987 to 2019 in each category, showing them in chronological order. Empirical literature is based on observation, narratives, or experience rather than theory or pure logic, as are theoretical studies. Empirical studies collect and analyze data based on observable and measured phenomena. The categories and subcategories of the empirical and theoretical studies are listed in Table 1. The appendix provides additional information on the empirical studies: type of design, population, number of participants, strength of evidence and limitations, and key findings.

Table 1.

Description of categories and content of articles in the review

Subject of research	Subcategories and no. of articles
	Total empirical references: N = 33	Total theoretical references: n = 40
Grief and suicide bereavement	Identifying common experiences and mental health outcomes in the suicide bereaved (n = 9); Azorina et al. (2019); Bailley et al. (1999); Begley & Quayle (2007); Bridge et al. (2003); Dutra et al. (2018); Dyregrov et al. (2014); Gould et al. (2018); Hoffman et al. (2010); Levi-Belz & Lev-Ari (2019) Evaluating effectiveness of interventions for grief (n = 3): Black & Urbanowicz (1987); Pfeffer et al. (2002); Shear et al. (2005)	Synthesizing what we know about grief and bereavement (n = 5): Boelen (2016); Cerel et al. (2008); Farberow et al. (1992); Moore & Freeman (1995); Young et al. (2012)
Postvention	Evaluating postvention interventions (n = 4): Callahan (1996); Carter & Brooks (1990); Hazell & Lewin (1993); Poijula, Dyregrov, et al. (2001) Postvention evaluated as part of larger prevention model (n = 4): Capps et al. (2019); Hazell (1991); Mirick et al. (2018); Sandor et al. (1994) Case studies on school response to suicide postvention (n = 2): Berman (1990); Roberts (1995) Archival research to observe characteristics between group participants in active postvention versus traditional postvention (n = 1): Cerel & Campbell (2008)	Providing rationale and advocating for postvention as suicide prevention (n = 11): Aguirre & Slater (2010); Andriessen (2009); Andriessen & Krysinska (2011); Dunne-Maxim et al. (1992); Erlich (2016); King (1998); Leenaars & Wenckstern (1999); Mauk (1994); Mauk & Rodgers (1994); Speaker & Peterson (2000); Wenckstern & Leenaars (1993) School postvention guidelines: Specific recommendations for planning and implementation in schools (n = 14): American Foundation for Suicide Prevention & Suicide Prevention Resource Center (2018); Celotta (1995); Cox et al. (2016); Fineran (2012); King (1999, 2001); Komar (1994); Leenaars & Wenckstern (1998); Maples et al. (2005); Peterson et al. (1993); Sandoval & Brock (1998); Siehl (1990); Talbott & Bartlett (2012); Tierney et al. (1990) Proposed school-based suicide prevention and postvention recommendations specifically for Native American youth (n = 1): Metha & Webb (1996) Proposed specific interventions for survivors of suicide (n = 2): Galway et al. (2019); Parrish & Tunkle (2005) Postvention for mental health professionals after a client death (n = 1): Veilleux & Bilsky (2016)
Suicide contagion, clustering, and risk	Cross-sectional studies indicating suicide contagion or clustering (n = 1): Johansson et al. (2006) Case studies on suicide contagion (n = 1): Bell et al. (2015) Natural research design indicating suicide contagion (n = 1): Poijula, Wahlberg, et al. (2001)	Synthesizing what is known about suicide contagion, imitative suicide-related behaviors, and clusters (n = 3): Insel & Gould (2008); Poland et al. (2019); Zenere (2009)
Suicide preparedness: Mental health workers	Evaluating training to affect practices, perceptions, knowledge about prevention/postvention (n = 7): Breux & Boccio (2019); Erlich et al. (2017); Grossman et al. (1995); King et al. (2000); Mackesy-Amiti et al. (1996); Müller et al. (2017); O’Neill et al. (2020)	Describing suicide prevention/postvention training for mental health workers (n = 3): Cazares et al. (2015); Scott (2015); Suldo et al. (2010)
Description of synthesizing review articles	Total synthesizing references: n = 12
Postvention as part of larger suicide prevention or youth mental health programming (n = 7)	Burns & Patton (2000); *Cox et al. (2012); *Robinson et al. (2013); Singer et al. (2019); *Szumilas & Kutcher (2011); Totaro et al. (2016); Weare & Nind (2011)
Suicide bereavement (n = 5)	*Andriessen, Krysinska, Hill, et al. (2019); *Currier et al. (2007); Jordan & McMenamy (2004); *Maple et al. (2014); *McDaid et al. (2008)

^*Indicates systematic review.

Universal Prevention: Protective Environments and Safe Coping Strategies for Students

Creating an accepting environment in which children and youth feel safe and connected to others, including adults, can reduce the risk of suicide. Protective factors include students’ feelings of connectedness in their school environment, supportive relationships with adults and the community, and adaptive coping skills. School connectedness serves as a buffer to suicide risk (Marraccini & Brier, 2017; Whitlock et al., 2014), and fostering positive relationships between students and social workers, teachers, and other adults in the school is important so that students have safe trusted adults with whom to speak during postvention crises (King, 2001; Poland et al., 2019). It is important to note that, although skills training for students presents as beneficial to suicide prevention initiatives (Gould, Jamieson, et al., 2003), “there is little evidence available which supports protective factors being frequent among those adolescents most at risk for suicide” (DeHart, 2019, p. 12). The AFSP and SPRC (2018) toolkit emphasizes the importance of adults encouraging students to utilize healthy coping skills, facilitating discussions around emotions and practical coping strategies such as exercising, identifying dependable adults, and using relaxation or distraction skills. These recommendations can be described as social and emotional learning.

Schools’ proactive implementation of social-emotional learning generally promotes student well-being and can support future suicide postvention activities (Weare & Nind, 2011). Although coping strategies may vary between individuals, this review did not identify specific coping strategies that are helpful when initiating school-based suicide postvention. Students, especially in younger grades, cannot be expected to know how to utilize such coping strategies without training, modeling, and practicing on a regular basis to strengthen and hone these skills. One such intervention with promising initial results is the Good Behavior Game (GBG; Wilcox et al., 2008). Aimed at socializing first and second grade students to “the student role,” the GBG offers a non-stigmatizing game to reduce aggressive disruptive behavior. In a longitudinal clinical trial, Wilcox and colleagues (2008) found that participating in the GBG and learning prosocial skills in early elementary school appeared to reduce suicidal ideation and suicide attempts in early adulthood.

The Signs of Suicide (SOS) is another promising program that contributes to help-seeking within the school environment, particularly for those students who received the intervention (Singer et al., 2019). The program evaluation demonstrated that providing student education on signs of suicide and mental health issues helped student peer leaders believe that there were adults ready to help and decreased stigma in addressing stress and mental health concerns, but this is not enough. Other literature indicates that programs that place greater emphasis on mental health awareness education, personal growth, and positive youth development are likely to have a greater impact on depressed mood and suicidal ideation, although further research is needed to verify any enduring effects (Aseltine & DeMartino, 2004). Overall, schools with cultures that foster trust between youth and school staff (Wyman et al., 2019) and encourage mental health help-seeking (Pisani et al., 2012) have lower rates of suicidality, whereas schools that stigmatize mental health help-seeking (Mueller & Abrutyn, 2016) or where popular youth report suicidality (Wyman et al., 2019) have higher rates. These findings have opened up new suicide prevention programming strategies that are particularly efficacious when youth leadership is emphasized (Wasserman et al., 2015; Wyman et al., 2010).

