Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: School Ment Health. 2021 Jun 22;14(1):125–135. doi: 10.1007/s12310-021-09459-1

Under the Surface: The Role of Covert Cues in Peer Suicide Risk Referrals

LaDonna L Gleason 1, Ansley M Bender 2, Jason I Chen 3, Melanie Bozzay 4, Renee Hangartner 5, Gabriela Romero 6, Christa D Labouliere 7, Meredith Elzy 8, Kimberley Gryglewicz 9, Marc S Karver 10
PMCID: PMC8903059  NIHMSID: NIHMS1734195  PMID: 35273652

Abstract

Suicidal thoughts and behaviors are highly prevalent among adolescents, and peers are often the first, and sometimes only, people to know about youth suicidality. Since many adolescents do not directly disclose suicidal thoughts, school-based suicide prevention programs aim to train youth to recognize warning signs of suicide in their peers that serve as “cues” to refer at-risk peers to an appropriate adult. However, peer-presented cues vary widely in presentation, and adolescents are more likely to recognize overt (i.e., obvious or explicit) as opposed to covert (i.e., hidden or implied) cues. The type of cue exhibited may, in turn, affect whether adolescents make a referral to an adult. The current study examined whether training suicide prevention influences referral intentions for overt and covert suicide cues. Participants included 244 high school students (54% female; Mage = 16.21) in the Southeastern United States who received suicide prevention training (SOS; Signs of Suicide) as part of their health curriculum. Prior to training, students endorsed higher referral intentions for peers exhibiting overt compared to covert cues. Training was associated with increased intentions to refer peers across cue type, but referral intentions for covert cues improved significantly from pre to post-training while those for overt cues remained high and stable. Findings suggest that suicide prevention training might differentially improve students’ ability to detect and respond appropriately to less obvious indicators of suicide risk. These findings may inform the adaptation and development of future, more nuanced school-based suicide prevention programming.

Keywords: youth suicide prevention, warning signs, overt suicide cues, covert suicide cues, referral intentions, gatekeeper training

Suicidal thoughts and behaviors among youth are pressing concerns, particularly given their high prevalence. Every year, approximately 16% of youth between the ages of 10–24 report seriously considering suicide, 13% plan an attempt, and 8% attempt suicide (Centers for Disease Control and Prevention, 2017; Eaton et al., 2012; Lowry et al., 2014). Sadly, youth mental health problems and broader distress often go undetected by adults. Research shows that friends are often the first, or sometimes only, ones to know that an individual is suicidal (Drum et al., 2009; Owens et al., 2009), and adolescents are more likely to disclose distress and suicidal ideation to peers rather than to adults or mental health professionals (Carter & Janzen, 1994; Drum et al., 2009; Fortune et al., 2008; Hazell & King, 1996; Hennig et al., 1998; Michelmore & Hindley, 2012). Yet, not all adolescents directly disclose suicidal ideation even to their friends, and better assessment of suicidal thoughts and behaviors has been indicated as a priority in the field of school psychology (Flett & Hewitt, 2013; Levitt et al., 2007).

One promising avenue for early detection of distress and prevention of suicide risk is the identification of youth suicide warning signs (American Foundation for Suicide Prevention, 2018; Centers for Disease Control and Prevention, 2017). Warning signs are proximal, acute indicators of distress (Rudd, 2003) that may range from overtly talking about suicide to engaging in uncharacteristically impulsive behaviors or expressing that one feels trapped or hopeless (King, 2006; Rudd et al., 2006). Research indicates that most individuals at risk for suicide display changes in behavior and mood (American Foundation for Suicide Prevention, 2018; Boccio, 2015) which could be observable by others as warning signs of distress. Given the amount of time that adolescents spend in school settings each day, fellow students are often best positioned to observe changes in behavior that may signify risk for suicide in a peer. Indeed, identification of warning signs by peers has been named as a priority for preventing adolescent suicide (King, 2006).

How fellow students evaluate peer warning signs of suicide and how they choose to respond can determine whether adolescents in distress are connected to the help they need in a timely manner. Ideally, peer disclosure or observation of warning signs would serve as “cues” for students to refer at-risk adolescents to an appropriate adult; however, research indicates a reluctance among this age group to talk to an adult or mental health professional about peers who express or exhibit risk for suicide (Amarasuriya et al., 2017; Barton et al., 2013; Carter & Janzen, 1994; Dunham, 2004; Kalafat & Elias, 1992; Kalafat et al., 1993; Lang & Lovejoy, 1997). This reluctance may stem in part from stigmatizing attitudes related to suicide or a lack of knowledge about suicide (Overholser et al., 1989; Schilling et al., 2016). For instance, students may not recognize certain warning signs, such as recent problems in school or isolating behavior, as cues for suicide and may instead treat them as problems meriting less concern (King et al., 2011; Mo et al., 2018). This in turn leads to failure to connect at-risk students with appropriate resources. Training students to recognize and respond effectively to suicide cues may be one way to connect more at-risk students to the services they need. To this end, school-based suicide prevention programming is commonly implemented in an effort to increase knowledge of warning signs and peer referral behavior (i.e., telling a trusted adult their concerns about a peer), with both school psychologists and adolescent students viewing curriculum-based suicide-prevention programs favorably (Eckert et al., 2003; Eckert et al., 2006).

While considerable research has been conducted on school-based suicide prevention programming’s effects on students’ help-seeking and suicidal thoughts and behaviors, less empirical attention has been paid to its effects on identification of warning signs in peers and how different types of warning signs affect students’ identification of peers’ risk. In particular, it is unclear if suicide prevention training may result in enhanced detection of subtler warning signs and cues that are often missed, ignored, or misunderstood. Suicide cues differ in several qualities, such as the extent to which they are overt rather than covert. Overt cues are defined as warning signs that more clearly and directly indicate risk for suicide, including talking about wanting to die, writing or expressing a goodbye message, discussing a suicide plan, or revealing access to lethal means. Conversely, covert cues are defined as warning signs which may be more ambiguous or subtle in signifying risk, including increased use of drugs or alcohol, social withdrawal, and expressed feelings of depression or hopelessness (American Association of Suicidology classification system; Rudd et al., 2006). Given that cues have vastly different presentations (King, 2006), the type of cues that students display likely affects how peers evaluate their risk and respond, including whether or not to involve an adult.

