Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: Depress Anxiety. 2020 Aug 12;38(1):48–56. doi: 10.1002/da.23087

Suicidal thoughts and behaviors in preadolescents: Findings and replication in two population-based samples

Rachel F L Walsh a, Ana E Sheehan b, Richard T Liu c,d
PMCID: PMC7785704  NIHMSID: NIHMS1643418  PMID: 32789968

Abstract

Introduction:

Given increasing concern in suicide in preadolescent children, this study aimed to characterize and identify potential indicators of risk for suicidal ideation (SI) and suicide attempts (SA) in this population.

Methods:

Data were drawn from two population-based samples of preadolescents: the 2007 and 2010 Minnesota Student Survey and analyses were restricted to 11- and 12-year-olds. Sociodemographic characteristics, childhood maltreatment, parental relations, peer relations, and school climate were examined in relation to past-year SI and SA. To examine correlates of SI, unconfounded by risk for SA, individuals with a history of SA were excluded from SI analyses. Correlates of SA were examined among individuals with past-year SI. Logistic regression analyses were conducted with past-year SI and SA as criterion variables.

Results:

Results from the 2007 and 2010 datasets were highly consistent. The prevalence of past-year SI was 9.28% and 9.25% in 2007 and 2010, respectively. Of the total sample, 1.90% and 1.87% reported a past-year SA (17.00% and 16.78% of those with past-year SI). Overall, effect sizes were generally modest to medium. The strongest effects were observed for sexual and physical abuse, parental support, and perceived safety at school (ps < 0.001). In multivariate analyses of SI and SA, sexual and physical abuse had the largest effect sizes (ORSI=2.18 [95% CI=1.90–2.51] – 2.96 [95% CI=2.69–3.26]; ORSA=1.55 [95% CI=1.29–1.86] – 2.26 [95% CI=1.82–2.80]).

Conclusions:

SI and SA occur at a concerning rate among preadolescent children. Screening for childhood sexual and physical abuse may be important for identifying those at risk for these clinical outcomes.

Keywords: suicide, suicidal ideation, early childhood

Introduction

Suicide is a major public health concern. Indeed, rates of suicide in the United States have risen 33% over the past two decades (Centers for Disease Control and Prevention, 2018). Child and adolescent suicide is of particular concern: from 2007 to 2017, suicide rates increased 56% among 10–24-year-olds (Centers for Disease Control and Prevention, 2019). In fact, suicide is the second leading cause of death among this demographic, second only to death by unintentional injury (Centers for Disease Control and Prevention, 2018). Considering these troubling statistics, a great deal of recent research has aimed to understand what may account for the high incidence of youth suicide. However, this research has almost exclusively focused on teenagers, with preadolescents being relatively neglected. This is particularly concerning because although preadolescent suicide occurs at a relatively low base rate, the prevalence in this age group has been increasing (Centers for Disease Control and Prevention, 2019). From 2007 to 2017, rates of suicide among 10–14-year-olds specifically have nearly tripled (Curtin & Heron, 2019). In response to this concerning development, the National Institute of Mental Health recently identified research on suicide in children as a priority and convened a meeting to guide future work in this area (National Institutes of Mental Health, 2019). It is therefore all the more necessary to characterize and identify potential indicators of risk for suicidal thoughts and behaviors in this age group to shape developmentally appropriate prevention and intervention efforts.

There is currently a paucity of studies on preadolescent suicide. Early research in this area s employed relatively small sample sizes (Pfeffer et al., 1984, 1988) and heavily relied on clinical samples (Pfeffer et al., 1989, 1998). Nevertheless, significant associations with psychiatric disorders, especially depression (Renaud et al., 2008; Shafii et al., 1988), a family history of psychopathology (Brent, 1995; Sarkar et al., 2010),and negative interpersonal factors such as physical and sexual abuse (Klimes-Dougan et al., 1999; Salzinger et al., 2007) have been observed. There is also evidence that preadolescents with conduct problems (e.g. shoplifting history (Shafii et al., 1985) and tendencies to engage in externalizing behaviors (e.g., aggression) are at increased risk for suicide (McGirr et al., 2008). Externalizing psychopathology may be particular to risk in preadolescents, with one recent study finding ADHD to be more common among preadolescents who died by suicide than among adolescent (Sheftall et al., 2016). Two recent studies utilized data from the Adolescent Brain Cognitive Development Study to examine risk and protective correlates of child suicidality, with an emphasis on familial influences. Internalizing and externalizing problems, family conflict, and low parental monitoring were all significantly associated with child suicide (DeVille et al., 2020; Janiri et al., 2020).

There are several significant methodological challenges, however, that have hindered overall progress in research in this area. Specifically, studies that do include preadolescents tend to combine them with adolescents (Gibbons et al., 2006; Goldstein et al., 2012). Given that suicidal thoughts and behaviors are more prevalent in teenagers than in children, it is likely the findings from studies of merged samples are largely reflective of adolescents and thus can speak little of preadolescents. Seeing as indicators of risk for suicide differ between adolescents and adults, and adolescents are developmentally different from preadolescents, it cannot be assumed that characteristics or processes applicable to older age groups are equally relevant to preadolescents (Goldston et al., 2009; Sheftall et al., 2016). As an example of developmental differences in the strength of associations between such characteristics and suicidal behavior, trait impulsivity has been associated with suicidal behavior in both adolescents and adults, but this association appears stronger in adolescents (Kasen et al., 2011; McGirr et al., 2008). Thus, merging preadolescents and adolescents makes it impossible to evaluate risk markers specifically in relation to the former age group.