Suicide Crisis Preparedness

Building relationships with parents, guardians, and community stakeholders.

Prior to any crisis planning or suicide postvention, building a foundation of collaborative efforts and cohesion among parents, guardians, schools, and community stakeholders is a critical component in suicide crisis preparedness, as well as active suicide postvention response (Cerel & Campbell, 2008; Cox et al., 2016; King, 2001). Policies must be established and agreed upon at the district and individual school level; for example, will district-level crisis teams and/or school crisis teams lead the intervention? Additionally, it is imperative to identify members of the crisis team and establish clear roles for members. Crisis team members should be broadly representative, extending beyond school counselors and social workers to include families, community members, and community mental health organizations. This diverse team is crucial to support students and their emotional well-being, especially because suicide has the potential to negatively impact the entire community (Erlich, 2016; King, 2001).

Developing relationships with community leaders and organizations can help ensure appropriate responses from various parts of students’ lives and streamline the process of disseminating information about resources to support individuals and families (Erlich, 2016; King, 2001). However, in the scope of this review, we found a lack of empirically based findings, recommendations, or studies on how to build relationships among these stakeholders to effectively support suicide prevention or postvention initiatives in the school. Although they flagged relationships with stakeholders as important, the studies reviewed here did not specify what kinds of relationships with which stakeholders matter for these teams.

Several articles strongly recommended that school-based mental health professionals, such as school social workers, psychologists, and counselors, are aware of outside community resources to which they can refer students and families in need of additional supports (Capps et al., 2019; Galway et al., 2019; Singer et al., 2019). Existing and established relationships between mental health organizations and school-based personnel would, in theory, make these referrals and a collaborative approach to suicide postvention easier for all involved. These calls for collaboration did not include specific action steps to build capacity or specify best-practice strategies to form and strengthen key relationships. Rather, the relationships among school staff, family, and community members were underscored as vital in theoretical articles or important to study in future research recommendations.

Training mental health professionals.

Although mental health practitioners are an integral part of crisis preparedness, the AFSP and SPRC (2018) toolkit for suicide postvention lacks guidelines on the training and supports necessary to assist students, families, and community members following a death by suicide. In an expansive literature review, Robinson and colleagues (2013) identified only twelve studies that evaluated gatekeeper training programs, focusing on identifying, referring, and responding to students at risk for suicide without particular attention to postvention. Although many of the studies lacked detailed descriptions of methodology and implementation of training, the main outcomes of these trainings focused on increased knowledge of youth suicide and confidence of mental health professionals in responding to suicidal students (Robinson et al., 2013). In one recent study, less than half of the surveyed North Carolina school psychologists indicated having formal postvention training (O’Neill et al., 2020). Breux and Boccio (2019) also found that staff members involved in suicide crisis response teams felt they needed more professional development and ongoing support in suicide postvention.

Despite the lack of strong evidence of efficacy through controlled design or randomized control trials (Robinson et al., 2013), positive effects in knowledge and confidence were noted when staff members engaged in gatekeeper training (Breux & Boccio, 2019). The literature suggests that this type of suicide prevention and crisis training is an important factor in school postvention efforts and planning because it can increase staff knowledge about how to assist youth in crisis, improve attitudes, and strengthen participants’ perceived self-efficacy.

Several studies showed that gatekeeper knowledge decreased over time from three-, six-, and nine-month follow-up (Robinson et al., 2013), indicating a need for regular and consistent training in any crisis, suicide prevention, and postvention programming (O’Neill et al., 2020; Suldo et al., 2010). Ongoing training can support school-based mental health professionals’ feelings of confidence in leading prevention and postvention efforts and can also mitigate any potential risks associated with inadequate suicide crisis interventions, including but not limited to suicide contagion (O’Neill, et al., 2020; Robinson et al., 2013). However, more research is necessary to determine which training modalities and curricula are most effective, particularly after a suicide occurs. Additional information is needed to develop standards of implementation to enable schools to streamline this process.

Seven of the thirty-three empirical studies focused on training, with an emphasis on knowledge, perceptions, or training experience so that social workers and other mental health professionals can better serve the school-based students following a death by suicide. However, these articles consisted of three observational studies using cross-sectional survey data (Erlich et al., 2017; King et al., 2000; O’Neill et al., 2020), three pretest-posttest designs (Breux & Boccio, 2019; Grossman et al., 1995; Mackesy-Amiti et al., 1996), and one study that relied on a semi-structured interview (Müller et al., 2017). Although pretest-posttest designs in general can be helpful in ascertaining participants’ increased knowledge, in this review the findings are limited by issues related to consistency and validity such as improvement in pretest-posttest scores based on exposure to repeated measures, potential self-selection bias, and a lack of comparison groups.

Formulating and preparing a crisis postvention plan.

This review found twenty-nine theoretical articles that advocated for schools to develop a postvention plan with specific recommendations for implementation and no empirical articles focused on postvention planning. The academic articles included a description of the critical components of the plan, such as having an established crisis plan before the event, mobilizing the crisis team, reaching out to the family to confirm death by suicide, media safe messaging, speaking to students in classrooms, and identifying students in need of support. Although all crisis plans should be formulated prior to an incident to support a rapid staff response and minimize the intensity and duration of negative effects (King, 2001; Owens, 2014), the empirical evidence supporting postvention planning and these specific recommendations is limited and vague.

Whereas evidence suggests that at least 90 percent of school districts in the United States have “some sort of crisis plan” (O’Neill, 2017, p. 26), most crisis plans are not comprehensive or issue-specific (many leave out postvention) nor are they communicated to and clearly understood by all school staff, including social workers, school counselors, and psychologists, with staff members varying by school and district. In DeHart’s (2019) literature review, numerous researchers emphasized the importance of a multifaceted approach to suicide prevention efforts, whereas “postvention activities were left unaddressed” (p. 55) even though they must be included to develop a comprehensive suicide prevention plan. This assertion is reflected even in the AFSP and SPRC (2018) toolkit, which recommends that schools develop and implement a comprehensive school-wide suicide prevention program including postvention before a crisis occurs.

How to implement postvention plans was not well articulated within the toolkit, especially for schools with scant resources, including those that may have insufficient mental health staff to meet high needs for mental health supports for risk assessments and individual follow-ups, for example. The complexity of developing such plans with attention to cultural and local specificity is not understood, although it is acknowledged. Owens (2014) emphasized a critical component: “the school crisis plan should both be based on a theoretical model and address the uniqueness of the school’s location and student population” (p. 16). How to navigate these complex issues was not described. Additionally, many empirical articles analyzed in this review presented with problems of diverse student populations, and the lack of specific research on different groups and contexts (rural, suburban, and urban schools) makes generalizability across schools tricky (see appendix for empirical article information).

The crisis response team.