Developmental theories of social cognition (e.g., the ability to understand the intentions, beliefs, and emotions of others in social contexts and to take the perspective of another in a communicative context to guide decision-making; Fiske, 2018; Tamnes et al., 2018) suggest that adolescents may still be developing the skills needed to recognize, understand, and respond to peers’ affective communications (Blakemore & Choudhury, 2006; Tamnes et al., 2018; Tousignant et al., 2017; Vetter et al., 2013), including those related to suicide. This may result in adolescents being more likely to miss suicide cues that are less overt, as the skills and prosocial behaviors related to recognition and understanding of suicide cues in peers may still be developing during the high school years. Furthermore, some theories of suicide (IPT; Joiner, 2005; van Orden et al., 2010) state that the desire for suicide is primarily driven by covert interpersonal cues — thwarted belongingness (i.e., an unmet need to belong; Baumeister & Leary, 1995) and the perception that one is a burden on others (Joiner, 2005). Since these cues are more common, suicidal peers may be more likely to express their covert indicators of suicidal ideation and behavior, such as burdensomeness, loneliness, exclusion, or social isolation (van Orden et al., 2010) rather than less-common overt cues, such as explicitly suicidal thoughts, suicidal statements, or plans of suicide. Youth may already be unlikely to seek help for a peer expressing covert cues of suicide, given low base rates (e.g. that most youth who express burdensomeness, loneliness, and other covert cues do not go on to attempt or die by suicide) and subjective norms of self-reliance (e.g. that these are issues youth can or should handle without adult intervention). Youth may further lack knowledge of how more common, covert cues are associated with heightened suicide risk (American Association of Suicidology, 2014), and may accordingly downplay the severity of risk in their peers and choose not to respond to covert cues of suicide. To this end, school-based trainings that emphasize recognition of more ambiguous warning signs, alongside peer response training, may be particularly integral for adolescents navigating this developmental stage of social cognition, especially if programs provide needed scaffolding to promote peer referral in response to both overt and covert cues.

Differentiating between type of presented suicide cue has been a neglected area of research. The extant literature is mixed as to whether adolescents consistently identify affective or behavioral cues of suicide risk. Some studies have suggested that affective cues, such as being depressed, were more easily identified as indicators of suicide risk than behavioral cues, such as engaging in risky behavior or increased consumption of alcohol (Lawrence & Ureda, 1990), whereas other studies have found the opposite (that behavioral symptoms were viewed as more serious indicators of suicide risk; Mueller & Waas, 2002) or that adolescents viewed affective and behavioral suicide cues as equally concerning (Lang and Lovejoy, 1997). What was consistent across studies was that overt cues of suicide, such as talking about death or expressing suicidal ideation, were seen as more serious than covert cues, such as dysphoria or skipping class (Lang and Lovejoy, 1997; Lawrence & Ureda, 1990). Likewise, studies examining how students responded to covert versus overt scenarios involving a suicidal peer found that that students referred peers to an authority or professional help more frequently when presented with overtly suicidal cues than when presented with more covert cues (e.g., depression; Dunham, 2004; Barton et al., 2013). Referrals to professional help (i.e., school counseling services) were the least frequent of all response patterns, with problem-solving help and offers of social support being the predominate responses when presented with covert cues (Barton et al., 2013). Taken together, it appears that type of suicide cue influences evaluation of risk and referral intentions, and that overt cues seem to be consistently identified as indicators of suicide risk more likely to warrant referral. This presents cause for concern given the considerable evidence for association between suicide and covert cues like social isolation (Calati et al., 2019; Trout, 1980), depression (Bhatia & Bhatia, 2007; Hawton et al., 2013), and hopelessness (Boergers et al., 1998; Zhang & Li, 2013). Adolescents may already be adept at recognizing overt cues as meriting concern even without specific training in suicide prevention, whereas programs that emphasize recognition of covert cues may result in increased numbers of referrals. However, to date, no studies have focused on whether suicide prevention training differentially impacts referral intentions in response to overt versus covert suicide cues.

We aimed to address this gap in the present study. Before and after participating in a school-based suicide prevention training, high school students were presented with overt and covert suicide cues and asked to report their referral intentions. The suicide prevention curriculum used was the Signs of Suicide (SOS) program, which has been extensively validated with high school students and shown to produce increased help-seeking and peer referral behavior (Aseltine, 2003; Aseltine & DeMartino, 2004; Aseltine et al., 2007; Schilling et al., 2016). To parse the effect of training on different types of suicide cues (overt vs. covert), the present study examined how students differentially formulate referral intentions for overt and covert suicide cues before and after receiving evidence-based suicide prevention training. To this end, two hypotheses were tested: 1) at baseline, students would report greater referral intentions for overt suicide cues relative to covert suicide cues, and 2) cue type would interact with time to predict referral intentions, so that referral intentions for covert suicide cues would increase pre- to post-training, whereas referral intentions for overt suicide cues would remain stable.

Method

Participants

Participants were 244 high school students in four schools in the Southeastern United States who participated in the SOS suicide-prevention curriculum as part of their health education class. Nearly 46% percent identified as male, 53.7% as female, and 0.4% identified as transgender. Of the students who opted to self-identify their race (98.2%), approximately 60.7% identified as White, 20.9% as Black or African American, 6.1% as Multiracial, 2.0% as Asian, and 0.4% identified as another racial group. Additionally, 18% percent identified as being of Hispanic or Latinx ethnicity. More than half of the sample was in 9th grade (65.2%), while fewer were in 10th (8.2%), 11th (8.2%), or 12th (18.4%) grades, as the suicide prevention curriculum was administered during a health class most commonly taken during freshman year. The mean age was 16.21 (SD = 1.74).

Measures

The Planned Behavior and Implementation Questionnaire (PBIQ) is a study-designed questionnaire (Gryglewicz et al., 2015; Totura, et al., 2008) widely-used in our school-based suicide prevention studies (Totura et al., 2019a; 2019b), based on the Theory of Planned Behavior (Ajzen, 1985; 2017) and theoretical factors associated with successful implementation (Fixsen, et al., 2005). The full questionnaire contained 48 pretest items and 54 posttest items, including demographic factors, youths’ suicide prevention knowledge, attitudes about suicide prevention, intentions to engage in referral behaviors, confidence in suicide prevention skills and knowledge (pre- and post-training), involvement in training, alliance with the teacher trainer, and fidelity of training implementation (post-training only). Subscales of the PBIQ have shown reliability in previous studies with Cronbach’s alpha scores consistently above 70 (α = 72, α = 75, α = 84 in Totura, et al., 2019a; α = 70, α = 77 in Totura, et al., 2019b). In this study, only the demographic, fidelity, and cue-related referral intention questions were used.