Another limitation is that much of the prior literature does not cleanly differentiate between characteristics associated with suicidal ideation and with suicide attempts. That is, many studies assessed suicidal ideation and suicide attempts together as a single outcome. Other studies examining risk for suicidal ideation tend to include individuals with a history of suicide attempts in the samples. Additionally, studies of risk for suicide attempts tend to compare individuals with suicide attempts to all other individuals, which in community samples are predominantly of individuals without attempts or suicidal ideation. Consequently, it is often unclear whether any observed findings are reflective of suicidal ideation, suicide, attempts, or both. For example, since individuals who attempt suicide also generally have suicidal ideation, in the aforementioned community samples, it becomes unclear whether any observed association is a function of suicidal ideation or suicide attempts. This concern is not inconsequential, because characteristics of risk for suicidal ideation are not necessarily the same as those for suicide attempts (Joiner, 2005; Klonsky et al., 2016; May & Klonsky, 2016; Van Orden et al., 2010). To address this limitation, there is a need for studies of “pure” suicidal ideation, unconfounded by suicide attempts (i.e., suicidal ideation in individuals with no history of suicide attempts) and studies that assess risk for suicide attempts among the subset of ideators, thereby identifying markers of risk specific to suicide attempts (Klonsky et al., 2016; Liu et al., 2014; May & Klonsky, 2016). This would allow for greater understanding of factors that give rise to “pure” suicidal ideation, as well as factors that may contribute to the transition from suicidal ideation to suicidal behaviors. This latter consideration is important because although suicidal ideation is a strong predictor of suicide attempts, the vast majority of ideators never make an attempt (Klonsky & May, 2014; Ten Have et al., 2009).

Potentially in part accounting for this limitation, the low base rates of suicidal ideation and behavior pose a significant challenge to meaningful progress. Although suicide is a leading cause of death, it is still relatively uncommon. Thus, large samples are often needed for sufficiently powered analyses.

The current study aims to address these limitations by assessing indicators of risk for “pure” suicidal ideation and suicide attempts using two population-based samples of preadolescents: the 2007 and 2010 Minnesota Student Survey (MSS). The large sample size of the MSS make the current study uniquely positioned to address the aforementioned limitations. Furthermore, to evaluate the robustness of any observed associations, the planned analyses will be repeated in these independent samples. Based on prior research on associations with suicide in other populations, this study will specifically examine sociodemographic characteristics (Nock et al., 2008), childhood maltreatment (Klimes-Dougan et al., 1999; Maniglio, 2011; Norman et al., 2012), parental relations (Brent et al., 1994; DeVille et al., 2020), peer relations, (Pfeffer, Cynthia et al., 1993), and school climate (Young et al., 2011) in relation to suicidal ideation and behavior among preadolescents. This will facilitate the comparison of indicators of risk in different domains.

Methods

Study samples and procedures

The MSS is a population-based surveillance system administered every three years to public school students by the Minnesota Departments of Education, Health, Human Services, and Public Safety. Students were asked to answer questions anonymously regarding various health risk behaviors such as substance use and violence, and protective factors like community engagement and family dynamics. Anonymous inquiry has been shown to increase openness in adolescent reporting of sensitive behaviors (Turner et al., 1998). Data were drawn from the 2007 and 2010 surveys; preadolescent students were not asked about suicidal thoughts and behaviors in subsequent iterations. The study samples were restricted to 11- and 12-year-old students. All public school districts in Minnesota were invited to participate; the majority agreed: 309 (91%) in 2007 and 295 (88%) in 2010. Student participation rates were also high; 81% of 6th graders participated in 2007 and 79% in 2010. This study used publicly available secondary data and was exempt from institutional review board review. Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Measures

Suicide variables

The main outcome variables were past-year suicidal ideation and past-year suicide attempts. These were each measured by a single item: “Have you ever thought about killing yourself” and “Have you ever tried to kill yourself?” Response options were as follows: “no,” “yes, during the last year,” and “yes, more than a year ago.” This single-item approach is consistent with that of other large population-based surveys (i.e., Youth Risk Behavior Survey), and psychometric analyses (May & Klonsky, 2011) have supported it validity for assessing suicidality, especially in large population-based studies.

Sociodemographic variables

Participants were asked to report on their race and gender. As a proxy for socioeconomic status (SES), students were asked “Do you currently get free or reduced-price lunch at school?” Free or reduced-price lunch has been recognized as a valid measure of SES (Nicholson et al., 2014) and has been used extensively in prior research (Baams, 2018; Barnes et al., 2010).

Childhood maltreatment

Two types of childhood maltreatment were measured: physical abuse and sexual abuse. To evaluate lifetime physical abuse, students were asked “Has any adult in your household ever hit you so hard or so often that you had marks or were afraid of that person?” Lifetime sexual abuse was measured by combining two items: “Has any older person outside the family touched you sexually against your wishes or forced you to touch them sexually?” and “Has any older/stronger member of your family touched you sexually or had you touch them sexually?”

Parental relations

The quality of parent-child relationships was assessed by asking “How much do you feel your parents care about you?” Response options were on a five-point Likert scale from “not at all” to “very much.” For analyses, responses were reverse coded such that a higher score reflected worse parental relationships.

Peer relations

To assess the quality of students’ friendships, participants were asked “How much do you feel friends care about you?” Response options were initially coded on a five-point Likert scale, ranging from “not at all” to “very much” and then reverse coded for analyses. Negative peer relations were measured with three questions. Bullying within the past month was assessed by asking “During the last 30 days, how often has another student or group of students made fun of or teased you in a hurtful way, or excluded you from friends or activities?” Participant response options were on a five-point Likert scale, ranging from “never” to “every day.” Students were administered two additional items about being threatened and physical assault at school, respectively, in past 12 months with dichotomous response options.

School variables

To evaluate general attitudes towards school, participants were asked “How do you feel about going to school?” Responses were recorded on a five-point Likert scale, with response options ranging from “I like school very much” to “I hate school.” Students were also asked about feeling safe at school on a four-point Likert scale ranging from “strongly agree” to “strongly disagree.” Finally, students were asked whether they felt teachers at school cared about them on a five-point Likert scale, with response options ranging from “not at all” to “very much.” Responses to this item were reverse coded so higher scores reflected a worse relationship.