In this review, we did not find empirical studies that focused solely on formulating a district-wide crisis response team or even individual school-based teams. Some theoretical studies defined suicide postvention and advocated for postvention and protocols as described above, but these reflected much of the same information provided by the AFSP and SPRC toolkit (2018) without delving into specific studies or other supporting evidence. According to this literature, there is a general consensus that school districts (and schools within them) should establish these teams and that the teams should include a diverse group of school-based professionals such as administration, school social workers, school counselors, school psychologists, nurses, and school resource officers (AFSP & SPRC, 2018). The criteria for involvement and the kinds of specific evidence-based frameworks these teams can use to navigate outreach, communication, and decision making were not described nor were the configurations and processes of these team studied.

Active Suicide Postvention

Safe messaging.

There are no empirical studies in this review that considered media safe messaging or the impacts of social media on suicide contagion even though teenagers are using social media ubiquitously and regularly. Theoretical articles encourage monitoring social media and limiting the spread of misinformation about a suicide or romanticizing suicide. It appears that many of the recommendations around safe messaging stem from one older empirical study that reflected how sensationalizing language can lead to a romantic view of suicide and lead to an increase in suicidal behavior or attempts (Callahan, 1996).

Scientific information that is disseminated widely appears to be derived mostly from three theoretical articles focusing on suicide contagion (Insel & Gould, 2008; Poland et al., 2019; Zenere, 2009). Poland and colleagues (2019) reported on the “well-established” notion that “at-risk individuals with a recent history of suicide attempt or a concurrent severe depression are more likely to attempt suicide in the wake of a media report of suicide” (p. 22). The rationale for safe messaging typically centers on reducing the risk of suicide contagion after a tragedy or media story (Gould, Jamieson, et al., 2003). Safe messaging to lessen the associated risk of suicide clusters and contagion includes omitting vivid details such as method of suicide, avoiding language that romanticizes or sensationalizes the suicide, referring to mental health resources, promoting help-seeking and coping with negative emotions, and using only neutral photos if any (Owens, 2014; Poland et al., 2019). There seems to be theoretical agreement about the importance of providing accurate information about suicide and emphasizing that suicide is not caused by a single event (AFSP & SPRC, 2018, p. 55). Additionally, a consistent recommendation involves schools closely coordinating with the family of the deceased, school personnel, and the larger community in working with the media to ensure safe messaging protocols that will lessen the likelihood of suicide contagion.

Indeed, recommendations can and should be utilized to support safe messaging for students and parents immediately following a suicide in their school, with scripts to ensure accuracy and uniformity of information (Owens, 2014). The postvention toolkit (AFSP & SPRC, 2018) offers some print templates and “language to avoid” for safe messaging, including specific verbiage and scripts that staff and administrators are recommended to use when discussing a suicide among students and parents. These recommendations include using language such as “died by suicide” or “killed himself or herself” rather than a “successful” suicide. However, some key guidance is missing. For example, not romanticizing suicide is a clear recommendation (AFSP & SPRC, 2018) based on theoretical work (Insel & Gould, 2008; Poland et al., 2019; Zenere, 2009), but even the user-friendly toolkit does not include specific language to avoid.

Minimizing suicide contagion.

Only three of the thirty-three empirical studies focused on suicide contagion and clustering, and all were descriptive epidemiologic studies, not intervention research. Of these three, only two focused on adolescent suicide (Johansson et al., 2006; Poijula, Wahlberg, et al., 2001), and the remaining one focused on higher education students with unknown demographic data. Both articles focusing on adolescent suicide supported the significance of identifying close friends to the deceased and identifying individuals more at risk for suicide.

More studies focused on reducing risk of suicide contagion within schools are needed. Our review found only two school-specific postvention studies with four publications describing them (Poijula, Dyregrov, et al., 2001; Poijula, Wahlberg, et al., 2001; Hazell, 1991; Hazell & Lewin, 1993). One study focused on adolescent suicide (Johansson et al., 2006) and described student suicide clusters in Sweden, establishing that suicide contagion occurred. However, the authors did not report on what else may have contributed to the contagion besides the fact that the deceased individuals knew one another. In the second youth-focused study, Poijula, Wahlberg, and Jokelainen (2001) devised a natural research design and noted suicide contagion in some high schools where there were no strategic, large-scale, school-wide postvention interventions. Their findings indicated that schools that initiated “talk-throughs” and psychological debriefings after a student suicide had no new suicides, further highlighting the importance of some form of school-based postvention.

Poijula, Dyregrov, and colleagues (2001) conducted research in three rural high schools in northern Finland after a suicide cluster involving all three schools. Using a natural research design, these researchers considered whether and how quickly schools initiated crisis intervention. They used additional suicides and the student body expressions of posttraumatic stress disorder (PTSD) and grief reaction on surveys as their outcomes (n = 89). Results indicated that timely and more thorough crisis intervention (in the form of first talk-throughs and psychological debriefing) may reduce the risk of contagion. Due to the small numbers and the low base rate of youth suicide, these authors characterized their findings as tentative.

In the second school-based postvention intervention study, Hazell and Lewin (1993) studied two schools in New South Wales, Australia, that had experienced a recent student suicide death. The first publication described the suicide events and the postvention 90-minute counseling sessions involving twenty-five students who were close peers of the deceased (Hazell, 1991). The second publication surveyed students in eighth through eleventh grade in the two schools eight months after the suicides and compared two groups of matched case and control students: those who were close to the deceased and either received counseling or did not receive counseling. The ninety-minute counseling sessions in this study were initiated by school staff, and the outcomes did not find differences in the suicide risk index between those who received counseling and those who did not. Again, these authors cautioned readers that their unconvincing results should not “be taken to be representative of other postvention counselling programs” (p. 108) due to the limited scope of the piloted counseling intervention. None of these school-based postvention research projects (Poijula, Dyregrov, et al., 2001; Poijula, Wahlberg et al., 2001; Hazell, 1991; Hazel & Lewin, 1993) reported on the racial/ethnic socioeconomic makeup of the student body or community characteristics (rural, urban, or suburban) of the schools, although these particulars could be important for generalizing and applying the findings.

Identifying individuals in need of additional supports.

Our review found common themes among both empirical and theoretical articles on postvention focused on identifying at-risk young people after a peer suicide. There are two postvention models described in this literature: traditional and active (Cerel & Campbell, 2008; Neimeyer et al., 2017). Traditional postvention encompasses activities or services suicide survivors seek out on their own. In contrast, the active model of postvention is initiated immediately after a suicide and involves seeking out and guiding survivors by offering supportive services and resources rather than relying on survivors to seek help on their own (Erlich, 2016), a process that may be thwarted due to perceived stigma (Galway et al., 2019). Delaying mental health services following a death by suicide not only increases the likelihood of suicide contagion but it also can affect the overall well-being of individuals and increase complicated grief reactions and problems in social connectedness (Azorina et al., 2019; Neimeyer et al., 2017). Despite these deleterious effects and barriers to help-seeking frequently faced by survivors (Aguirre & Slater, 2010; Andriessen, Krysinska, Kõlves, et al., 2019; Galway et al., 2019), the traditional postvention model is most commonly utilized, although it is estimated that only 25 percent of survivors actively seek support on their own (Aguirre & Slater, 2010).