Demographics factors were assessed by five items including age, grade, gender, race, and ethnicity. Training fidelity was assessed by eight items, six assessing required elements of SOS training as outlined in the manualized protocol (e.g. “we learned the warning signs of suicide”) and two assessing optional elements (e.g., “we discussed suicide statistics”) to which participants responded “yes” or “no.” Referral intentions were assessed by 12 items on a Likert-scale ranging from 1 (very likely) to 5 (very unlikely) asking about their likelihood of telling an adult if a friend exhibited various warning signs of suicide, reflecting both overt suicide cues (e.g. told you his/her plan for suicide, said or wrote a goodbye message; n=6) and covert suicide cues (e.g. seemed depressed most of the time, was avoiding you and other friends; n=6). Reliability for the scale was strong with α = 0.91 at pre and α = 0.95 at post. Items were reverse-scored, such that higher scores indicated greater referral intentions, and participants’ responses were averaged to produce mean referral intentions. For the purposes of this project, 19 undergraduate research assistants (RAs), knowledgeable in youth suicide prevention, and 5 of the authors of this paper independently classified the 12 intentions items as either “overt” or “covert” in concert with the American Association of Suicidology classification system (Rudd et al., 2006). Coder agreement was high (82.6%) and any discrepancies were resolved via group review and consensus among the project team.

Procedure

Evaluation

Students completed all study measures during health education classes. Measures were completed in paper format, with pre-training measures completed one week prior to the suicide prevention training and post-training measures completed immediately following training.

Suicide Prevention Training

Student participants received the 60-minute Signs of Suicide (SOS) suicide prevention training during their health education class. This training followed a manualized protocol and was led by one designated health teacher in each school who had been trained by programmatic staff with expertise in suicide prevention trainings, including SOS. Before the trainings were implemented, programmatic staff shared manualized material with each school’s health teacher, coached them in delivery of the protocol, and modeled delivery of the training with fidelity.

The SOS curriculum is designed to raise awareness about suicide, while also educating students on the identification of associated warning signs and risk factors (Aseltine, 2003; Aseltine & DeMartino, 2004). SOS incorporates both screening (e.g., assess which students may require intervention) and gatekeeper training (e.g., teach how to identify warning signs of suicide and respond effectively; Katz et al., 2013). SOS teaches students to “ACT,” or to acknowledge warning signs of suicide observed in peers, to express care for the student, and to tell an adult. The SOS program was included on SAMHSA’s National Registry of Effective Programs (Suicide Prevention Resource Center, nd), and has extensive research demonstrating that participating in the curriculum improves knowledge and attitudes, including intervening with at-risk peers (Aseltine & DeMartino, 2004; Aseltine et al., 2007; Schilling et al., 2016), increased help-seeking (up to 60%; Aseltine, 2003), significant reductions in suicidal ideation (Schilling et al., 2016; even among students with elevated baseline ideation, Schilling et al., 2014), and reduced rates of suicide attempts (as much as 40–64% less in randomized control trials; Aseltine & DeMartino, 2004; Aseltine et al., 2007; Schilling et al., 2016).

Data Analysis

Of the 246 participants who were trained and evaluated, 80.89% (n = 199) had complete data. Missingness ranged from 1–4 items per case and were spread throughout the dataset without a discernable pattern. Because the study was adequately powered with the 199 complete cases, listwise deletion was chosen as an acceptable method of handling missing data for this study.

Prior to hypothesis testing, all variables were screened for statistical assumptions. To address negative skew and leptokurtosis, logarithmic transformations were conducted, and analyses were run with and without transformed values. Results did not differ, so untransformed results are reported to ease interpretation. Likewise, associations between referral intentions and all demographic variables were examined, and analyses were run with and without including significant demographic variables as covariates; results did not differ, so results are presented without covariates to ease interpretation. All hypotheses were tested with repeated measures analysis of variance (ANOVA). A two-way repeated measures ANOVA with time, cue type, and their interaction as predictors of referral intentions was used to determine if baseline intentions varied between covert and overt cues (H1) and if changes in intentions from pre- to post-training varied based on cue type (H2). In exploratory analyses, a repeated measures multiple ANOVA (MANOVA) was run with posthoc, item-level ANOVAs using a Bonferroni correction to determine which specific overt and covert cues were most likely to produce referral intentions.

Results

Preliminary Results

Descriptive statistics and correlation coefficients are presented in Table 1. Prior to hypothesis testing, associations between referral intentions and all demographic variables were examined; gender was the only demographic variable significantly related to referral intentions, in that female participants (n=103) were significantly more likely than male participants to endorse referral intentions for both overt and covert cues of suicide at both timepoints (Pre Overt: F2, 196 = 9.30, p < .001, η2 = .08; Pre Covert: F2, 241 = 7.07, p = .001, η2 = .06; Post Overt: F2, 241 = 5.12, p = .007, η2 = .05; Post Covert: F2, 241 = 5.18, p = .006, η2 = .05). Analyses were initially run using gender as a covariate; however, as gender was not found to be significant in any model and results did not differ based on gender’s inclusion, results are presented without covariates to improve parsimony. The intraclass correlation coefficient (< 2%) showed little effect by the clustering variable (school) on the outcome variable, therefore subsequent analyses were run without nested models. As expected, the SOS suicide prevention training was administered with high fidelity (88.9%), and referral intentions increased from pre- to post-training (F1, 198 = 29.59, p < .001, d = 0.39; see Figure 1 and Table 1), in concert with the previous literature on the SOS program.

Table 1.

Descriptive statistics and correlations for referral intentions by cue type and time.

Referral Intentions Mean SD Min Max Skew Kurtosis 1. 2. 3. 4. 5. 6.
1. Pre-training covert cues 3.83 0.85 1.00 5.00 −1.04 1.11 1.00
2. Pre-training overt cues 4.50 0.73 1.00 5.00 −2.69 9.04 .75** 1.00
3. Post-training covert cues 4.24 0.83 1.00 5.00 −1.98 4.80 .66** .66** 1.00
4. Post-training overt cues 4.55 0.81 1.00 5.00 −2.83 8.62 .58** .69** .86** 1.00
5. Pre-training total cues 4.17 0.74 1.00 5.00 −1.92 5.14 .95** .93** .70** .67** 1.00
6. Post-training total cues 4.39 0.79 1.00 5.00 −2.57 7.53 .64** .70** .97** .96** .71** 1.00
Open in a new tab

Mean, SD, Skew, Kurtosis, Correlations

Note:

**

p < .01

Figure 1.

Figure 1.

Open in a new tab

Changes in student referral intentions from pre- to post-suicide prevention training in response to overt and convert suicide cues

Hypothesis Testing

At baseline, a significant effect of cue type (F1, 198 = 269.08, p < .001, d = 1.12) was found, wherein referral intentions in response to covert cues were significantly lower than in response to overt cues (supporting H1). As hypothesized (H2), only referral intentions for covert cues increased from pre- to post-training (F1, 198 = 65.79, p < .001) and a significant interaction effect was found (F1, 198 = 65.55, p < .001) wherein referrals in response to covert cues increased significantly from pre- to post-training (d = 0.59), whereas referrals in response to overt cues remained stable (d = 0.08; see Figure 1).