Data analysis

Data were analyzed in 2019. A series of univariate logistic regression analyses were conducted separately for 2007 and 2010 with past-year suicidal ideation and past-year suicide attempts as the criterion variables. In analyses of correlates of suicidal ideation, participants were excluded if they ever made a suicide attempt. This approach was utilized to validly examine correlates of “pure” suicidal ideation, unconfounded by correlates specific to suicide attempts (Walsh et al., 2018). Individuals who reported lifetime suicidal ideation but not past-year suicidal ideation, were also excluded to allow for greater temporal precision in evaluating correlates of suicidal ideation.

As in prior studies (Stewart et al., 2017), logistic regression analyses of suicide attempts were restricted to participants who reported past-year suicidal ideation to better understand factors that may contribute to the transition from ideation to attempt (Cheek et al., 2015). Participants were excluded from the suicide attempt analyses if they endorsed a suicide attempt in their lifetime but not within the past year.

Sociodemographic characteristics (i.e.,gender and socioeconomic status), family variables (i.e., parental support), childhood maltreatment (i.e. sexual abuse and physical abuse), school variables (i.e., teacher support, attitude towards school, perceived safety at school), and peer relations (perception of friends caring, bullying, threatening behaviors, and physical assault at school) were assessed individually as candidate predictor variables. Each predictor found to be significant at p < 0.05 was included in the corresponding final multivariate logistic regression model with suicidal ideation and suicide attempts, respectively, as the criterion variables.

Results

In total, there were 47,667 11- and 12-year-old participants in the 2007 dataset, and 45,450 in the 2010 dataset. After excluding individuals with lifetime (but not current) suicidal ideation and past-year suicide attempts, the 2007 and 2010 datasets for analyses of suicidal ideation included 42,149 and 40,359 preadolescents, respectively. Descriptive information regarding the study variables are presented in Table 1. In 2007, 3,911 (9.28%) preadolescents reported past-year suicidal ideation. Similarly, there were 3,735 (9.25%) individuals with past-year suicidal ideation in 2010. Correlations among all study variables are presented in Table 2. Correlation coefficients derived from the 2007 data were largely consistent with the values from the 2010 dataset.

Table 1.

Sample characteristics for 2007 and 2010 data.

2007
(n = 42,149)
 2010
(n = 40,359)
% or mean (SD) % or mean (SD)
Sociodemographic variables
 Gender (female) 50.4% 51.0%
 Free or reduced lunch 27.3% 30.6%
 Race (white) 75.7% 74.7%
Family variables
 Low parental support 1.12 (.49) 1.13 (.52)
Childhood maltreatment
 Lifetime sexual abuse 3.8% 3.7%
 Lifetime physical abuse 9.2% 9.8%
School variables
 Low teacher support 2.16 (1.06) 2.14 (1.07)
 Negative attitude towards attending school 2.38 (1.08) 2.38 (1.08)
 Low perceived safety at school 1.53 (.63) 1.52 (.63)
Peer relations variables
 Low peer social support 1.91 (.98) 1.88 (.97)
 Bullying victim 1.84(1.06) 1.81 (1.04)
 Threatened at school 22.5% 23.3%
 Physically assaulted at school 54.9% 53.1%
Open in a new tab

Table 2.

Correlations between study variables (n2007 = 42,149 and n2010 = 40,359)

Variable 1 2 3 4 5 6 7 8 9 10 11 12
1. Gender (female) – .02b −.01a .08c −.01a −.06c −.22c −.03c −.23c .02c −.13c −.18c
2. Free or reduced lunch .01 – .05c .06c .07c .02c .01 .07c .07c .02c .06c .04c
3. Low parental support <.01 .06c – .07c .21c .25c .13c .15c .20c .13c .11c .10c
4. Lifetime sexual abuse .06c .06c .09c – .16c .05c .02c .07c .02c .12c .12c .09c
5. Lifetime physical abuse <.01 .09c .23c .16c – .11c .07c .12c .10c .18c .18c .16c
6. Low teachers support −.05c .02c .25c .05c .11c – .31c .29c .39c .15c .14c .16c
7. Negative attitude towards attending school −.18c <.01 .13c .04c .08c .32c – .28c .22c .12c .13c .16c
8. Low perceived safety at school −.03c .07c .16c .07c .14c .29c .28c – .23c .24c .25c .21c
9. Low peer social support −.23c .06c .21c .02c .11c .39c .22c .24c – .22c .14c .17c
10. Bullying victim .01 .02c .12c .11c .18c .15c .12c .25c .24c – .36c .38c
11. Threatened at school −.13c .07c .11c .12c .19c .14c .13c .26c .15c .38c – .36c
12. Physically assaulted at school −.16c .05c .10c .10c .16c .16c .16c .23c .18c .40c .39c –
Open in a new tab

Note: Values above the diagonal line represent correlations from 2007 data. Values below the diagonal line are from the 2010 data.

a

p < 0.05

b

p < 0.01

c

p < 0.001

Results of univariate and multivariate logistic regression analyses for past-year suicidal ideation are presented in Table 3. Variables that reached significance (p < 0.05) at the univariate level were entered into the multivariate analyses. All variables emerged as significant (ps <0.001) in both the univariate and multivariate analyses. In the multivariate analyses, odds ratios ranged from 1.09 (95% CI = 1.05 – 1.13) to 2.96 (95% CI =2.69 – 3.26).

Table 3.