Active postvention means that school personnel immediately work to identify individual students who are at risk after a suicide, which include close friends and other peers who were not necessarily close to the deceased (Poland et al., 2019). The SPRC (n.d.) recommends using screening tools and mental health professionals to identify at-risk students during postvention following a student death. Cha and colleagues (2018) conducted a school-based postvention intervention in Korea within one week of a high school student’s death by suicide. Researchers engaged in active postvention initiatives by utilizing screening tests such as those measuring PTSD, anxiety, and depressive symptoms; educational sessions with students, parents, and school staff to normalize grief reactions and discuss adaptive coping strategies; and additional follow-up interviews and psychiatric supports for students with indicated PTSD symptoms. Active postvention was statistically significant in improving PTSD, anxiety, and depressive and complicated grief symptoms of students within the first five-month follow-up. In general, active postvention efforts may normalize the grief reactions and stigma of the trauma of death by suicide and resolve any acute issues or psychological distress (i.e., depression, anxiety, and complicated grief) before these disturbances result in chronic health problems, maladaptive coping strategies, or other significant mental health disorders (Cerel & Campbell, 2008; Cha et al., 2018; Levi-Belz & Lev-Ari, 2019).

Although screening for suicide and related risk factors may be important for identifying and referring students to additional prevention and postvention supports, current school-based best practices for screening have not yet been established and may subsequently miss identifying some students in need (Singer et al., 2019). According to Hazell and Lewin’s 1993 intervention study, history of suicidal ideation and behaviors were the strongest predictors of current or recent ideation and behavior, whereas friendships with an individual with a suicide attempt or death added a significant but small contribution to the overall risk of contagion. Unless staff are aware that screening should consider both history of suicidal ideation and behavior as well as relationship closeness with the decedent, they may miss identifying those in greatest need.

To identify individuals in need of additional mental health supports, aspects of the SOS school-based program may provide helpful insights. Signs of Suicide includes awareness education and screenings, and intervention studies indicate some positive impacts on self-reported suicide behavior and attempts (Aseltine & DeMartino, 2004; Singer et al., 2019). The SOS suicide prevention program includes a brief screening tool to screen students for depression; afterward they can indicate whether they want to talk with a trusted adult in the school building. Although more research and testing must be done to develop a standardized tool for identifying individuals in need of additional supports, utilizing a tool such as this, in combination with others, may encourage some individuals to seek help or allow staff to actively identify youth at risk of suicide.

It is important to add a note of caution here. Although screening programs can assist in identifying individuals at risk who would otherwise avoid help-seeking behaviors, they can be problematic in two main ways. First, there is a wide range (4–45%) of screened students identified as requiring further support in reported studies (Robinson et al., 2013). This range is likely a product of school climate, student gender (girls are more likely to screen positive), and cultural and social meanings associated with the screening tool itself (e.g., mental health stigma). Because of this, screening tools may not detect at-risk students. The second issue with screening is the need for accessible and culturally relevant mental health services for those who screen positive and ensuring that enough mental health staff are available to meet with students who screen positive or at high risk during assessments. Best practices should include age and culturally appropriate screening tools, as well as trained mental health professionals who are aware of the signs of suicide risk; know the student body; are aware of relationships impacted by any student death; and have the appropriate infrastructure, including trained mental health staff available to support services for those who are identified. To quickly pivot to actively screening, supporting, and educating students and their families requires preparations, protocols, and community-school relationships.

Additionally, social workers and other school staff should focus postvention efforts on identifying and supporting family members and friends of the deceased student to offer immediate mental health supports (Owens, 2014). The need for mental health supports for the bereaved is clear; people bereaved by suicide have increased risks of suicide or suicide attempts, although they often report receiving less support than others during their bereavement (Azorina et al., 2019). Additionally, survivors of suicide are at a higher risk of developing complicated grief (Levi-Belz & Lev-Ari, 2019). Postvention needs to be immediate to mitigate risks of suicide contagion and to help individuals cope with grief and trauma. However, efficient and targeted interventions should be offered for several months because PTSD symptoms can last up to six months after a suicide cluster in a school setting (Poland et al., 2019).

Grief and bereavement supports.

Research on suicide bereavement emphasizes the importance of normalizing grief reactions to lessen the effects of complicated grief and stigmatization (AFSP & SPRC, 2018); however, the state of the science leaves many unanswered questions. Of twelve empirical studies focused on grief, bereavement, or intervention, only seven focused specifically on suicide bereavement (Azorina et al., 2019; Begley & Quayle, 2007; Bridge et al., 2003; Dutra et al., 2018; Gould et al., 2018; Hoffman et al., 2010; Levi-Belz & Lev-Ari, 2019). Many described the experience of loss through qualitative thematic interviews following traumatic death (Dutra et al., 2018; Dyregrov et al., 2014; Hoffman et al., 2010), case control studies of students who lost a friend or classmate to suicide (Gould et al., 2018), and phenomenological research about the experience of losing an immediate family member to suicide (Begley & Quayle, 2007). This literature base also includes cross-sectional survey research with individuals who were bereaved by suicide or other traumatic deaths (Azorina et al., 2019; Bailley et al., 1999) and a cohort study of individuals in Israel bereaved by suicide and their complicated grief reactions and self-disclosure over time (Levi-Belz & Lev-Ari, 2019).

The review found three grief and bereavement intervention studies. Of these studies, two focused on interventions with bereaved children (Black & Urbanowicz, 1987; Pfeffer et al., 2002), and one focused on an intervention for children bereaved by suicide (Pfeffer et al., 2002). The remaining article evaluated a program offering grief supports for adults, who could then support bereaved children (Shear et al., 2005). All of these studies support the recommendation to initiate immediate grief supports to lessen anxiety and depressive symptoms in suicide survivors. Additionally, there is modest research indicating that grief-specific therapeutic intervention may be more beneficial than general psychotherapy (Black & Urbanowicz, 1987).

Suicide has a large impact. Research estimates that between 115 and 135 people are exposed to each suicide (Cerel et al., 2018; Survivors of Suicide Loss Task Force, 2015), many of whom report this experience as having a detrimental impact or causing major disruptions in their lives. Approximately 50 percent of the U.S. population has experienced a suicide loss (Feigelman et al., 2018). A study indicates that suicide survivors who do not self-disclose experience higher rates of complicated grief and suicidality than those who talk about the experience and thus have less grief difficulties and mental health impairments (Levi-Belz & Lev-Ari, 2019). Although postvention supports are critical and often called for in the literature, “only a single cross-sectional study has directly addressed the specific role of S-D [self-disclosure] in CG [complicated grief] among SUSs [suicide survivors]” (Levi-Belz & Lev-Ari, 2019, p. 2). No studies have examined the role of self-disclosure over time as a protective factor against depression, suicidal ideation, and complicated grief in survivors of suicide (Levi-Belz & Lev-Ari, 2019).

Azorina and colleagues (2019) published an empirical study exploring the perceived impact of suicide bereavement on social relationships. They found social withdrawal of the bereaved (43%); attachments affected by fear of loss, which resulted in either overprotection or withdrawal (36%); social discomfort of friends and family over death (24%); and shared bereavement experiences resulting in closeness or avoidance (21%). Relationships often became strained following the suicide, and many responders felt uncomfortable with how others perceived their grief reactions, either feeling pressure to get over the death quickly or feeling that they did not show enough emotion (Azorina et al., 2019). This research highlighted a need for social supports for suicide survivors, especially peer supports with shared experiences during their grieving.