Exploratory Analyses

Follow-up exploratory analyses were conducted to determine which specific covert cues were most likely to produce referral intentions. After training, cues most likely to result in peer referral included “Told you his/her plan for suicide” (M = 4.79, SD = 0.74), “Said or wrote a goodbye message” (M = 4.71, SD = 0.81), and “Made a comment about wanting to die” (M = 4.63, SD = 0.84), all overt cues.

Repeated measures MANOVA indicated significant changes in referral intentions for covert cues from pre- to post-training, F6, 198 = 14.10, p < .001, ηp2 = .31. Follow-up post hoc analyses revealed that referral intentions were most likely to improve as a result of training for the following cues: depression (F1,198 = 41.07, p < .001, ηp2 = .17), hopelessness (F1,198 = 31.55, p < .001, ηp2 = .14), and asking to keep depression or hopelessness a secret (F1,198 = 54.93, p < .001, ηp2 = .22). Referral intentions for overt cues did not change significantly from pre to post-training training, likely due to ceiling effects resulting from very high referral intentions endorsed at baseline. Item-level analyses for individual cues are presented in Table 2.

Table 2.

Exploratory item-level analysis of which specific overt and covert cues were most likely to result in referral intentions

Cue Pre-Training M (SD) Post-Training M (SD) Cohen’s d F pa
Covert Cues
Was avoiding you and other friends 3.49 (1.25) 3.85 (1.08) 0.29 18.83 <.001
Seemed depressed most of the time 3.87 (1.08) 4.34 (0.98) 0.44 41.07 < .001
Began drinking more alcohol 3.97 (1.11) 4.26 (1.00) 0.29 17.51 <.001
Began using more drugs 4.16 (1.10) 4.37 (1.00) 0.21 9.58 .002
Talked to you about feeling depressed or hopeless 3.95 (1.05) 4.35 (0.95) 0.38 31.55 < .001
Asked you to keep [depression or hopelessness] a secret 3.56 (1.32) 4.25 (1.05) 0.48 54.93 < .001
Overt Cues
Made a comment that life wasn’t worth living anymore 4.37 (0.97) 4.58 (0.86) --- --- ---
Made a comment about wanting to die 4.48 (0.88) 4.63 (0.84) --- --- ---
Told you his/her plan for suicide 4.79 (0.74) 4.72 (0.86) --- --- ---
Told you he/she has a weapon 4.31 (1.12) 4.31 (1.08) --- --- ---
Showed you a weapon 4.34 (1.08) 4.35 (1.07) --- --- ---
Said or wrote a goodbye message 4.71 (0.81) 4.69 (0.87) --- --- ---
Open in a new tab
a

Significance values for exploratory analyses were adjusted with Bonferroni corrections

Discussion

This study presents an initial exploration of how cue type (i.e. overt versus covert) influences changes in referral intentions following youth suicide prevention training. As expected, training was associated with increases in referral intentions from pre- to post- training, replicating past research validating the SOS program (Aseltine & DeMartino, 2004; Aseltine et al., 2007; Schilling et al., 2016). However, student intentions to refer a suicidal peer differed depending on whether the peer displayed suicide cues that were overt or covert. At baseline, participants reported higher referral intentions for overt cues than covert cues. However, referral intentions for covert cues improved significantly from pre to post-training while those for overt cues remained stable, suggesting that SOS training might differentially improve students’ ability to detect and respond appropriately to less obvious indicators of suicide risk. These findings may inform the adaptation and development of future youth suicide prevention programming.

At baseline, students endorsed high referral intentions for overt cues (e.g., all in the agree-to-strongly agree range at baseline), with the highest endorsed cues being if another student revealed having a plan for suicide or wrote a goodbye message. This is unsurprising, as a sizeable proportion of adolescents (up to 24%) have been exposed to peers who have attempted or died by suicide and may have previously observed overt markers of suicide risk (Bearman & Moody, 2004; Mitchell et al., 2019; Randall et al., 2015; Swanson & Colman, 2013). While the high level of referral intentions is a positive phenomenon, it is important to recognize that suicide attempt survivors often report negative reactions from friends when displaying overt cues of suicide risk (Frey et al., 2017). Specifically, attempt survivors report experiencing stigmatizing comments and anxiety from friends and family that can lead to feeling like a burden upon others (Frey et al., 2017). This is especially concerning, since perceiving that one is a burden upon others (i.e. perceived burdensomeness) has been identified as a key driver of suicidal desire (e.g. Joiner, 2005; Joiner et al., 2002). Given our findings, adolescents have high intentions to refer peers exhibiting overt cues of suicide risk, even prior to gatekeeper training. Because most youth know to refer overt cues, training should emphasize teaching a more empathic, sensitive response to covert cues.

In contrast to overt cues, referral intentions for covert cues at baseline were significantly lower, supporting our hypothesis (H1). This supports earlier findings that suggested adolescents are more likely to refer peers with overt behaviors when given hypothetical scenarios with covert or overt suicidal behaviors (Dunham, 2004; Barton, 2013). Recognizing the discrepancy in referral intentions at baseline is important, as it emphasizes a specific need for improvement in responding to less obvious signs of suicidal thoughts or behavior in untrained adolescents. Our findings that suicide prevention training differentially improved referral intentions for covert cues suggest that improving the recognition of these cues has the potential to increase referral behavior.

As hypothesized (H2), students showed significant increases in referral intentions for covert cues but not overt cues from before to after training. Given the observed ceiling effect for overt cues, these findings can improve the specificity of gatekeeper programs by suggesting that there is more potential for improvement when we target covert cues in training. School psychologists may consider more directly targeting covert cues and their importance in suicide prevention training in light of low levels of referral intentions at baseline and the increase in these intentions post-training. Covert cue recognition and referral is critical in identifying adolescents at risk of suicide, so the increase in referral intentions for these cues in our sample is encouraging. Given the subtle nature of covert cues, recognizing these among peers may be a particularly nuanced skill for adolescents to develop. Indeed, because suicide is a relatively low base-rate behavior, the majority of youth who display a covert cue will not go on to attempt suicide, making it more difficult for peers to differentiate when a specific cue may be a warning sign. As such, an item level discussion may be warranted to begin exploring referral behavior in the context of specific covert cues.