Univariate and multivariate predictors of past-year suicidal ideation

Univariate Multivariate
2007 2010 2007 2010
Predictor Odds Ratio (95% CI) Odds Ratio (95% CI) Odds Ratio (95% CI) Odds Ratio (95% CI)
Sociodemographic variables
 Gender (female) 1.11 (1.04–1.19)** 1.17 (1.09–1.25)*** 1.47 (1.34–1.58)*** 1.50 (1.38–1.63)***
 Free or reduced lunch 1.46 (1.36–1.57)*** 1.52 (1.42–1.63)*** 1.20 (1.11–1.31)*** 1.22 (1.13–1.33)***
Family variables
 Low parental support 2.07 (1.98–2.16)*** 2.25 (2.16–2.35)*** 1.41 (1.34–1.50)*** 1.56 (1.48–1.65)***
Childhood maltreatment
 Lifetime sexual abuse 4.70 (4.19–5.27)*** 4.91 (4.37–5.51)*** 2.18 (1.90–2.51)*** 2.23 (1.93–2.58)***
 Lifetime physical abuse 5.67 (5.24–6.14)*** 5.66 (5.22–6.13)*** 2.96 (2.69–3.26)*** 2.52 (2.28–2.79)***
School variables
 Low teachers support 1.51 (1.47–1.56)*** 1.57 (1.53–1.62)*** 1.09 (1.05–1.13)*** 1.11 (1.06–1.15)***
 Negative attitude towards 1.44 (1.40–1.48)*** 1.45 (1.40–1.49)***
 attending school 1.20 (1.16–1.25)*** 1.14 (1.10–1.19)***
 Low perceived safety at school 2.07 (1.97–2.17)*** 2.21 (2.11–2.32)*** 1.19 (1.12–1.26)*** 1.19 (1.12–1.27)***
Peer relations variables
 Low peer social support 1.50 (1.45–1.54)*** 1.59 (1.54–1.64)*** 1.14 (1.10–1.19)*** 1.13 (1.09–1.18)***
 Bullying victim 1.70 (1.65–1.74)*** 1.74 (1.70–1.79)*** 1.25 (1.21–1.30)*** 1.27 (1.22–1.31)***
 Threatened at school 3.24 (3.03–3.47)*** 3.60 (3.36–3.86)*** 1.47 (1.34–1.61)*** 1.64 (1.49–1.80)***
 Physically assaulted at school 3.02 (2.79–3.26)*** 3.23 (2.98–3.49)*** 1.56 (1.42–1.72)*** 1.51 (1.37–1.67)***
Open in a new tab

Note. n2007 = 42,149, n2010 = 40,359

*

p < 0.05

**

p < 0.01

***

p < 0.001

CI = confidence interval

To examine correlates of suicide attempts, analyses were restricted to all individuals with past-year suicidal ideation, with and without past-year suicide attempts, (n2007 = 4,712; n2010 = 4,488). Among past-year ideators, 801 (17.00%) and 753 (16.78%) individuals reported past-year suicide attempts in the 2007 and 2010 datasets, respectively. The prevalence of past-year suicide attempts among the entire sample was 1.90% and 1.87%, respectively.

Results of univariate and multivariate logistic regression analyses for past-year suicide attempts among past-year ideators are presented in Table 4. For both 2007 and 2010 all variables except for gender reached significance at the univariate level (ps < 0.001). All variables that were significant at the univariate level were entered into the multivariate analyses. In both the 2007 and 2010 multivariate analyses socioeconomic status, parental support, physical and sexual abuse, and perceived safety at school remained significant (ps < 0.01). In the 2007 model, but not the 2010 model, teacher support also remained significant (p < 0.05). Overall, most of the peer relationship variables (except for being threatened at school in the 2007 dataset) emerged as non-significant in the multivariate model, ps>0.05.

Table 4.

Univariate and multivariate predictors of past-year suicidal attempts

Univariate Multivariate
2007 2010 2007 2010
Predictor Odds Ratio (95% CI) Odds Ratio (95% CI) Odds Ratio (95% CI) Odds Ratio (95% CI)
Sociodemographic variables
 Gender (female) 1.04 (0.89–1.21) 1.05 (0.89–1.22) -- --
 Free or reduced lunch 1.58 (1.35–1.86)*** 1.51 (1.28–1.77)*** 1.43 (1.19–1.70)*** 1.42 (1.19–1.70)***
Family variables
 Low parental support 1.45 (1.36–1.55)*** 1.31 (1.23–1.40)*** 1.24 (1.14–1.35)*** 1.16 (1.06–1.26)***
Childhood maltreatment
 Lifetime sexual abuse 2.56 (2.13–3.09)*** 2.87 (2.38–3.46)*** 1.96 (1.58–2.43)*** 2.26 (1.82–2.80)***
 Lifetime physical abuse 2.12 (1.81–2.48)*** 2.30 (1.96–2.70)*** 1.55 (1.29–1.86)*** 1.79 (1.48–2.17)***
School variables
 Low teachers support 1.34 (1.26–1.43)*** 1.20 (1.13–1.28)*** 1.12 (1.03–1.21)** 1.02 (0.94–1.11)
 Negative attitude towards attending school 1.22 (1.14–1.29)*** 1.18 (1.11–1.26)*** 1.06 (0.98–1.14) 1.06 (0.98–1.15)
 Low perceived safety at school 1.61 (1.47–1.78)*** 1.47 (1.34–1.62)*** 1.24 (1.10–1.40)*** 1.27 (1.12–1.43)***
Peer relations variables
 Low peer social support 1.13 (1.06–1.21)*** 1.10 (1.03–1.17)** 0.95 (0.87–1.02) 0.99 (0.91–1.08)
 Bullying victim 1.22 (1.15–1.29)*** 1.15 (1.08–1.21)*** 1.07 (1.00–1.15) 0.99 (0.91–1.06)
 Threatened at school 1.83 (1.56–2.14)*** 1.57 (1.33–1.84)*** 1.29 (1.05–1.58)* 1.22 (0.99–1.50)
 Physically assaulted at school 1.33 (1.09–1.61)** 1.46 (1.19–1.80)*** 0.87 (0.68–1.10) 1.06 (0.82–1.36)
Open in a new tab

Note. n2007 = 4,712, n2010 = 4,488

*

p < 0.05

**

p < 0.01

***

p < 0.001

CI = confidence interval

Discussion

Given the rise in suicidal thoughts and behaviors among preadolescents (Centers for Disease Control, 2018), it is essential to characterize at-risk individuals to inform intervention and prevention efforts for this population. To address this need, the current study employed two large population-based samples of preadolescents to assess indicators of risk for suicidal ideation and suicide attempts in four domains: family, childhood maltreatment, school, and peer relations. In the analyses of suicidal ideation, unconfounded with suicide attempts, all variables emerged as significant in the multivariate models. For multivariate models examining suicide attempts among children with past-year ideation, only demographic characteristics, family relations, childhood maltreatment, and several school-related variables remained significant. Several of these findings warrant discussion.