Overall, there are few controlled evaluation studies for suicide survivor bereavement, and “little is known about what treatments, programs and group formats are beneficial for what survivors regarding age, gender, kinship, in what time after the suicide” (Andriessen, 2009, p. 45). In general, social workers, administrators, and other school staff need to lessen the amount of time that passes between a suicide death and receipt of support services by student survivors. Postvention is important because it provides an opportunity to safely express grief and psychological pain and to increase one’s sense of belonging and connectedness. Without support groups, some survivors will avoid help-seeking behaviors (Aguirre & Slater, 2010; Galway et al., 2019; Levi-Belz & Lev-Ari, 2019).

Many of the long-term bereavement needs of student suicide survivors are potentially beyond the scope of school personnel’s capacities and/or responsibilities. However, research suggests that schools could help by providing bereavement supports for student suicide survivors, immediately and with a larger range of supports offered; more long-term therapeutic aids; and consistent follow-up by staff to ensure engagement (Owens, 2014; Poland et al., 2019). One such broader approach was described by Galway and colleagues (2019), who explored the possibility of referring bereaved peers to non-medical community-based supports, in conjunction with existing suicide postvention supports to address and reduce stigma. There is some evidence suggesting that it can “improve connectedness, increase sense of empowerment, and reduce health service use” (p. 2), as well as increasing help-seeking behaviors, improving agency integration, and providing access to a multitude of suicide bereavement supports. This intervention is yet to be tested for efficacy.

It is important to note that most suicide bereavement studies found in this review focus on White females in urban and suburban communities. Of the seven studies on those bereaved by suicide, only three involved children or adolescents (Gould et al., 2018; Hoffman et al., 2010; Pfeffer et al., 2002). Two of these studies included primarily White students (Gould et al., 2018; Pfeffer et al., 2002), and one did not specify the ethnicities of the all-female participants (Hoffman et al., 2010). Of the remaining four studies on adults bereaved by suicide, three had a majority of White female participants from areas such as the United Kingdom, Ireland, or Israel (Azorina et al., 2019; Begley & Quayle, 2007; Levi-Belz & Lev-Ari, 2019). The remaining study did not report gender or ethnicity of participants (Dutra et al., 2018). Because suicide and death practices are shaped by gender, culture, and context, the bereavement studies in this review indicate a lack of knowledge reflecting the experiences of rural males and people from diverse populations.

Limitations

This review of the literature supporting current recommendations is limited by our reliance on peer-reviewed journal articles and our focus on postvention and search terms that reflect that narrow focus. More is known about suicide prevention in schools than is indicated by our review due to our focus on postvention or the period of time after a suicide occurs. Excluding books or other types of literature narrowed our review and may have inadvertently excluded some work that is particularly important to schools (i.e., Erbacher et al., 2015). Our review is focused on identifying the state of evidence for postvention recommendations and strategies, which are critical to move the field of suicide prevention forward.

Although postvention is often motivated by a desire to prevent suicide contagion or a suicide cluster, it is beyond the scope of this review to assess the state of knowledge about suicide contagion or suicide clusters. Nevertheless, it is crucial to understand the mechanisms that drive events such as contagion and clustering if we are to design effective postvention strategies that thwart them (Abrutyn et al., 2019; Mueller & Abrutyn, 2016). Understanding what contributes to suicide clustering and how it works are important directions for future postvention research. This article focuses on the school context; however, there may be important social and environmental influences beyond schools that affect postvention and students’ suicide risk.

Discussion

Most suicide prevention programs focus on the outcomes of suicidal ideation and behaviors, with the main goal being to observe a decrease in these phenomena post-intervention. However, because death by suicide is generally rare, postvention interventions are hard to plan and carry out in real time. Additionally, suicidal ideation is relatively common, with almost 20 percent of high schoolers reporting seriously considering suicide during the past twelve months (Centers for Disease Control and Prevention, 2020). These complexities in assessment plus the reliance on self-report survey data to measure suicidal risk and ideation create difficulties in consistency and accuracy (Singer et al., 2019), making it difficult to accurately assess suicide-related outcomes (Robinson et al., 2013). Postvention research is also challenging due to ethical concerns that prevent randomized controlled trials between suicide prevention and postvention intervention groups and control groups. Such issues combine to create gaps in our knowledge base, with little known about effective postvention services, supports, and components (Andriessen, Krysinska, Kõlves, et al., 2019).

Comprehensive suicide prevention includes work across the prevention spectrum from universal to postvention. Postvention is an understudied area. This review of the school suicide postvention and suicide bereavement literature indicates multiple knowledge gaps, particularly in intervention research for suicide postvention in the school setting. There is little empirical research to identify feasible best practices for schools to implement after a suicide death to reduce the risk of suicidality in youth. This review also found areas in which the limited research evidence is amplified beyond its original scope. Findings and suggestions for future research efforts can be summarized as follows:

• Building relationships with parents, guardians, families, and community stakeholders and developing a crisis postvention plan, including an identified crisis response team with representatives from the school, community, and families, is key to postvention. The complexity of developing such plans with attention to cultural and local specificity is not understood, although it is acknowledged. Building trusting relationships over time can be key to the success of such an effort (Goebert et al., 2018).

• Efforts must be made to incorporate effective evidence-based postvention training for school staff, including undergraduate and graduate level preparation, district- and school-level training, and continuing education. Social workers and other mental health practitioners require more postvention training earlier in their careers, especially because they are considered first responders. Many of the articles in this review outlined recommendations for comprehensive programming, including gatekeeper training. Importantly, postvention protocols must be taught and made regularly available to all school staff. Without receiving adequate training and support around postvention and ways to mitigate suicide contagion, ill-equipped mental health professionals may inadvertently increase the likelihood of suicide contagion (Poijula, Wahlberg, et al., 2001; Poland et al., 2019; Zenere, 2009).

• Active suicide postvention includes safe messaging to reduce the risk of suicide contagion. Although more studies focused on reducing the risk of suicide contagion in schools are needed, specific recommendations on language to utilize, scripts, and templates are available in the AFSP and SPRC toolkit (2018) and should be considered by school personnel. Safe messaging includes refraining from use of romanticized language (Poland et al., 2019); however, more guidance around language to avoid should be provided and explored by researchers.

• After a student dies by suicide, it is important to support all students and identify vulnerable students within a school (Gould, Greenberg, et al., 2003; Robinson et al., 2013). Our current practice of utilizing self-report survey data runs the risk of missing students in need of support and at risk of suicide themselves, especially in view of the well-documented feelings of shame and stigma that are often attached to suicide. At-risk students may not accurately report their symptoms and feelings due to social desirability bias; subsequently they may miss follow-up support from school-based mental health providers and first responders. Reaching out proactively and quickly to offer mental health supports to suicide survivors while also encouraging students to self-refer is an important part of suicide postvention, but how to do this in culturally responsive and respectful ways is an important area for future research.

• Social workers and other mental health clinicians must utilize empirically supported grief interventions that extend beyond the scope of the school and delve into longer term supports for all age groups. Although research is limited (Pfeffer et al., 2002), support groups for survivors show promise in this review. However, research is needed on longer term supports both within and outside of schools. These activities and responsibilities cannot solely fall on schools given the significant needs of students, lack of funding, and broad scope of the training and responsibilities of social workers and other school staff. Additionally, most existing studies on suicide prevention and postvention revolve around middle and high school students. Postvention should also be developed for younger children in order to decrease suicide risk when a peer suicide (even suicide of an older peer) occurs, especially as younger students are an age group that is at growing risk of suicidal behavior (Ayer et al., 2020; Hedegaard et al., 2018). Building bridges between research on and clinical support of child and adolescent bereavement more generally (such as those by Judi’s House/JAG Institute [2021] or the Dougy Center [2021]) and research on school and community-based postvention may be an effective way forward.