Students displayed increases on the majority of covert cue items to levels comparable to those of overt cues (i.e. in the agree-to-strongly agree range) following training, with the exception of social isolation (“avoiding you and other friends”). Participants continued to have lower referral intentions for peers who exhibited social isolation: a critical predictor of suicidal behavior. This is highly concerning given that social isolation is a consistently strong predictor of suicidal ideation and behavior (van Orden et al., 2010). Studies show that socially withdrawn youth have greater difficulty connecting with peers and are less interested in engaging with peers (Coplan et al., 2013; Kingery et al., 2010). Thus, it is possible that youth report lower referral intentions for social withdrawal because they have poorer relationships with these youth and may be less motivated to intervene with them. It is also possible that due to less frequent interactions with socially withdrawn youth, students may feel they lack sufficient information from peer encounters to determine if a person is at risk, leading them to report intentions closer to the midpoint of the scale. The lower level of referral intentions with socially withdrawn youth is particularly concerning when interpreted in the context of qualitative research findings that people tend to engage in avoidant reactions with suicidal individuals (e.g. disengaging from socializing with the suicidal individual or talking about their suicidality; Frey et al., 2017). This pattern of social withdrawal of the suicidal individual and avoidance of peers contributes to a thwarted need to belong, a key driver of suicidal desire in Joiner’s (2005) theory of suicidal behavior. School psychologists may wish to focus on increasing awareness of social withdrawal as a potential suicide warning sign based on the limited effectiveness of referral training on responses to this covert cue.

While increasing knowledge and awareness of suicide warning signs among peers remains an important primary component of prevention, information alone may not be sufficient to foster peer referral behavior among all adolescents. Persistent stigma around mental health issues in general, and suicide specifically, could increase some adolescents’ feelings of discomfort with engaging peers in conversation about suicide. Additionally, barriers to understanding suicide communication, like mismatched words and affect, indirectness or politeness, and other face-saving behaviors add complexity to an already difficult task for a young person (Owen et al., 2012). Given the still developing nature of social cognition in adolescence and the influence of self-efficacy in feeling confident to perform newly learned skills (Bandura, 1983), prevention training that incorporates in-vivo practice and exposure to uncomfortable conversation may enhance learning and encourage future behavior. Future studies could expand upon overt and covert cue distinction in behavioral learning contexts (i.e., in-vivo roleplay) to include assessment of self-efficacy. There remains considerable opportunity to target covert cues in suicide prevention training that includes modeling and roleplay practice in addition to information sharing.

The results of this study should be interpreted in light of several limitations. All measures were self-report and thus subject to associated biases. However, self-report measures are a feasible, efficient means by which to begin studying referral intentions, and such measures may yield greater self-disclosure than interviews or other data collection formats when studying suicide-related constructs (Corrigan & Watson, 2002; Kaplan et al., 1994). Nonetheless, future studies should consider employing alternative methodologies, including behavioral observations or focus groups. The study likewise may have been limited by the brevity of the suicide cues measure; that is, the included cues may not have fully captured the broad range of overt and covert cues that would signal the need for referral to a trusted adult. However, we identified cues based on the literature (e.g. American Association of Suicidology’s consensus on warning signs for suicide; Rudd et al., 2006) and those which are taught in suicide prevention trainings. Further, the present classification of overt and covert cues demonstrated strong inter-rater reliability and internal consistency and was derived from measures that have been well-validated and widely-used in other suicide prevention studies. Nonetheless, researchers may wish to build on the current findings by proposing and testing other overt and covert cues which may conceivably be influenced by referral training.

An additional limitation of the study was the inability to assess actual referral behavior; instead, referral intentions were assessed. Given the nascent nature of this research, this again presented as a feasible means of beginning to study the present effects without having to conduct a lengthy follow-up. Additionally, intentions are among the best predictors of actual behavior (e.g. Armitage & Conner, 2001; Winkelnkemper, 2014). For example, a recent study of a gatekeeper training demonstrated that greater intentions to apply recently learned behavioral knowledge were a strong predictor of actual engagement in suicide prevention behaviors (Kuhlman et al., 2017). This provides further support for the acceptability of measuring behavioral intentions as a proxy for behavior as originally proposed by Ajzen (1985; 2017) in the Theory of Planned Behavior. Future researchers are encouraged to extend this work by examining how youth suicide prevention trainings may uniquely influence referral behaviors in response to covert cues of suicide.

As a further limitation, the study was not able to utilize a control group due to limitations of the school setting; as such, changes in referral intentions may be subject to threats to internal validity such as history or maturation. However, we intentionally selected a short duration of time between pre and post-training assessment to reduce the likelihood of significant maturation, as well as selected a training program that was well-validated as an effective tool for increasing peer referral (e.g. Aseltine, 2003; Aseltine & DeMartino, 2004; Aseltine et al., 2007; Schilling et al., 2016). As such, while we cannot entirely rule out the role of history, maturation, or other third variables being responsible for differential improvement in referral intentions to overt versus covert cues, it is highly probable that training influenced referral increases. In a within-subjects design such as this one, we are directly comparing youth’s referral intentions to overt cues to their own referral intentions to covert cues, controlling for their baseline intentions. Given the nature of our exploration of differential responses to overt versus covert cues, a within-subjects design is appropriate (Leong & Austin, 2016). It is recommended that future studies incorporate a control group to account for potential threats to internal validity and rule out additional uncontrolled confounding variables that may affect results. An additional limitation is the possibility of selection bias. That is, it is possible that adolescents who provided consent or assent with parental consent differed meaningfully from those who did not. Future research may wish to assess for characteristics which may both predict study participation and influence referral intentions. However, a strength of our sample was its fairly balanced ratio of male and female students and moderate diversity with regards to race (e.g. nearly 40% nonwhite). Finally, it is unclear as to the degree to which changes in referral intentions would be sustained over time, given our inability to collect follow-up data as a result of limitations associated with conducting research in the school setting. However, other studies have documented a sustained effect of SOS training on intentions and perceived behavioral control to engage in best-practice suicide prevention behaviors at one-month follow-up (Gryglewicz et al., 2015). Examining the stability of improved intentions using a longitudinal study design with follow-up presents as an additional opportunity for future research.

Despite these limitations, this study was the first to evaluate how training impacts differences in referral intentions based on covert versus overt cues. The study was strengthened by its inclusion of a large, diverse sample and a within-subjects design. Results suggest that students have lower referral intentions for covert cues at baseline, but that referral training is effective for increasing attention to these important, more subtle markers of suicide risk. Future research should consider enhanced designs including longer term follow-up, incorporation of measures of actual referral behavior over the follow-up period, comparison with a control condition who does not receive suicide prevention training, and further exploration of a broader range of overt and covert cues. Such studies may facilitate the development of more nuanced suicide prevention training programs.