Notably, results were highly consistent from 2007 to 2010. Almost all variables that emerged as significant in one year remained significant in the other year. Effect sizes were largely comparable, and, occasionally virtually identical. This is likely because both sample sizes were very large and highly representative of the same population, which reduces statistical error due to sampling. The observed consistency between years suggests the findings are robust and stable, thereby increasing confidence in their interpretation. Given the recent emphasis on reproducibility in science (Maxwell et al., 2015), this is a particular strength of this study.

In analyses of suicidal ideation, effect sizes ranged from ORs of 1.09 to 2.96. Generally, factors related to gender, physical safety/violence, and parental support exhibited the largest effect sizes (OR=1.41–2.96). The smallest effects were observed among variables related to school and peer relations. Among the three types of social support assessed (i.e., parent, teacher, and peer), parental support appeared to have the largest effect, which is consistent with the view that parental relationships have a larger role in preadolescence (Steinberg & Silverberg, 1986). That is, children spend more time interacting with family than peers, often relying on parents for emotional support (Ryan & Lynch, 1989). This differs from adolescence, which is characterized by a transition to greater autonomy from their parents and increasing affiliation with same-aged peers (Choudhury et al., 2006; Wentzel, 1998), making them more susceptible to peer influence. This finding builds on the existing literature; whereas past studies have found parental (DeVille et al., 2020) and peer influences (Pfeffer et al., 1993), respectively to predict risk for suicidal outcomes among preadolescents, to our knowledge ours is the first to evaluate the relative effects of parental and peer influences. This finding is also consistent with prominent theories of suicide, which hold that social disconnectedness with important figures within the social network may feature prominently in risk for suicidal ideation (Van Orden et al., 2010). When the most salient social relationships for children this age (i.e., parental relationships) are compromised feelings of social isolation may manifest, thereby increasing risk of suicidal ideation.

That factors associated with physical safety/violence (i.e., sexual and physical abuse, and threat of or being the victim of assault) are linked to suicidal ideation may initially seem to be inconsistent with theoretical perspectives of suicide (Klonsky et al., 2016; Van Orden et al., 2010) that associate physically painful experiences and ones involving threats to physical safety more with suicidal behavior than ideation. However, given that these three experiences all occur within the interpersonal context, the social disconnectedness inherent in these experiences may be driving their association with suicidal ideation.

In analyses of suicide attempts, similar patterns emerged, except, unlike in the case of suicidal ideation, peer relations and school variables were largely not significant. Gender is the only variable that did not reach significance at the univariate level. This is in line with the developmental course of suicidal behavior and other forms of psychopathology (e.g., depression), for which sex differences do not emerge until adolescence (Hankin et al., 1998; Nock et al., 2008). In contrast, experiences associated with more severe forms of violence (i.e., sexual and physical abuse) appeared to be the main predictors of suicide attempts. It is important to note, however, that these analyses were aimed at predicting suicide attempts among preadolescents with active suicidal ideation (i.e., differentiating between suicidal ideation and attempts). Thus, sexual and physical abuse may be particularly potent predictors when considering the stringency of these analyses.

Several limitations must be considered when interpreting these findings. First, the data are cross-sectional, and thus preclude inferences of directionality. However, given the relative want of research on preadolescent suicide, the current findings provide a meaningful foundation for future work in this area, offering candidates for longitudinal evaluations of potential risk factors (Faraone et al., 2000; Kraemer, 1997). Additionally, although findings were replicated in a second independent sample, the two samples were drawn from the same population (Minnesota residents). It will be important to replicate these findings in other populations to determine the extent of their generalizability. Lastly, the indicators of risk investigated in this study are not comprehensive and it will be important to examine other factors such as early-onset puberty, irritability, and symptoms of psychopathology.

These findings have several clinical implications. First, it is important to note that overall, the observed effect sizes were relatively modest to medium. This supports the notion that suicidal thoughts and behaviors are multi-determined, that there are numerous factors that give rise to their occurrence (Brent & Mann, 2005; Smith et al., 2012). These findings thus underscore the challenge of predicting risk for and preventing the occurrence of suicidal thoughts and behaviors, particularly the need simultaneously to consider multiple indices of risk. Second, the current findings suggest that sexual and physical abuse may be especially important in screening for suicidal ideation and attempts. These findings, however, also highlight the difficulty of assessing suicide risk, as many individuals may be uncomfortable disclosing such personally sensitive and traumatic experiences as childhood abuse (Lemaigre et al., 2017). Finally, the challenges indicated by the current findings support the need for additional research identifying predictors of suicidal thoughts and behaviors in preadolescents. Individuals who experience severe mental health difficulties at a young age tend to have poor prognosis (Jaffee et al., 2002; Korczak & Goldstein, 2009; Weissman et al., 1999). Thus, identifying additional risk indicators for early targeted prevention efforts with preadolescents could have a substantial impact in interrupting the increasingly severe course often observed with suicidal behavior (Goldston et al., 2015).