• Postvention interventions recommendations amplify limited research findings. Robinson and colleagues (2013) observed only two postvention studies with four publications describing them (Poijula, Dyregrov, et al., 2001; Poijula, Wahlberg, et al., 2001; Hazell, 1991; Hazell & Lewin, 1993). These postvention studies have small sample sizes and inconclusive findings, and they reflect particular populations. Importantly, these studies are cited extensively in the suicide postvention literature although the authors cautioned against this. These two research projects were conducted two decades ago and do not offer strong or conclusive evidence of best practices. Further research into the kinds of counseling most beneficial for school-based postvention is another important focus of future study, since the benefits of 90-minute counseling sessions after a student suicide death were inconclusive (Hazell & Lewin, 1993).

In this review, we have described and considered the evidence that supports our national recommendations for school postvention from the AFSP and SPRC toolkit, identifying its strengths and limitations. The purpose is to identify areas of amplification of evidence, gaps in knowledge, and specific recommendations informed by research. This scoping review has exposed gaps in our knowledge base and complexities that require more and broader research to better understand and address postvention in schools. We must create more empirically supported, culturally responsive, immediate, active postvention interventions for various representative populations. Currently, no comprehensive studies exist that explore and focus on suicide postvention for adolescents and children in the school setting, and there is a dearth of suicide postvention strategies, policies, and procedures that have demonstrated efficacy (Andriessen, Krysinska, Kõlves, et al., 2019). Given methodological challenges, more emphasis on accumulating observational empirical case studies may be an important direction for future research. This review aims to help schools decipher and apply what we know from scientific postvention research and to spur targeted research to contribute to the actionable scientific knowledge base. The goal is not only to help develop evidence-based protocols for school crisis plans but also to highlight the need for more research areas within school-based suicide postvention.