As expected, due to the novel nature of our research question, many additional questions remain in the context of understanding referral intentions in the context of covert and overt suicidal behavior cues. For example, future research needs to expand on these findings to explore reasons that baseline discrepancies exist in the referral of overt and covert behaviors. It remains unclear whether this is due to a lack of recognition of these cues as suicide warning signs or if this is due to a reticence to disclose someone’s mental health problems due to ambiguity or a perceived desire for privacy. In the context of gatekeeper trainings, there has been limited research on covert cues. Future research should explore which covert cues may be most difficult for youth to identify – and why – to inform future programming. Finally, future studies that follow up after gatekeeper trainings are delivered should also explore how peer relationships and interpretation of covert behaviors contribute to responses.

Acknowledgments:

This work was supported by the Substance Abuse and Mental Health Services Administration (SAMHSA; grant number U79 SM060427–01). This publication was made possible, in part, by grant number T32-GM081740 from National Institutes of Health – National Institute of General Medical Sciences (NIH-NIGMS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of SAMHSA, the NIGMS or NIH.

Funding

This work was supported by the Substance Abuse and Mental Health Services Administration [grant number U79 SM060427–01]. This publication was made possible in part by Grant Number T32-GM081740 from NIH-NIGMS.

Footnotes

Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.

Compliance with Ethical Standards

Research involving Human Participants

The project was approved by the University of South Florida Institutional Review Board (IRB) and performed in accordance with ethical standards consistent with the 1964 Declaration of Helsinki and its later amendments.

Informed Consent

Parental consent and youth assent were required for students under 18 years of age, whereas students over 18 years of age provided informed consent. Assent and consent documents were distributed and collected by teachers prior to the beginning of the study.

Conflict of Interest

The authors declare that they have no conflicts of interest to report.

Availability of Data and Material

For ethical and legal reasons, we are unable to publicly share the database used for this study. However, individual requests will be considered with data sharing agreements and IRB approvals in place.

Code Availability

Not applicable.

Contributor Information

LaDonna L. Gleason, University of South Carolina.