Acknowledgments:

This study was supported in part by National Science Foundation Graduate Research Fellowships to Rachel Walsh and Ana Sheehan, and the National Institute of Mental Health of the National Institutes of Health under awards RF1 MH120830, R01MH101138, R01MH115905, and R21MH112055 to Richard Liu.

Footnotes

Conflict of Interest: None

Data availability statement:

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

References

  1. Baams L (2018). Disparities for LGBTQ and gender nonconforming adolescents. Pediatrics, 141 10.1542/peds.2017-3004 [DOI] [PubMed] [Google Scholar]
  2. Barnes AJ, Eisenberg ME, & Resnick MD (2010). Suicide and self-injury among children and youth with chronic health conditions. Pediatrics, 125, 889–895. 10.1542/peds.2009-1814 [DOI] [PubMed] [Google Scholar]
  3. Brent DA (1995). Risk Factors for Adolescent Suicide and Suicidal Behavior: Mental and Substance Abuse Disorders, Family Environmental Factors, and Life Stress. Suicide and Life‐Threatening Behavior, 25, 52–63. 10.1111/j.1943-278X.1995.tb00490.x [DOI] [PubMed] [Google Scholar]
  4. Brent DA, & Mann JJ (2005). Family genetic studies, suicide, and suicidal behavior. American Journal of Medical Genetics - Seminars in Medical Genetics, 133 C, 13–24. 10.1002/ajmg.c.30042 [DOI] [PubMed] [Google Scholar]
  5. Brent DA, Perper JA, Moritz G, Liotus L, Schweers J, Balach L, & Roth C (1994). Familial risk factors for adolescent suicide: a case‐control study. Acta Psychiatrica Scandinavica, 89, 52–58. 10.1111/j.1600-0447.1994.tb01485.x [DOI] [PubMed] [Google Scholar]
  6. Centers for Disease Control and Prevention (CDC). (2018). Suicide Mortality in the United States, 1999 – 2017. NCHS Data Brief. https://www.cdc.gov/nchs/data/databriefs/db330-h.pdf [PubMed] [Google Scholar]
  7. Centers for Disease Control and Prevention (CDC). (2019). Death Rates Due to Suicide and Homicide Among Persons Aged 10–24: United States, 2000–2017. NCHS Data Brief. https://www.cdc.gov/nchs/data/databriefs/db352-h.pdf [PubMed] [Google Scholar]
  8. Cheek SM, Nestor BA, & Liu RT (2015). Substance use and suicidality: Specificity of substance use by injection to suicide attempts in a nationally representative sample of adults with major depression. Depression and Anxiety, 33, 541–548. 10.1002/da.22407 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Choudhury S, Blakemore SJ, & Charman T (2006). Social cognitive development during adolescence. Social Cognitive and Affective Neuroscience. 10.1093/scan/nsl024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Curtin SC, & Heron M (2019). Death Rates Due to Suicide and Homicide Among Persons Aged 10–24: United States, 2000–2017. NCHS Data Brief, 352, 1–8. [PubMed] [Google Scholar]
  11. DeVille DC, Whalen D, Breslin FJ, Morris AS, Khalsa SS, Paulus MP, & Barch DM (2020). Prevalence and Family-Related Factors Associated With Suicidal Ideation, Suicide Attempts, and Self-injury in Children Aged 9 to 10 Years. JAMA Network Open, 3 10.1001/jamanetworkopen.2019.20956 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Faraone SV, Biederman J, Mick E, Williamson S, Wilens T, Spencer T, Weber W, Jetton J, Kraus I, Pert J, & Zallen B (2000). Family study of girls with attention deficit hyperactivity disorder. American Journal of Psychiatry, 157, 1077–1083. 10.1176/appi.ajp.157.7.1077 [DOI] [PubMed] [Google Scholar]
  13. Gibbons RD, Hur K, Bhaumik DK, & Mann JJ (2006). The relationship between antidepressant prescription rates and rate of early adolescent suicide. American Journal of Psychiatry, 163, 1898–1904. 10.1176/ajp.2006.163.11.1898 [DOI] [PubMed] [Google Scholar]
  14. Goldstein TR, Ha W, Axelson DA, Goldstein BI, Liao F, Gill MK, Ryan ND, Yen S, Hunt J, Hower H, Keller M, Strober M, & Birmaher B (2012). Predictors of prospectively examined suicide attempts among youth with bipolar disorder. Archives of General Psychiatry, 69, 1113–1122. 10.1001/archgenpsychiatry.2012.650 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Goldston DB, Daniel SS, Erkanli A, Heilbron N, Doyle O, Weller B, Sapyta J, Mayfield A, & Faulkner M (2015). Suicide attempts in a longitudinal sample of adolescents followed through adulthood: Evidence of escalation. Journal of Consulting and Clinical Psychology, 83, 253–264. 10.1037/a0038657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Goldston DB, Daniel SS, Erkanli A, Reboussin BA, Mayfield A, Frazier PH, & Treadway SL (2009). Psychiatric Diagnoses as Contemporaneous Risk Factors for Suicide Attempts Among Adolescents and Young Adults: Developmental Changes. Journal of Consulting and Clinical Psychology, 77, 281–290. 10.1037/a0014732 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hankin BL, Abramson LY, Moffitt TE, Angell KE, Silva PA, & McGee R (1998). Development of depression from preadolescence to young adulthood: Emerging gender differences in a 10-year longitudinal study. Journal of Abnormal Psychology, 107, 128–140. 10.1037/0021-843X.107.1.128 [DOI] [PubMed] [Google Scholar]
  18. Jaffee SR, Moffitt TE, Caspi A, Fombonne E, Poulton R, & Martin J (2002). Differences in early childhood risk factors for juvenile-onset and adult-onset depression. Archives of General Psychiatry, 59, 215 10.1001/archpsyc.59.3.215 [DOI] [PubMed] [Google Scholar]