Appendix. Empirical Article Information

Authors	Study design	Population (age, gender, ethnicity)	No. of participants	Strength of evidence/limitations	Key findings
Azorina et al. (2019)	National cross-sectional design; survey data	Young adults bereaved by suicide in the United Kingdom Age range 18–40 years (mean = 45 years) Male (N = 83) Female (N = 416 White (N = 461) Other ethnicities unknown	N = 499	Large sample size contributes to more generalizability; potential selection bias and social desirability bias; potential recall bias	Main themes of adults bereaved by suicide included social discomfort of friends and family over death, social withdrawal, shared bereavement experiences resulting in closeness or avoidance, and attachments affected by fear of loss that resulted in either overprotection or withdrawal.
Bailley et al. (1999)	Survey research	Bereaved university students from an introductory psychology course Age range 18–64 years Male (N = 90) Female (N = 253) Unknown gender (N = 7) White (N = 307) Other ethnicities unknown	Total bereaved participants (N = 350) Cause of death: Suicide (N = 34) Accident (N = 57) Unanticipated natural (N = 102) Anticipated natural (N = 157)	Lower quality of evidence due to this being an observational study; difficulty generalizing due to homogeneity across individuals and small sample size of bereaved suicide survivors; potential self-selection bias	Compared to other bereaved individuals, suicide survivors experienced more grief reactions and feelings of rejection and responsibility and increased levels of shame and perceived stigma.
Begley & Quayle (2007)	Phenomenological research	Adults living in Ireland bereaved by suicide and recruited from the Living Links network for adult suicide survivors Age range 27–72 years Male (N = 3) Female (N = 5) Ethnicities unknown	N = 8	Difficulty generalizing due to homogeneity across individuals and small sample size; potential self-selection bias	Themes of bereaved include early attempts to “control the impact” of the death, the need to “make sense” of the death, feeling uncomfortable in social situations, and a sense of “purposefulness” in their lives following the suicide death. Findings suggest bereavement is shaped by the individual’s experiences with the deceased and their perceived social interactions after the death.
Bell et al. (2015)	Case studies and interviews	Individuals in the United Kingdom, connected to students attending higher education institutions who died by suicide Ages, genders, and ethnicities unknown	Total N = 73 Family members N = 29 Students’ friends N = 12 Higher education institution staff N = 17 Coroners and procurators of fisca records N = 15	Lacking scientific rigor; potential research bias; self-referral bias	The idea of suicide gets transmitted through the “power of ideas” (how the unthinkable can become thinkable).
Berman (1990)	Case study on school response to suicide postvention	Two individual school-aged youth (12 and 13 years old) who died by suicide Male (N = 2) Southeast Asian (N = 1) White (N = 1)	N = 2	Small sample size makes the findings difficult to generalize. Sampling protocols, specifics of school response, and interventions are unknown.	Schools must be prepared with suicide crisis protocol to respond appropriately with postvention. Schools should also foster relationships in the community and consider cultural implications with suicide and diverse populations.
Black & Urbanowicz (1987)	Randomized controlled trial; intervention study	Children < 16 years old who had at least one parent die Male (N = 42) Female (N = 41) Ethnicity unknown	N = 83	Higher quality of evidence due to randomized trials; potential attrition bias	There was some indication that the treatment group benefitted from brief intervention (“very modest” findings).
Breux & Boccio (2019)	One-group pretest-posttest design	School personnel attending the Creating Suicide Safety in Schools (CSSS) workshop Mean age = 41.77 years Male (N ≈ 78) Female (N ≈ 484) Ethnicities unknown	N = 562	Potential attrition bias following the 3-month follow-up; potential issues with internal consistency and validity; practice effects in the pretests and posttests; lack of comparison groups	After the CSSS workshop, there were significant gains in the knowledge area, especially for those who lacked previous training in suicide. School-based personnel identified barriers to implementation such as insufficient time and resources; lack of support from staff, parents, and administration to address the importance of suicide prevention; need for more professional development and ongoing support; and clearer suicide prevention, intervention, and postvention protocols.
Bridge et al. (2003)	Prospective design; cohort study	Adolescents exposed to a friend’s suicide Age range unknown, average age 18.3 years Gender and ethnicities unknown, but predominantly White population	Total N = 274 Exposed to suicide N = 129 Control group N = 145	Lower quality of evidence due to this being an observational study; possible recall bias; potential sampling bias due to snowball sampling technique; strong inter-rater reliability; difficulty generalizing due to homogeneity of participants	Adolescents exposed to a suicide had an increased risk of developing new-onset major depressive disorder (MDD) within 1 month after exposure. There were no differences in the incidence of MDD between the groups in months two to six. Prior history of alcohol use increased the risk of exposure to suicide and increased the risk of new-onset MDD. Youths with a family history of MDD and feeling responsible for the death were at a higher risk of MDD.
Callahan (1996)	Case study	Postvention in a middle school following the suicides of two students Postvention participants were adolescents, ages unknown Gender and ethnicities unknown	N = unknown	Lacking scientific rigor; difficulty generalizing results to the wider population; unknown sample size and demographics; potential researcher bias; inability to replicate	This is a middle school postvention that resulted in adverse effects (suicide contagion), leading to improvements in postvention methods. Recommended changes included having school mental health professionals meet with students in need for pre-arranged shorter time periods, requiring staff to follow up and inform parents of any reported suicidal ideation, establishing more open communication between teachers and mental health staff, and providing individualized instead of whole-group grief processing services in the school.
Capps et al. (2019)	Case study	High school students requiring Prevention of Escalating Adolescent Crisis Events (PEACE) protocol crisis intervention from a high school in rural western North Carolina Age range 13–18 years Male (N = 34) Female (N = 43) Transgender (N = 1) White (N = 68) Non-White (N = 10)	Total number of crisis events N = 78 Total number of students involved N = 58	Lacking scientific rigor; difficulty generalizing results to the wider population; potential research bias; self-referral bias	At this time, researchers cannot make conclusions about the effectiveness of the expansion of the PEACE protocol. However, results support the idea that crises can be managed without hospitalizing children, as well as the importance of safety planning and means restriction to reduce risk of suicide.
Carter & Brooks (1990)	Case study on postvention intervention	High school students, aged 14–18 from a magnet school who were bereaved after a friend completed suicide Male (N = 2) Female (N = 4) Ethnicities unknown	N = 6	Potential recall bias; lacking scientific rigor; difficulty generalizing results to the wider population; small sample size.	One and a half years after the death of their friend by suicide, all six postvention participants were alive and high functioning. They continue to report positive outcomes in the long-term during informal checks. Five of the six participants indicated that they felt the postvention group was helpful to them.
Cerel & Campbell (2008)	Archival research	Suicide survivors presenting for treatment at the Baton Rouge Crisis Center Age range 18–89 years Active model of postvention (APM) Male (N ≈ 46) Female (N ≈ 104) White (N = 147) African American (N = 3) Traditional passive postvention (PP) Male (N ≈ 55) Female (N ≈ 151); White (N ≈ 201) African American (N ≈ 5	Total N = 356 APM N = 150 Traditional PP N = 206	Potential self-selection bias; potential confounding variables; homogeneity between participants makes it difficult to generalize	Those who presented for the APM were treated sooner (48 versus 97 days) and were more likely to have been a survivor of a violent suicide (gunshot wound or hanging). They were more likely to attend survivor support groups and attend a higher number of groups.
Dutra et al. (2018)	Interview	Health professionals and family bereaved by suicide Health professionals Age range 28–58 years Genders and ethnicities unknown Family members Age range 24–51 years Male (N = 1) Female (N = 6) Ethnicities unknown	Total N = 20 Health professionals N = 13 Family members N = 7	Lack of scientific rigor; small sample size makes generalizing information difficult; potential responder bias	Themes of the suicide bereaved include being in a “state of shock,” living with the suffering and effects of the loss of the family member, and rebuilding their life in the aftermath.
Dyregrov et al. (2014)	Interview	Bereaved individuals Age range 35–55 years, with an average of 46 years Male (N = 10) Female (N = 20) Norwegian (N = 10) Indigenous Sámi (N = 10) Norwegian/Sámi (N = 8) Minority Kven (N = 2)	N = 30	Lack of scientific rigor; homogeneity of participants makes generalizing information difficult; potential responder bias	Findings indicate that need for help following a traumatic death is universal, and response from providers should be quick and proactive. Barriers to help included stigma and lack of cultural and historical knowledge and implications.
Erlich et al. (2017)	Survey research	Psychiatrists attending Group for the Advancement for Psychiatry meeting to identify postvention training and efforts Average age of 56.8 years Male (N ≈ 65) Female (N ≈ 25) Ethnicities unknown	N = 90	Lower quality of evidence due to this being an observational study; potential self-selection bias; low response rate; potential social desirability bias	Postvention efforts need to be improved, especially as related to mental health clinicians and client loss.
Gould et al. (2018)	Case control study	Students in 12 high schools in New York State, 6 schools where a suicide occurred and 6 schools where there were no suicides Average age of 15.5 years Students who experienced a classmate suicide Male (N ≈ 1,661) Female (N ≈ 1,204) White (N ≈ 2,492) African American (N ≈ 65) Hispanic (N ≈ 117) Asian (N ≈ 88) Other (N ≈ 103) Students who did not experience a classmate suicide Male (N ≈ 1,403) Female (N ≈ 1,016) White (N ≈ 1,879) African American (N ≈ 133) Hispanic (N ≈ 179) Asian (N ≈ 91) Other (N ≈ 137)	Total N = 5,284 Students who experienced a classmate suicide N = 2,865 Students who did not experience a classmate suicide N = 2,419	Lower quality of evidence due to this being an observational study; potential self-selection bias and lower participation rates from schools; potential responder bias	Students with negative life events had a risk of serious suicidal ideation and behaviors and less help-seeking behaviors. Students who were friends with the deceased (though not close) had the greatest risk of serious suicidal ideation and behavior.