Ansley M. Bender, University of South Florida

Jason I. Chen, Oregon Health and Science University

Melanie Bozzay, Brown University

Renee Hangartner, University of South Florida

Gabriela Romero, UAB College of Nursing

Christa D. Labouliere, Columbia University

Meredith Elzy, University of Tampa

Kimberley Gryglewicz, University of Central Florida

Marc S. Karver, University of South Florida

References

  1. Ajzen I (1985). From intentions to actions: A theory of planned behavior. In Kuhl J & Beckmann J (Eds), Action Control. SSSP Springer Series in Social Psychology Springer. [Google Scholar]
  2. Ajzen I (2017). Constructing a theory of planned behavior questionnaire. 2006. Ref Type: Online Source http://people.umass.edu/aizen/tpb.html
  3. Amarasuriya SD, Reavley NJ, Rossetto A, & Jorm AF (2017). Helping intentions of undergraduates towards their depressed peers: a cross-sectional study in Sri Lanka. BioMed Central Psychiatry, 17(1), 40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. American Association of Suicidology. (2014). Depression and suicide risk https://www.suicidology.org/portals/14/docs/resources/factsheets/2011/depressionsuicide2014.pdf
  5. American Foundation for Suicide Prevention. (2018). Risk factors and warning signs https://afsp.org/about-suicide/risk-factors-and-warning-signs/
  6. Armitage CJ, & Conner M (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471–499. [DOI] [PubMed] [Google Scholar]
  7. Apperly IA, Carroll DJ, Samson D, Humphreys GW, Qureshi A, & Moffitt G (2010). Why are there limits on theory of mind use? Evidence from adults’ ability to follow instructions from an ignorant speaker. The Quarterly Journal of Experimental Psychology, 63(6), 1201–1217. [DOI] [PubMed] [Google Scholar]
  8. Aseltine RH (2003). An evaluation of a school based suicide prevention program. Adolescent & Family Health, 3, 81–88. [Google Scholar]
  9. Aseltine RH, & DeMartino R (2004). An outcome evaluation of the SOS suicide prevention program. American Journal of Public Health, 94(3), 446–451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Aseltine RH, James A, Schilling EA, & Glanovsky J (2007). Evaluating the SOS suicide prevention program: A replication and extension. BioMed Central Public Health, 7(1), 161–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bhatia SK, & Bhatia SC (2007). Childhood and adolescent depression. American Family Physician, 75(1), 73–80. [PubMed] [Google Scholar]
  12. Bandura A (1983). Self-efficacy determinants of anticipated fears and calamities. Journal of Personality and Social Psychology, 45(2), 464. [Google Scholar]
  13. Barton AL, Hirsch JK, & Lovejoy MC (2013). Peer response to messages of distress. Crisis, 34(3), 183–191. [DOI] [PubMed] [Google Scholar]
  14. Baumeister RF & Leary MR (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://www.ncbi.nlm.nih.gov/pubmed/7777651 [PubMed] [Google Scholar]
  15. Bearman PS, & Moody J (2004). Suicide and friendships among American adolescents. American Journal of Public Health, 94(1), 89–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Blakemore SJ, & Choudhury S (2006). Development of the adolescent brain: Implications for executive function and social cognition. Journal of Child Psychology and Psychiatry, 47(3‐4), 296–312. [DOI] [PubMed] [Google Scholar]
  17. Boccio DE (2015). A school-based suicide risk assessment protocol. Journal of Applied School Psychology, 31(1), 31–62. [Google Scholar]
  18. Boergers J, Spirito A, & Donaldson D (1998). Reasons for adolescent suicide attempts: Associations with psychological functioning. Journal of the American Academy of Child & Adolescent Psychiatry, 37(12), 1287–1293. [DOI] [PubMed] [Google Scholar]
  19. Calati R, Ferrari C, Brittner M, Oasi O, Olié E, Carvalho AF, & Courtet P (2019). Suicidal thoughts and behaviors and social isolation: A narrative review of the literature. Journal of Affective Disorders, 245(15), 653–667. [DOI] [PubMed] [Google Scholar]
  20. Carter SP, & Janzen HL (1994). Peer counseling or peer support—There is a difference. Canadian Journal of School Psychology, 10(1), 36–42. [Google Scholar]
  21. Centers for Disease Control and Prevention. (2017). Suicide Among Youth https://www.cdc.gov/healthcommunication/toolstemplates/entertainmented/tips/SuicideYouth.html
  22. Choudhury S, Blakemore SJ, & Charman T (2006). Social cognitive development during adolescence. Social Cognitive and Affective Neuroscience, 1(3), 165–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Coplan RJ, Rose-Krasnor L, Weeks M, Kingsbury A, Kingsbury M, & Bullock A (2013). Alone is a crowd: Social motivations, social withdrawal, and socioemotional functioning in later childhood. Developmental Psychology, 49(5), 861–875. [DOI] [PubMed] [Google Scholar]
  24. Corrigan PW, & Watson AC (2002). The paradox of self‐stigma and mental illness. Clinical Psychology: Science and Practice, 9(1), 35–53. [Google Scholar]
  25. Decety J, & Lamm C (2006). Human empathy through the lens of social neuroscience. The Scientific World Journal, 6, 1146–1163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Drum DJ, Brownson C, Burton Denmark A, & Smith SE (2009). New data on the nature of suicidal crises in college students: Shifting the paradigm. Professional Psychology: Research and Practice, 40(3), 213–222. [Google Scholar]
  27. Dunham K (2004). Young adults’ support strategies when peers disclose suicidal intent. Suicide and Life-Threatening Behavior, 34(1), 56–65. [DOI] [PubMed] [Google Scholar]
  28. Eaton DK, Kann L, Kinchen S, Shanklin S, Flint KH, Hawkins J, Harris WA, Lowry R, McManus T, Chyen D, Whittle L, Lim C, & Wechsler H (2012). Youth risk behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report: Surveillance Summaries, 61(4), 1–162. https://www.jstor.org/stable/e24806045 [PubMed] [Google Scholar]
  29. Eckert TL, Miller DN, DuPaul GJ, & Riley-Tillman TC (2003). Adolescent suicide prevention: School psychologists’ acceptability of school-based programs. School Psychology Review, 32(1), 57–76. [Google Scholar]
  30. Eckert TL, Miller DN, Riley-Tillman TC, & DuPaul GJ (2006). Adolescent suicide prevention: Gender differences in students’ perceptions of the acceptability and intrusiveness of school-based screening programs. Journal of School Psychology, 44(4), 271–285. [Google Scholar]
  31. Fiske ST (2018). Social beings: Core motives in social psychology United States: Wiley. [Google Scholar]
  32. Fixsen DL, Naoom SF, Blase KA, Friedman RM, & Wallace F (2005). Implementation research: A synthesis of the literature Tampa, FL: University of South Florida, Louis de la Parte Florida Mental Health Institute. [Google Scholar]
  33. Flett GL, & Hewitt PL (2013). Disguised distress in children and adolescents “flying under the radar”: Why psychological problems are underestimated and how schools must respond. Canadian Journal of School Psychology, 28(1), 12–27. [Google Scholar]
  34. Fortune S, Sinclair J, & Hawton K (2008). Help-seeking before and after episodes of self-harm: A descriptive study in school pupils in England. BioMed Central Public Health, 8(1), 369–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Frey LM, Hans JD, & Cerel J (2017). An interpretive phenomenological inquiry of family and friend reactions to suicide disclosure. Journal of Marital and Family Therapy, 43(1), 159–172. [DOI] [PubMed] [Google Scholar]
  36. Gryglewicz K, Gleason L, Chen J, Steding L, Monahan MF, Jordan-Arthur B, Brown R, Bozzay M, Romero G, Rakhshani A, Sulbaran A, Ruiz Chaves D, Witmeier M, & Karver M (2015). Florida Youth Suicide Prevention Project Washington DC: Substance Abuse and Mental Health Service Administration. [Google Scholar]
  37. Hawton K, i Comabella CC, Haw C, & Saunders K (2013). Risk factors for suicide in individuals with depression: a systematic review. Journal of Affective Disorders, 147(1–3), 17–28. [DOI] [PubMed] [Google Scholar]
  38. Hazell P, & King R (1996). Arguments for and against teaching suicide prevention in schools. Australian and New Zealand Journal of Psychiatry, 30(5), 633–642. [DOI] [PubMed] [Google Scholar]
  39. Hennig CW, Crabtree CR, & Baum D (1998). Mental health CPR: Peer contracting as a response to potential suicide in adolescents. Archives of Suicide Research, 4(2), 169–187. [Google Scholar]
  40. Joiner. (2005). Why People Die by Suicide Harvard University Press. https://www.google.com/books/edition/Why_People_Die_by_Suicide/YqN-8XHDNX8C?hl=en&gbpv=1&dq=inauthor:%22Thomas+Joiner%22&printsec=frontcover [Google Scholar]