  19. Janiri D, Doucet GE, Pompili M, Sani G, Luna B, Brent DA, & Frangou S (2020). Risk and protective factors for childhood suicidality: a US population-based study. The Lancet Psychiatry, 7, 317–326. 10.1016/S2215-0366(20)30049-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Joiner T (2005). Why people die by suicide. In Harvard University Press, Cambridge, MA. [Google Scholar]
  21. Kasen S, Cohen P, & Chen H (2011). Developmental course of impulsivity and capability from age 10 to age 25 as related to trajectory of suicide attempt in a community cohort. Suicide and Life-Threatening Behavior, 41, 180–192. 10.1111/j.1943-278X.2011.00017.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Klimes-Dougan B, Free K, Ronsaville D, Stilwell J, Jean Welsh C, & Radke-Yarrow M (1999). Suicidal ideation and attempts: A longitudinal investigation of children of depressed and well mothers. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 651–659. 10.1097/00004583-199906000-00011 [DOI] [PubMed] [Google Scholar]
  23. Klonsky ED, & May AM (2014). Differentiating suicide attempters from suicide ideators: A critical frontier for suicidology research. Suicide and Life-Threatening Behavior, 44, 1–5. 10.1111/sltb.12068 [DOI] [PubMed] [Google Scholar]
  24. Klonsky ED, May AM, & Saffer BY (2016). Suicide, Suicide Attempts, and Suicidal Ideation. Annual Review of Clinical Psychology, 12, 307–330. 10.1146/annurev-clinpsy-021815-093204 [DOI] [PubMed] [Google Scholar]
  25. Korczak DJ, & Goldstein BI (2009). Childhood Onset Major Depressive Disorder: Course of Illness and Psychiatric Comorbidity in a Community Sample. Journal of Pediatrics, 155, 118–123. 10.1016/j.jpeds.2009.01.061 [DOI] [PubMed] [Google Scholar]
  26. Kraemer HC (1997). Coming to Terms With the Terms of Risk. Archives of General Psychiatry, 54, 337 10.1001/archpsyc.1997.01830160065009 [DOI] [PubMed] [Google Scholar]
  27. Lemaigre C, Taylor EP, & Gittoes C (2017). Barriers and facilitators to disclosing sexual abuse in childhood and adolescence: A systematic review. Child Abuse and Neglect, 70, 39–52. 10.1016/j.chiabu.2017.05.009 [DOI] [PubMed] [Google Scholar]
  28. Liu RT, Case BG, & Spirito A (2014). Injection drug use is associated with suicide attempts but not ideation or plans in a sample of adolescents with depressive symptoms. Journal of Psychiatric Research, 56, 65–71. 10.1016/j.jpsychires.2014.05.001 [DOI] [PubMed] [Google Scholar]
  29. Maniglio R (2011). The role of child sexual abuse in the etiology of suicide and non-suicidal self-injury. Acta Psychiatrica Scandinavica, 124, 30–41. 10.1111/j.1600-0447.2010.01612.x [DOI] [PubMed] [Google Scholar]
  30. Maxwell SE, Lau MY, & Howard GS (2015). Is psychology suffering from a replication crisis?: What does “failure to replicate” really mean? American Psychologist, 70, 487–498. 10.1037/a0039400 [DOI] [PubMed] [Google Scholar]
  31. May A, & Klonsky ED (2011). Validity of suicidality items from the youth risk behavior survey in a high school sample. Assessment, 18, 379–381. 10.1177/1073191110374285 [DOI] [PubMed] [Google Scholar]
  32. May A, & Klonsky ED (2016). What Distinguishes Suicide Attempters From Suicide Ideators? A Meta-Analysis of Potential Factors. Clinical Psychology: Science and Practice, 23, 5–20. 10.1111/cpsp.12136 [DOI] [Google Scholar]
  33. McGirr A, Renaud J, Bureau A, Seguin M, Lesage A, & Turecki G (2008). Impulsive-aggressive behaviours and completed suicide across the life cycle: A predisposition for younger age of suicide. Psychological Medicine, 38, 407–417. 10.1017/S0033291707001419 [DOI] [PubMed] [Google Scholar]
  34. National Institutes of Mental Health. (2019). Identifying Research Priorities in Child Suicide Risk. https://www.nimh.nih.gov/news/events/announcements/identifying-research-priorities-in-child-suicide-risk.shtml
  35. Nicholson LM, Slater SJ, Chriqui JF, & Chaloupka F (2014). Validating Adolescent Socioeconomic Status: Comparing School Free or Reduced Price Lunch with Community Measures. Spatial Demography, 2, 55–65. 10.1007/bf03354904 [DOI] [Google Scholar]
  36. Nock MK, Borges G, Bromet EJ, Cha CB, Kessler RC, & Lee S (2008). Suicide and suicidal behavior. Epidemiologic Reviews, 30, 133–154. 10.1093/epirev/mxn002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Norman RE, Byambaa M, De R, Butchart A, Scott J, & Vos T (2012). The Long-Term Health Consequences of Child Physical Abuse, Emotional Abuse, and Neglect: A Systematic Review and Meta-Analysis. PLoS Medicine, 9 10.1371/journal.pmed.1001349 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pfeffer Cynthia, R., Klerman G,L, Hurt Stephen, W., Kakuma, T, Peskin JR, & Siefker CA (1993). Suicidal Children Grow Up: Rates and Psychosocial Risk Factors for Suicide Attempts during Follow-up. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 106–113. 10.1097/00004583-199301000-00016 [DOI] [PubMed] [Google Scholar]
  39. Pfeffer CR, McBride PA, Anderson GM, Kakuma T, Fensterheim L, & Khait V (1998). Peripheral serotonin measures in prepubertal psychiatric inpatients and normal children: Associations with suicidal behavior and its risk factors. Biological Psychiatry, 44, 568–577. 10.1016/S0006-3223(98)00020-1 [DOI] [PubMed] [Google Scholar]
  40. Pfeffer CR, Plutchik R, & Mizruchi M (1988). Normal Children at Risk for Suicidal Behavior: A Two-Year Follow-up Study. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 34–41. 10.1097/00004583-198801000-00006 [DOI] [PubMed] [Google Scholar]
  41. Pfeffer CR, Plutchik R, Mizruchi MS, Faughnan L, Mintz M, & & Shindledecker R (1989). Changes in suicidal behavior in child psychiatric inpatients. Acta Psychiatrica Scandinavica, 79, 431–435. [DOI] [PubMed] [Google Scholar]
  42. Pfeffer CR, Zuckerman S, Plutchik R, & Mizruchi MS (1984). Suicidal Behavior in Normal School Children: A Comparison with Child Psychiatric Inpatients. Journal of the American Academy of Child Psychiatry, 23, 416–423. 10.1016/S0002-7138(09)60319-8 [DOI] [PubMed] [Google Scholar]
  43. Renaud J, Berlim MT, McGirr A, Tousignant M, & Turecki G (2008). Current psychiatric morbidity, aggression/impulsivity, and personality dimensions in child and adolescent suicide: A case-control study. Journal of Affective Disorders, 15, 381–407. 10.1016/j.jad.2007.05.013 [DOI] [PubMed] [Google Scholar]
  44. Ryan RM, & Lynch JH (1989). Emotional autonomy versus detachment: revisiting the vicissitudes of adolescence and young adulthood. Child Development, 60, 340–356. 10.1111/j.1467-8624.1989.tb02720.x [DOI] [PubMed] [Google Scholar]
  45. Salzinger S, Rosario M, Feldman RS, & Ng-Mak DS (2007). Adolescent suicidal behavior: Associations with preadolescent physical abuse and selected risk and protective factors. Journal of the American Academy of Child and Adolescent Psychiatry, 46, 859–866. 10.1097/chi.0b013e318054e702 [DOI] [PubMed] [Google Scholar]
  46. Sarkar M, Byrne P, Power L, Fitzpatrick C, Anglim M, Boylan C, & Morgan S (2010). Are Suicidal Phenomena in Children Different to Suicidal Phenomena in Adolescents? A Six-Year Review. Child and Adolescent Mental Health, 15, 197–203. 10.1111/j.1475-3588.2010.00567.x [DOI] [PubMed] [Google Scholar]
  47. Shafii M, Carrigan S, Whittinghill JR, & Derrick A (1985). Psychological autopsy of completed suicide in children and adolescents. American Journal of Psychiatry, 142, 1061–1064. 10.1176/ajp.142.9.1061 [DOI] [PubMed] [Google Scholar]
  48. Shafii M, Steltz-Lenarsky J, Derrick AMC, Beckner C, & Whittinghill JR (1988). Comorbidity of mental disorders in the post-mortem diagnosis of completed suicide in children and adolescents. Journal of Affective Disorders, 15, 227–233. 10.1016/0165-0327(88)90020-1 [DOI] [PubMed] [Google Scholar]
  49. Sheftall AH, Asti L, Horowitz LM, Felts A, Fontanella CA, Campo JV, & Bridge JA (2016). Suicide in elementary school-aged children and early adolescents. Pediatrics, 138 10.1542/peds.2016-0436 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Smith AR, Ribeiro JD, Mikolajewski A, Taylor J, Joiner TE, & Iacono WG (2012). An examination of environmental and genetic contributions to the determinants of suicidal behavior among male twins. Psychiatry Research, 197, 60–65. 10.1016/j.psychres.2012.01.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Steinberg L, & Silverberg SB (1986). The vicissitudes of autonomy in early adolescence. Child Development, 57, 841–851. 10.1111/j.1467-8624.1986.tb00250.x [DOI] [PubMed] [Google Scholar]
  52. Stewart JG, Glenn CR, Esposito EC, Cha CB, Nock MK, & Auerbach RP (2017). Cognitive control deficits differentiate adolescent suicide ideators from attempters. Journal of Clinical Psychiatry, 78 10.4088/JCP.16m10647 [DOI] [PubMed] [Google Scholar]
  53. Ten Have M, De Graaf R, Van Dorsselaer S, Verdurmen J, Van’t Land H, Vollebergh W, & Beekman A (2009). Incidence and course of suicidal ideation and suicide attempts in the general population. Canadian Journal of Psychiatry, 54, 824–833. [DOI] [PubMed] [Google Scholar]
  54. Turner CF, Ku L, Rogers SM, Lindberg LD, Pleck JH, & Sonenstein FL (1998). Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science, 280, 867–873. 10.1126/science.280.5365.867 [DOI] [PubMed] [Google Scholar]
  55. Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, & Joiner TE (2010). The Interpersonal Theory of Suicide. Psychological Review, 117, 575–600. 10.1037/a0018697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Walsh RFL, Sheehan AE, & Liu RT (2018). Prospective prediction of first lifetime onset of suicidal ideation in a national study of substance users. Journal of Psychiatric Research, 107 10.1016/j.jpsychires.2018.09.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Weissman MM, Wolk S, Wickramaratne P, Goldstein RB, Adams P, Greenwald S, Ryan ND, Dahl RE, & Steinberg D (1999). Children with prepubertal-onset major depressive disorder and anxiety grown up. Archives of General Psychiatry, 56, 794–801. 10.1001/archpsyc.56.9.794 [DOI] [PubMed] [Google Scholar]
  58. Wentzel KR (1998). Social relationships and motivation in middle school: The role of parents, teachers, and peers. Journal of Educational Psychology, 90, 202–209. 10.1037/0022-0663.90.2.202 [DOI] [Google Scholar]
  59. Young R, Sweeting H, & Ellaway A (2011). Do schools differ in suicide risk? the influence of school and neighbourhood on attempted suicide, suicidal ideation and self-harm among secondary school pupils. BMC Public Health, 11 10.1186/1471-2458-11-874 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

RESOURCES