Grossman et al. (1995)	One-group pretest-posttest design and survey research	Training provided to school and community caregivers Age, gender, and ethnicity of survey respondents not reported	N = 263	Potential self-selection bias; potential issues with internal consistency and validity; practice effects in the pretests and posttests; lack of comparison groups	After Preparing for Crisis trainings, pretests and posttests indicated that there was a statistically significant increase in knowledge and skills in year one for participants.
Hazell (1991)	Case study on postvention intervention	High-risk adolescent students with relationships to the deceased following three high school student suicides in the same area of Australia Exact ages unknown Genders and ethnicities unknown	N = 25 participants from one school; number of participants from two additional schools unreported	Lacking scientific rigor; unknown sample size and demographics make generalizing results to the wider population difficult	Clinician discussions with students included three main takeaways: Though there are generally suicide risk warning signs, they are difficult to interpret. Irritability and difficulty concentrating or sleeping are common grief reactions. If any students feel depressed or suicidal they must speak with an adult.
Hazell & Lewin (1993)	Nonrandomly controlled trials; intervention study	Students from two high schools in the Lower Hunter region of New South Wales who returned consent forms. One private school and one public school Students from grades 8–11 who experienced a student death by suicide in their school Age, gender, and ethnicities unknown.	Initial postvention, N = 806; 8-month follow-up, N = 126	Lower strength of evidence due to nonrandom assignment; potential attrition and selection bias; no baseline measure of depression and retrospective baseline data on suicidal ideation and behavior; difficulty generalizing findings due to unknown variables of the studied population	A history of suicidal ideation and behaviors was the strongest predictor of current or recent ideation and behavior. Friendships with an individual with a suicide attempt or completion added a significant but “small” component to the overall risk of contagion.
Hoffman et al. (2010)	Phenomenological research, lived experience	Individuals from South Africa bereaved by suicide in late adolescence Age range 17–22 years Male (N = 0) Female (N = 5) Ethnicities unreported	N = 5	Lack of scientific rigor; difficulty generalizing findings due to homogeneity of participants and small sample size	Among suicide survivors, guilt, self-blame, blaming others or God, anger, loss or restriction of “self,” depression, negative behavioral coping patterns, changes in relationship dynamics, and suicidality were prevalent themes in experiences.
Johansson et al. (2006)	Cross-sectional study	Adolescent completed suicides 1981–2000 in Sweden Age range 13–18 years Male (N = 66) Female (N = 22) Ethnicities unknown	N = 88	Difficulty generalizing due to homogeneity across individuals; missing information on data quality; susceptible to bias; potential confounding variables	Studies of teenage suicides in Sweden during 1981–2000. There were two clusters including six teenagers confirmed, and suicide contagion was established within each cluster.
King et al. (2000)	Survey research	School counselor members of the American School Counselor Association Age range 22–69 years; mean = 50 years Male (N ≈ 43) Female (N ≈ 143); White (N ≈ 173) Other ethnicities unknown	N = 186	Lower quality of evidence due to this being an observational study; difficulty generalizing due to homogeneity across individuals; potential self-selection bias	Many school counselors feel ill-equipped to identify a student at risk of suicide, and more training is necessary.
Levi-Belz & Lev-Ari (2019)	Longitudinal cohort study	Individuals in Israel bereaved by a suicide loss Age range 18–70 years Male (N = 24) Female (N = 132) Ethnicities unknown	N = 156	Lower quality of evidence due to this being an observational study; potential reporting bias; lack of control group; snowball sampling technique and sample may decrease the generalizability of the data	Complicated grief lessened over time; survivors with low self-disclosure experienced more complicated grief at both the initial survey and the 18-month follow-up survey.
Mackesy-Amiti et al., (1996)	One-group pretest-posttest design	School personnel engaged in a Prepared for Crisis training and evaluation Male (N = 58) Female (N = 144) Unknown gender (N = 3) Ages and ethnicities unknown	N = 205	Potential issues with internal consistency and validity; practice effects in the pretests and posttests; lack of comparison groups; only one training program with one group of trainers, resulting in no replication studies; potential attrition bias	Researchers developed a knowledge measure to evaluate a suicide postvention training that was found to have construct validity. This supports the notion that training individuals in suicide prevention and postvention efforts can result in knowledge gains.
Mirick et al. (2018)	Survey research	Students from 16 high schools and 12 middle schools from 13 communities in the Northeast United States where there was a student death by suicide Descriptive characteristics of all students were not collected, but these were primarily White suburbs to urban areas with racial diversity Female (N ≈ 4,962) Male (N ≈ 5,022) Ethnicities unknown	N = 9,984	Missing information on demographics on many of the students, which leads to difficulty generalizing; however, large sample size contributes to more generalizability. Screening tools utilized in this research have been validated with White samples, but not with diverse cultural groups. Potential issues related to social desirability bias.	After conducting 10 years of universal school-based screenings for 13 communities as a postvention intervention, a nonprofit trauma agency reported on barriers, strategies for working with school personnel and stakeholders, lessons learned, and recommendations for school screenings. These screenings are based on those utilized in the Signs of Suicide suicide prevention program. Major findings indicate that, for a successful universal screening program, preparatory work, planning, and training must be done with school staff; schools must promote buy-in from families; and schools must establish relationships with families, community stakeholders, and local agencies to promote support, resources, and referrals if necessary.
Müller et al. (2017)	Semi-structured interview	Health professionals from Psychosocial Care Center in Brazil Age range 29–55 years Genders and ethnicities unknown	N = 5	Lack of scientific rigor; small sample size makes generalizing information difficult; potential responder bias	Mental health professionals cite the importance of networking, teamwork, humanized care, and family care when engaged in postvention.
O’Neill et al. (2020)	Survey research	School psychologists employed in the North Carolina public school system during the 2016–2017 academic school year Age, genders, and ethnicities unknown	N = 111	Lower quality of evidence due to this being an observational study; low response rate; potential nested data; potential social desirability bias; difficulty generalizing	Results indicate inadequately trained school-based mental health professionals with gaps in knowledge and lower rates of self-efficacy and perceived knowledge. They are often crisis first responders, but more than half of respondents feel ill-equipped to intervene during postvention.
Pfeffer et al. (2002)	Nonrandomly controlled trials; intervention study	Children bereaved by suicide Bereavement intervention group (after attrition) Average age: 9.8 years Male (N = 13) Female (N = 19) White (N = 25) African American (N = 5) Hispanic (N = 2) Control group (after attrition) Average age: 12.2 years Male (N = 5) Female (N = 4) White (N = 8) African American (N = 1) Hispanic (N = 0)	Total N = 41 Bereavement intervention group (after attrition) N = 32 Control group (after attrition) N = 9	Issues with recruitment and retention led to possible attribution bias; difficulty generalizing due to attrition rates; potential biases due to nonrandom assignment	Changes in anxiety and depressive symptoms were significantly greater among children who received the intervention than in those who did not.
Poijula, Dyregrov, et al. (2001)	Survey research	High school adolescents from a home-room class that experienced a student death by suicide Ages 14–17 years Male (N = 46) Female (N = 43) Ethnicities unknown	N = 89	Lower quality of evidence due to this being an observational study; no reliability and validity data for the questionnaire measuring reactions to school crisis interventions following a student death	Six months following the death of a classmate by suicide, 30% of classmates had scores suggesting PTSD and 9.8% high-intensity grief. Inadequate crisis intervention was a risk factor for high-intensity grief, and friendship was a predictor of both PTSD and high-intensity grief.
Poijula, Wahlberg, et al. (2001)	Natural research design	High school students from three schools in small rural communities of northern Finland that experienced a student suicide (N = 6) Ages 13–17 years: Male (N = 46) Female (N = 43) Ethnicities unknown	N = 89	Research based on a small number of cases, which is not optimal for statistical analysis; lacking sufficient statistical power	The suicide contagion hypothesis was supported. In schools where mental health professionals performed “adequate” talk-throughs and psychological debriefings following a suicide, there were no new suicides.
Roberts (1995)	Case study on school response to suicide postvention	Case study on a small rural public school’s immediate and long-range postvention responses to a student suicide during the summer, when school was out of session Individual was a 14-year- old 8th grade male, race unknown	N = 1 (number of students interviewed was not included)	Difficult to generalize findings due to small sample size and specifics of the rural school; lacking scientific rigor	All schools should have suicide prevention and postvention in their curriculum, with regular staff trainings of identified crisis response members. Psychological autopsies can be a valuable tool to help schools assess their postvention response and identify students in need.
Sandor et al. (1994)	Quasi-experimental study with two intervention conditions and one control condition	Suicide survivor group Teenagers bereaved by a peer suicide, ages 14 −17 years old Male (N = 5) Female (N = 10) All White Comparison group Teenagers ages 14 −18 years old Male (N = 6) Female (N = 13) White (N = 17) Hispanic (N = 2)	Total N = 34 Suicide survivor group N = 15 Comparison group N = 19	Small sample size; difficulty of generalizability; there cannot be study replications for consistency; risk of confounding bias	This study supports the notion that effective postvention efforts must include the community, and mental health providers should extend postvention efforts in the longer term for best results.
Shear et al. (2005)	Randomized controlled trial study	Bereaved adults ages 18–85 years Male (N = 12) Female (N = 83) Ethnicities unknown	Total N = 95 Inter-personal psychotherapy (N = 46) Complicated grief treatment (N = 49)	Higher quality of evidence due to randomized trials; difficulty generalizing due to homogeneity across individuals; potential selection bias; attrition rate (26%); potential confounding variables (45% of participants were on psychotropic medication)	Findings show that both interpersonal psychotherapy and grief treatment will improve complicated grief symptoms, but that complicated grief treatment is an improved treatment relative to interpersonal psychotherapy, showing higher response rates and faster time to response.