  41. Joiner T, Pettit JW, Walker RL, Voelz ZR, Cruz J, Rudd MD, & Lester D (2002). Perceived burdensomeness and suicidality: Two studies on the suicide notes of those attempting and those completing suicide. Journal of Social and Cliical Psychology, 21(5), 531–545. [Google Scholar]
  42. Kalafat J, & Elias M (1992). Adolescents’ experience with and response to suicidal peers. Suicide and Life-Threatening Behavior, 22(3), 315–321. [PubMed] [Google Scholar]
  43. Kalafat J, Elias M, & Gara MA (1993). The relationship of bystander intervention variables to adolescents’ responses to suicidal peers. Journal of Primary Prevention, 13(4), 231–244. [DOI] [PubMed] [Google Scholar]
  44. Kaplan ML, Asnis GM, Sanderson WC, Keswani L, de Lecuona JM, & Joseph S (1994). Suicide assessment: Clinical interview vs. self-report. Journal of Clinical Psychology, 50(2), 294–298. [DOI] [PubMed] [Google Scholar]
  45. Katz C, Bolton SL, Katz LY, Isaak C, Tilston‐Jones T, Sareen J, & Team SCSP (2013). A systematic review of school‐based suicide prevention programs. Depression and Anxiety, 30(10), 1030–1045. [DOI] [PubMed] [Google Scholar]
  46. King. (2006). Practical strategies for preventing adolescent suicide. Prevention Researcher, 13(3), 8–11. http://www.tpronline.org/product_toc/13-3-TOC-preview.pdf [Google Scholar]
  47. King KA, Strunk CM, & Sorter MT (2011). Preliminary effectiveness of Surviving the Teens® suicide prevention and depression awareness program on adolescents’ suicidality and self-efficacy in performing help-seeking behaviors. Journal of School Health, 81(9), 581–590. [DOI] [PubMed] [Google Scholar]
  48. Kingery JN, Erdley CA, Marshall KC, Whitaker KG, & Reuter TR (2010). Peer experiences of anxious and socially withdrawn youth: An integrative review of the developmental and clinical literature. Clinical Child and Family Psychology Review, 13(1), 91–128. [DOI] [PubMed] [Google Scholar]
  49. Kuhlman ST, Walch SE, Bauer KN, & Glenn AD (2017). Intention to enact and enactment of gatekeeper behaviors for suicide prevention: An application of the theory of planned behavior. Prevention Science, 18, 704–715. [DOI] [PubMed] [Google Scholar]
  50. Lang A, & Lovejoy MC (1997). Perceptions of suicide risk and the helpfulness of intervention strategies: A comparison of students and mothers. Suicide and Life-Threatening Behavior, 27(4), 362–372. [PubMed] [Google Scholar]
  51. Lawrence MT, & Ureda JR (1990). Student recognition of and response to suicidal peers. Suicide and Life-Threatening Behavior, 20(2), 164–176. [PubMed] [Google Scholar]
  52. Leong FTL & Austin JT (Eds.). (2016). The Psychology Research Handbook: A Guide for Graduate Students and Research Assistants (2nd Ed.). Sage Publications. [Google Scholar]
  53. Levitt JM, Saka N, Romanelli LH, & Hoagwood K (2007). Early identification of mental health problems in schools: The status of instrumentation. Journal of School Psychology, 45(2), 163–191. [Google Scholar]
  54. Lowry R, Crosby AE, Brener ND, & Kann L (2014). Suicidal thoughts and attempts among US high school students: Trends and associated health-risk behaviors, 1991–2011. Journal of Adolescent Health, 54(1), 100–108. [DOI] [PubMed] [Google Scholar]
  55. Michelmore L, & Hindley P (2012). Help-seeking for suicidal thoughts and self-harm in young people: A systematic review. Suicide and Life-Threatening Behavior, 42(5), 507–524. [DOI] [PubMed] [Google Scholar]
  56. Mitchell KJ, Turner HA, & Jones LM (2019). Youth exposure to suicide attempts: relative impact on personal trauma symptoms. American Journal of Preventive Medicine, 56(1), 109–115. [DOI] [PubMed] [Google Scholar]
  57. Mo PK, Ko TT, & Xin MQ (2018). School-based gatekeeper training programmes in enhancing gatekeepers’ cognitions and behaviours for adolescent suicide prevention: A systematic review. Child and Adolescent Psychiatry and Mental Health, 12(1), 1–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Mueller MA, & Waas GA (2002). College students’ perceptions of suicide: The role of empathy on attitudes, evaluation, and responsiveness. Death Studies, 26(4), 325–341. [DOI] [PubMed] [Google Scholar]
  59. Overholser JC, Hemstreet AH, Spirito A, & Vyse S (1989). Suicide awareness programs in the schools: Effects of gender and personal experience. Journal of the American Academy of Child & Adolescent Psychiatry, 28(6), 925–930. [DOI] [PubMed] [Google Scholar]
  60. Owen G, Belam J, Lambert H, Donovan J, Rapport F, & Owens C (2012). Suicide communication events: Lay interpretation of the communication of suicidal ideation and intent. Social Science & Medicine, 75(2), 419–428. [DOI] [PubMed] [Google Scholar]
  61. Owens C, Owen G, Lambert H, Donovan J, Belam J, Rapport F, & Lloyd K (2009). Public involvement in suicide prevention: Understanding and strengthening lay responses to distress. BioMed Central Public Health, 9(1), 308–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Randall JR, Nickel NC, & Colman I (2015). Contagion from peer suicidal behavior in a representative sample of American adolescents. Journal of Affective Disorders, 186(1), 219–225. [DOI] [PubMed] [Google Scholar]
  63. Rudd MD (2003). Warning signs for suicide? Suicide and Life-Threatening Behavior, 33(1), 99–100. [DOI] [PubMed] [Google Scholar]
  64. Rudd MD, Berman AL, Joiner TE Jr, Nock MK, Silverman MM, Mandrusiak M, van Orden K & Witte T (2006). Warning signs for suicide: Theory, research, and clinical applications. Suicide and Life-Threatening Behavior, 36(3), 255–262. [DOI] [PubMed] [Google Scholar]
  65. Schilling EA, Aseltine RH, & James A (2016). The SOS suicide prevention program: Further evidence of efficacy and effectiveness. Prevention Science, 17(2), 157–166. [DOI] [PubMed] [Google Scholar]
  66. Schilling EA, Lawless M, Buchanan L, & Aseltine RH (2014). “Signs of Suicide” shows promise as a middle school suicide prevention program. Suicide and Life-Threatening Behavior, 44(6), 653–667. [DOI] [PubMed] [Google Scholar]
  67. Suicide Prevention Resource Center (nd). SOS Signs of Suicide middle school and high school prevention programs. Archived NREPP registry https://nrepp.samhsa.gov/ProgramProfile.aspx?id=85
  68. Swanson SA, & Colman I (2013). Association between exposure to suicide and suicidality outcomes in youth. Canadian Medical Association Journal, 185(10), 870–877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Tamnes CK, Overbye K, Ferschmann L, Fjell AM, Walhovd KB, Blakemore SJ, & Dumontheil I (2018). Social perspective taking is associated with self-reported prosocial behavior and regional cortical thickness across adolescence. Developmental Psychology, 54(9), 1745–1757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Totura CMW, Labouliere CD, Gryglewicz K, & Karver MS (2019a). The role of youth trainee-trainer alliance and involvement in school-based prevention: A moderated-mediation model of student gatekeeper suicide prevention training. Administration and Policy in Mental Health, 46, 209–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Totura CMW, Labouliere CD, Gryglewicz K, & Karver MS (2019b). Adolescent decision-making: The value of perceived behavioral control in predicting engagement in suicide prevention behaviors. Journal of Youth and Adolescence, 48, 1784–1795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Totura CMW, Tarquini S, Naoom S, Karver MS, & Kutash K (2008). Planned Behavior & Implementation Questionnaire Unpublished Tampa, FL: University of South Florida. [Google Scholar]
  73. Tousignant B, Eugène F, & Jackson PL (2017). A developmental perspective on the neural bases of human empathy. Infant Behavior and Development, 48, 5–12. [DOI] [PubMed] [Google Scholar]
  74. van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, & Joiner TE Jr (2010). The interpersonal theory of suicide. Psychological Review, 117(2), 575–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Trout DL (1980). The role of social isolation in suicide. Suicide and Life‐Threatening Behavior, 10(1), 10–23. [DOI] [PubMed] [Google Scholar]
  76. Vetter NC, Altgassen M, Phillips L, Mahy CE, & Kliegel M (2013). Development of affective theory of mind across adolescence: disentangling the role of executive functions. Developmental Neuropsychology, 38(2), 114–125. [DOI] [PubMed] [Google Scholar]
  77. Winkelnkemper P (2014). Die Theorie des geplanten Verhaltens. Metaanalyse und Strukturgleichungsmodelle [The theory of planned behavior. Meta-analysis and structural equation models] (Unpublished diploma thesis). University of Giessen, Germany.
  78. Zhang J, & Li Z (2013). The association between depression and suicide when hopelessness is controlled for. Comprehensive Psychiatry, 54(7), 790–796. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES