Proximal Risk of Suicide-Related Behavior in Black Youth With a Parental History of Mood Disorder

Abstract

Background:

Suicide rates have risen disproportionately in Black youth, yet little is known about risk factors specific to this population.

Aims:

We sought to identify proximal risk factors for suicide-related behavior (SRB) in a sample of Black youth at high risk for mood disorders.

Methods:

The sample, drawn from a longitudinal study, included 156 Black youth aged 17 years or younger at baseline. Participants underwent baseline diagnostic and clinical assessment, including suicidal ideation and SRB, and yearly assessments for new SRB. Observations were censored at the time point closest to the first episode of SRB or the last observation for those without SRB. Chi-square, Fisher’s exact test, t-tests, and Firth logistic regression were used.

Results:

Twelve participants experienced SRB during follow-up. The average age of participants at baseline was 12.6 years (SD = 2.7). Participants were 44% female and 10% identified as Hispanic. Self-reported irritability, aggression, suicidal ideation, and a depression diagnosis were significant at the univariate level. Medium to large effect sizes were found for self-reported impulsive aggression, parental substance use disorder, and childhood abuse, all consistent with prior reports to be associated with SRB in youth. Using Firth logistic regression, younger age (β = –.60; p = .001) and depression diagnosis (β = 2.32, p = .006) were proximal risk indicators of SRB.

Conclusions:

Risk factors identified in this study are consistent with previous reports in majority White samples. A depression diagnosis was the strongest indicator, highlighting the importance of early identification of depression in Black youth, particularly in younger youth with a family history of mood disorder.

Keywords

suicide and self-destructive behavior adolescents adolescence minority populations

Suicide was the second leading cause of death in youth ages 10 to 14 years. and the third leading cause of death in youth ages 14 to 19 years in 2024 (Centers for Disease Control and Prevention [CDC], 2025). Suicide rates in youth increased by 52.2% since 2000 (CDC, 2023). In recent years, suicide rates have risen disproportionally in Black youth, especially in preteen children. Bridge et al. (2018) examined United States mortality data from 2001–2015 and found that Black youth ages 5 to 12 years were twice as likely to die by suicide compared to White youth in the same age range, with reversed trends in older adolescents (Bridge et al., 2018; Ruch et al., 2024). In addition, there has been an upward trend in suicide rates among Black youth across all age groups, as well as an increase in the number of girls who died by suicide, most notably in 12 to 14-year-olds (Sheftall et al., 2022). The suicide rate for Black males aged 15 to 24 years increased by 47% from 2013 to 2019 (Ramchand et al., 2021). Because suicide rates in Black youth have historically been much lower than rates in White youth, research into risk and protective factors for Black youth has been minimal to date.

The rates of suicidal ideation and suicide attempts among Black youth, based on the National Youth Risk Behavior Survey data, show an increase from 2007 to 2021 in those seriously considering suicide (13.2% vs. 21.6%), making a suicide plan (9.5% vs. 17.7%), suicide attempts (7.7% vs. 14.5%), and attempt with injury requiring treatment (2.3% vs. 4.4%) (CDC, 2024). The objective of this study was to identify proximal risk factors for suicide-related behavior (SRB) in a sample of Black youth at high risk for mood disorder by virtue of a parental history of mood disorder. We have previously examined risk factors in the offspring of parents with a history of suicide attempt compared to parents with no history of suicide attempt in the full sample (Brent et al., 2002; 2003, 2004, 2015; Melhem et al., 2007, 2019; J. Zelazny et al., 2021; J. Zelazny et al., 2019). However, this is the first analysis in this sample of risk factors specific to Black youth. The aim of this secondary data analysis was to explore risk factors specific to the Black youth in this study. We hypothesized that we would find similar proximal risk indicators for SRB in Black youth of parents with mood disorders to those reported in our prior studies for the full sample, with regards to contributions of family history and parental attempt status as well as offspring mood disorder, impulsive aggression, and history of childhood abuse. However, we are aware that findings might diverge in relation to risk factors more commonly found in minoritized populations.

Method

Design

This is a secondary data analysis of the Familial Pathways to Early Onset Suicide Attempt Study (FamPath), a longitudinal study that enrolled 334 parents with mood disorders and their offspring (N = 711) across two sites—the University of Pittsburgh and New York State Psychiatric Institute (NYSPI) (Brent et al., 2002, 2015).

Sample

The study was approved by the institutional review boards at both the University of Pittsburgh and NYSPI. As part of the larger study, parents and offspring were followed over a 15-year period from 1997 to 2012. The primary aim of the larger FamPath study was to determine whether offspring of parents with mood disorders who also attempted suicide were more likely to attempt suicide than offspring of parents with mood disorders who did not attempt suicide and to explore differences in other characteristics between the two offspring groups. Parents were primarily recruited from clinical settings, including inpatient units and outpatient clinics. Full description of recruitment methods and inclusion/exclusion criteria was previously described for the parent study (Brent et al., 2002). Parents were eligible to participate in the study if they met DSM-IV criteria for a unipolar or bipolar mood disorder and had at least one offspring age 10 or older who was willing and able to participate in the study. Therefore, this sample of Black youth was specifically recruited due to parental history of affective disorders, including about half with a parental history of suicide attempt.

Data Collection

Procedures

Parents and their offspring aged 10 years and older underwent baseline and yearly follow-up diagnostic and clinical assessments. Baseline interviews were conducted in-person as home visits or clinic visits (participant choice). Follow-up interviews alternated between phone and in-person every other year. Parents and offspring were compensated for their participation. Clinical interviewers were psychiatric nurses or master’s or PhD-level clinicians. The primary outcome for this analysis was suicide-related behavior (SRB), defined as a suicide attempt, interrupted attempt, aborted attempt, or severe suicidal ideation resulting in an emergency referral occurring during follow-up, as determined by interview with the Columbia Suicide History Form Follow-up version (Mann et al., 1992).

Clinician-Administered Interview Measures

At baseline, demographic information, including site, age, race, ethnicity, sex, and socioeconomic status were collected. Youth were assessed using the Schedule for Affective Disorders and Schizophrenia for School-Age Children—Present and Lifetime Version (K-SADS-PL) using DSM-IV criteria (American Psychiatric Association, 1994) as the parent study was launched prior to the release of DSM 5. The K-SADS-PL has been shown to generate reliable and valid psychiatric diagnoses (Kaufman et al., 1997). At each follow-up, data were collected about the presence of psychiatric disorders since the last study assessment, including onset and offset dates and the number of episodes. SRB data were collected at each time point using the Columbia Suicide History (Mann et al., 1992), which incorporates the definitions used in the Columbia Suicide Reclassification project for actual, interrupted, and aborted attempts with good reliability and validity (Posner et al., 2007). Aggression was measured with the intake and follow-up versions of the Brown-Goodwin Lifetime History of Aggression interview (Brown & Goodwin, 1986). Consensus meetings were held weekly to assess reliability and agreement between assessors on psychiatric diagnoses, aggression, and suicidal behavior classification.

Self-Reported Measures

All self-report measures were chosen based on their reliability and validity for youth between 10 and 17 (Brent et al., 2002). Depressive symptoms, hopelessness, and suicidal ideation were measured in youth aged 14 and older using the Beck Depression Inventory (Beck et al., 1961), the Beck Hopelessness Scale (Beck et al., 1974), and the Beck Scale of Suicidal Ideation (Beck et al., 1979), and in youth 10 to 13 years of age using the Child Depression Inventory (Kovacs, 1985), Hopelessness Scale for Children (Kazdin et al., 1986), and the Child Scale for Suicidal Ideation (Kazdin et al., 1983). Impulsive aggression was assessed using the Buss-Durkee Hostility Inventory (Buss & Durkee, 1957) for youth 14 and older and the Children’s Hostility Inventory (Kazdin et al., 1987) for youth 10 to 13. Impulsivity was measured using the Emotionality, Activity, Sociability and Impulsivity scale (Plomin, 1983). History of exposure to childhood abuse was measured using the Childhood Trauma Questionnaire (Bernstein et al., 1994) and an adapted instrument from the Abuse Dimensions Inventory (Chaffin et al., 1997).

Family Measures

Parental diagnoses were ascertained using the Structured Clinical Interview for DSM-IV (Spitzer et al., 1992), and parental history of suicidal behavior was assessed using the Columbia Suicide History (Mann et al., 1992). Family cohesion and support were assessed using the Family Adaptability and Cohesion Evaluation (FACES) (Olsen et al., 1982) with both parent and child self-report. Family conflict was measured for parent and child with the self-reported Conflict Tactics Scale (CTS) (Straus, 1979).

Statistical Analysis

Statistical analyses were performed using Stata version 18.0 (StataCorp., 2023). We first compared Black youth with and without a history of SRB at baseline for group differences in demographic and clinical characteristics using 2-tailed t-tests, Chi-square, and Fisher’s Exact tests. Self-report measures across the age bands were z-transformed for data analysis.

For the longitudinal analyses, we employed the methods used in previous reports (Melhem et al., 2007). We examined proximal risk indicators of SRB in three ways: (1) using measures collected at the time point closest to the suicidal event or at the time of censored observation (for those without SRB); (2) using the most severe observation for each risk indicator up to the suicidal event or censored observation; and (3) using all observations up to the suicidal event or censored time point and taking the mean of the measures over all time points. We controlled for parental history of suicide attempt in all models because these data were from a study of the transmission of suicidal behavior, and, by design, half of the youth participants were offspring of parents who attempted suicide.

We evaluated group differences on demographic and clinical characteristics between those with and without incident SRB using 2-tailed t-tests, Chi-square, and Fisher’s Exact tests. We then used univariate logistic regression for each of the significant variables, controlling for age, sex, site, and parental history of attempt and clustering by family relationship. Significant variables were entered in the multivariate Firth’s logistic regression models. Due to the small sample size and our analysis of rare events, we used a Firth correction to reduce bias (Puhr et al., 2017). We then used a backward stepwise approach to obtain the most parsimonious set of risk factors. For the backward stepwise regression, controlling for age, sex, and parental suicide attempt status, variables were eliminated until all remaining variables were significant at p < .05.

Results

Description of Overall Sample at Baseline

The sample consisted of 156 Black youth with a mean age of 12.6+ 2.7 years, 44% female, and 10% Hispanic. The average household income of the sample was in the range of $15,000–19,999 per year. At baseline, parents had completed an average of 12.67 years of education.

Baseline Comparisons of Youth With and Without SRB

At baseline, 18 youth had a history of SRB prior to their baseline assessment, and 138 had no such history (Table 1). The baseline SRB group was older (M = 13.8, SD = 2.9 vs. M = 12.5 SD = 2.6; t = –1.97, p = .05), more likely to be female (72% vs. 40%; χ² = 6.78, p = .01), and had higher scores on self-reported impulsivity (M = .47, SD = .97 vs. M = –.08, SD = .95; t = –2.20, p = .03), lifetime aggression (M = 1.38, SD = 1.31 vs. M = –.003, SD = 1.04; t = –4.57, p < .001), impulsive aggression (M = 1.35, SD = .72 vs. M = .15, SD = .92; t = –4.58, p < .001), irritability (M = 1.31, SD = 1.43 vs. M = .06, SD = 1.05; t = –4.29; p < .001), depression (M = 1.93, SD = 1.31 vs. M = .07, SD = .89; t = –7.43, p < .001), hopelessness (M = 1.47, SD = 1.40 vs. M = .03, SD = .83; t = –5.98, p < .001), and suicidal ideation (M = 2.16, SD = 1.78 vs. M = .005, SD = .96; t = –7.57; p < .001) (Table 1). Those with baseline SRB were also more likely to report childhood physical or sexual abuse (70.59% vs. 37.78%; χ² = 6.64, p = .01) and to have a diagnosis of depressive disorder (78% vs. 23%, χ² = 10.47, p < .001), anxiety disorder (61% vs. 24%, χ² = 10.88, p = .001), post-traumatic stress disorder (39% vs. 7%: Fisher’s Exact Test, p < .001), and oppositional defiant disorder (ODD) (41% vs. 14%; χ² = 7.73, p = .005). Overall, at baseline, the SRB group had more co-occurring diagnoses (M = 3.06, SD = 1.30 vs. M = 1.01, SD = 1.23, t = –6.57, p < .001). The SRB group was also more likely to have a parent diagnosed with bipolar disorder (56% vs. 30%; χ² = 4.83, p = .03) and to have lower total scores on measures of family cohesion (M = 44.5, SD = 8.99 vs. M = 49.68, SD = 8.93, t = 2.18, p = .03) (Table 1).

Table 1.

Baseline Characteristics Comparing Black Youth With and Without Baseline SRB.

Baseline Characteristics	No SRB (N = 138)	SRB (N = 18)	Test statistic	p-value
Demographic Characteristics
Age at baseline Mean,SD (range)	12.48, 2.59 (12.04–12.9)	13.77, 2.94 (12.31, 15.24)	t = –1.97, df = 154	.05
Pitt Site—N, %	107, 78%	14, 78%	χ² = .00	.98
Female Sex* – N, %	55, 40%	13, 72%	χ² = 6.78	.009
Hispanic Ethnicity—N, %	13, 9%	2, 11%	FET	.72
Household SES Mean, SD (range)	3.18, 2.30 (2.79, 3.56)	3.44, 2.20 (2.35, 4.54)	t = –.47, df = 153	.64
Parent years of education Mean, SD (range)	12.71, 2.15 (9–20)	12.39, 1.42 (9–14)	t = 0.62, df = 154	.54
Youth Clinical and Diagnostic Characteristics
Impulsivity* Mean, SD (range)	−.08, .95 (–.24, .08)	.47, .97 (–.04, .99)	t = –2.20, df = 146	.03
Aggression* Mean, SD (range)	−.003, 1.04 (–.19, .18)	1.38, 131 (.62, 2.13)	t = –4.57, df = 131	<.001
Impulsive Aggression* Mean, SD (range)	.15, .92 (–.01, .31)	1.35, .72 (.91, 1.79)	t = –4.58, df = 140	<.001
SR Irritability* Mean, SD (range)	.06, 1.05 (–.12, .24)	1.31, 1.43 (.55, 2.08)	t = –4.29, df = 142	<.001
SR Depression* Mean, SD (range)	.07, .89 (–.09, .22)	1.93, 1.31 (1.23, 2.63)	t = –7.43, df = 142	<.001
SR Hopelessness* Mean, SD (range)	.03, .83 (–.11, .18)	1.47, 1.40 (.73, 2.22)	t = –5.98, df = 143	<.001
SR Suicidal Ideation* Mean, SD (range)	.005, .96 (–.16, .17)	2.16, 1.78 (1.21, 3.11)	t = –7.57, df = 141	<.001
Childhood abuse (physical or sexual)*—N, %	40, 37.38%	12, 70.59%	χ² = 6.64	.01
Depressive Disorder Dx*-N, %	32, 23%	14, 78%	FET	<.001
Bipolar Disorder Dx-N, %	2, 1%	0, 0%	FET	1.00
Psychotic Disorder Dx-N, %	0, 0%	0, 0%	−	−
Anxiety Disorder Dx*-N, %	33, 24%	11, 61%	χ² = 10.88	.001
Post-Traumatic Stress Disorder* Dx-N, %	9, 7%	7, 39%	FET	<.001
Attention Deficit Hyperactivity Disorder Dx—N, %	27, 21%	7, 41%	χ² = 3.45	.06
Oppositional Defiant Disorder Dx*-N, %	18, 14%	7, 41%	χ² = 7.73	.005
Conduct Disorder Dx –N, %	9, 7%	3, 18%	FET	.15
Number of Dx* Mean, SD (range)	1.01, 1.23 (0–7)	3.06, 1.30 (0–5)	t = –6.57, df = 154	<.001
Parental characteristics
Parental attempt–N, %	91, 66%	15, 83%	FET	.18
Parental History of Childhood Abuse–N, %	84, 80%	14, 82%	FET	1.00
Parental Dx of Bipolar Disorder*–N, %	41, 30%	10, 56%	χ² = 4.83	.03
Parental Dx of Alcohol or substance use disorder–N, %	94, 68%	14, 78%	FET	.59
Parent Dx of Post-traumatic stress disorder–N, %	79, 57%	14, 78%	χ² = 2.79	.10
Family Characteristics
Family Adaptability and Cohesion Scale (FACES)
Child Report* Mean, SD (range)	49.68, 8.93 (48.01, 51.30)	44.5, 8.99 (39.71, 49.29)	t = 2.18, df = 136	.03
Parent Report Mean, SD (range)	52.48, 8.5 (50.97, 53.99)	51.79, 7.08 (47.7, 55.88)	t = .29, df = 135	.77
Conflict Tactics Scale (CTS)
Child Report Mean, SD (range)	48.9, 51.18 (39.73, 58.07)	75.07, 56.56 (42.41, 107.73)	t = –1.79, df = 134	.07
Parent Report Mean, SD (range)	60.7, 50.26 (51.62, 69.78)	87.64, 71.34 (46.45, 128.83)	t = –1.81, df = 131	.07

SRB: suicide-related behavior. FET: Fisher’s Exact Test. SR: self-reported.

Income Bracket: 1 = <$10,000; 2 = $10,000–14,999; 3 = $15,000–19,999; 4 = $20,000–29,999; 5 = $30,000–39,999; 6 = $40,000–49,999; 7 = $50,000–59,999; 8 = $60,000–69,999; 9 = >$70,000.

and bolded values indicate significance at p<0.5.

We examined each of the above variables’ associations with baseline SRB using logistic regression, while controlling for age, sex, site, and parental suicide attempt and clustering for relationship status. Those with a history of SRB at baseline were more likely to be female (OR = 4.05, 95% CI = 1.14, 11.56, p = .009; have self-reported impulsivity (OR = 2.09, 95% CI = 1.30, 3.36, p = .002), self-reported aggression (OR = 2.9, 95% CI 1.30, 3.36, p = .001), self-reported impulsive aggression (OR = 4.73, 95% CI = 2.41, 9.28. p < .001), self-reported depression (OR = 3.97, 95% CI = 2.61, 6.04, p < .001), self-reported irritability (OR = 2.05, 95% CI = 1.32, 3.17, p = .001), self-reported hopelessness (OR = 3.44, 95% CI = 2.00, 5.93, p < .001), self-reported suicidal ideation (OR = 3.00, 95% CI = 1.92, 4.71, p < .001), a history of childhood maltreatment (OR = 5.25, 95% CI = 1.54, 17.93, p = .008), baseline depression diagnosis (OR = 10.04, 95% CI = 3.27, 30.87, p < .001, baseline anxiety diagnosis (OR = 3.00, 95% CI = 1.33, 6.79, p = .008), post-traumatic stress disorder (PTSD) baseline diagnosis (OR = 5.47, 95% CI = 1.12, 26.68, p = .036), attention deficit hyperactivity disorder baseline diagnosis (OR = 5.18, 95% CI = 1.98, 13.55, p = .001), and ODD baseline diagnosis (OR = 8.38, 95% CI = 2.36, 29.70, p = .001) (Table 2).

Table 2.

Baseline Characteristics—Logistic Regression Controlling for Age, Sex, Site, and Parental Suicide Attempt Clustered by Family Membership.

Baseline Characteristics	Odds Ratio	95% Confidence Interval	z	p
Age	1.14	.85, 1.53	0.87	0.39
Sex*	4.05	1.4, 11.56	2.61	0.009
Site	.83	.27, 2.55	−0.32	0.75
Parental Suicide Attempt	2.69	.59, 12.20	1.28	0.20
Age	1.14	.87, 1.50	0.94	0.35
Sex*	3.92	1.39, 11.02	2.59	0.01
Site	.67	.20, 2.21	−0.66	0.51
Parental Suicide Attempt	2.15	.52, 8.95	1.05	0.29
Parental Bipolar Disorder	2.56	.57, 11.45	1.23	0.22
Age	1.18	.89, 1.56	1.13	0.26
Sex*	4.97	1.42, 17.40	2.51	0.01
Site	.60	.21, 1.75	−0.93	0.35
Parental Suicide Attempt	2.59	.66, 10.12	1.36	0.17
Self-Reported Impulsivity*	2.09	1.30, 3.36	3.04	0.002
Age	1.04	.76, 1.43	0.24	0.81
Sex*	9.23	2.26, 37.71	3.09	0.002
Site	.64	.21, 1.90	−0.80	0.42
Parental Suicide Attempt	1.96	.50, 7.76	0.96	0.34
Self-Reported Aggression*	2.90	1.52, 5.54	3.22	0.001
Age	1.07	.80, 1.42	0.46	0.65
Sex	2.04	.36, 11.67	0.80	0.42
Site	.66	.14, 3.11	−0.53	0.60
Parental Suicide Attempt	1.19	.25, 5.63	0.22	0.83
Self-Reported Impulsive Aggression*	4.73	2.41, 9.28	4.52	<0.001
Age	1.06	.84, 1.33	0.46	0.65
Sex	3.80	.84, 17.17	1.73	0.08
Site	.38	.13, 1.12	−1.75	0.08
Parental Suicide Attempt	1.97	.61, 6.34	1.13	0.26
Self-Reported Depression*	3.97	2.61, 6.04	6.44	<0.001
Age	1.04	.75, 1.45	0.25	0.80
Sex	4.81	.85, 27.2	1.78	0.08
Site	.82	.27, 2.47	−0.35	0.73
Parental Suicide Attempt	2.70	.57, 12.81	1.25	0.21
Self-Reported Irritability*	2.05	1.32, 3.17	3.21	0.001
Age	1.15	.87, 1.52	0.99	0.32
Sex*	4.80	1.42, 16.22	2.53	0.01
Site	.74	.24, 2.27	−0.53	0.60
Parental Suicide Attempt	2.66	.53, 13.32	1.19	0.24
Self-Reported Hopelessness*	3.44	2.00, 5.93	4.47	<0.001
Age	1.18	.90, 1.56	1.21	0.23
Sex	2.41	.55, 10.53	1.17	0.24
Site	.31	.08, 1.26	−1.63	0.10
Parental Suicide Attempt	.94	.22, 4.04	−0.08	0.93
Self-Reported Suicidal Ideation*	3.00	1.92, 4.71	4.79	<0.001
Age	1.19	.87, 1.62	1.08	0.28
Sex*	5.56	1.92, 16.12	3.16	0.002
Site	1.08	.32, 3.63	0.13	0.90
Parental Suicide Attempt	2.15	.45, 10.21	0.96	0.34
History of Childhood Abuse*	5.25	1.54, 17.93	2.64	0.008
Age	1.05	.82, 1.33	0.36	0.72
Sex*	3.82	1.26, 11.56	2.37	0.02
Site	.98	.23, 4.19	−0.03	0.97
Parental Suicide Attempt	2.30	.53, 9.91	1.12	0.26
Depression Baseline Dx*	10.04	3.27, 30.87	4.03	<0.001
Age	1.09	.79, 1.49	0.52	0.61
Sex*	3.35	1.11, 10.12	2.15	0.03
Site	.84	.26, 2.70	−0.29	0.77
Parental Suicide Attempt	2.17	.48, 9.84	1.01	0.32
Anxiety Disorder Baseline Dx*	3.00	1.33, 6.79	2.64	0.008
Age	1.09	.80, 1.48	0.54	0.59
Sex*	3.25	1.12, 9.44	2.17	0.03
Site	.74	.22, 2.49	−0.49	0.62
Parental Suicide Attempt	1.81	.43, 7.57	0.81	0.42
PTSD Baseline Dx*	5.47	1.12, 26.68	2.10	0.036
Age	1.27	.94, 1.70	1.58	0.11
Sex*	6.53	2.20, 19.37	3.38	0.001
Site	.72	.21, 2.48	−0.52	0.60
Parental Suicide Attempt	2.71	.69, 10.64	1.43	0.15
ADHD Baseline Dx*	5.18	1.98, 13.55	3.35	0.001
Age	1.19	.87, 1.63	1.07	0.29
Sex*	6.92	1.77, 27.09	2.78	0.005
Site	.74	.21, 2.55	−0.48	0.63
Parental Suicide Attempt	2.63	.61, 11.29	1.30	0.19
Oppositional Defiant Disorder Baseline Dx*	8.38	2.36, 29.70	3.29	0.001
Age	1.21	1.45, 14.03	1.02	0.31
Sex*	4.51	1.45, 14.03	2.60	0.009
Site	.92	.30, 2.77	−0.15	0.88
Parental Suicide Attempt	2.81	.64, 12.22	1.37	0.17
Conduct Disorder Baseline Dx	2.13	.49, 9.38	1.00	0.32
Age	1.16	.79, 1.71	0.77	0.44
Sex*	3.48	1.16, 10.47	2.22	0.026
Site	.78	.24, 2.58	−0.41	0.68
Parental Suicide Attempt	2.58	.52, 12.78	1.16	0.25
FACES Total Score	.95	.89, 1.01	−1.72	0.09

and bolded values indicate significance at p<0.5.

Longitudinal Comparison of Youth With and Without SRB on Follow-Up

Over the 15-year follow-up period, seven incidents of first-onset SRB occurred. In addition, a total of five individuals with a history of SRB at baseline engaged in seven additional SRB events. For those with more than one SRB event over follow-up, we used the first event occurring during follow-up. Therefore, the analyses included 12 SRB events occurring over the follow-up period of an average of 8.0 years (SD = 4.0, range = 1–15.2), with no significant difference in years of follow-up between those with SRB and those without (8.7±3.4 vs. 7.9±4.0, p = .47). An average of 5.33 months elapsed between the suicidal event and the time point prior.

We examined longitudinal data in the three ways described earlier. Results for each of these approaches identified consistent proximal risk indicators. Therefore, we report only on the time closest to the event, focusing on measures of risk most proximal to the SRB event. Upon follow-up, 12 youth (8.63%) endorsed SRB since baseline (Table 3). The new onset SRB group was younger (M = 12.8, SD = 2.7 years vs. M = 16.6, SD = 2.1 years; t = 5.79, p < .001) and had higher scores on self-reported aggression (M = 1.00, SD = 1.24 vs. M = .12, SD = .99; t = –2.85, p = .005), impulsive aggression (M = 1.33, SD = 1.17 vs. M = .29, SD = .93; t = –3.39; p = .002), irritability (M = 1.42, SD = 1.23 vs. M = .44, SD = 1.41; t = –2.27, p = .03), depression (M = 1.39, SD = 1.54 vs. M = .23, SD = 1.14; t = –3.16, p = .002), hopelessness (M = .83, SD = 1.51 vs. M = –.002, SD = .95; t = –2.65, p = .009) and suicidal ideation (M = 1.89, SD = 1.54 vs. M = .17, SD = 1.21; t = –4.46, p < .001). The SRB group was more likely to report physical or sexual abuse in childhood (80% vs. 30%, Fisher’s Exact Test = .008) and to be diagnosed with a depressive disorder (75% vs.23%, Fisher’s Exact Test < .001). The SRB group had a higher number of co-occurring diagnoses (M = 2.0, SD = 1.35 vs. M = .98, SD = 1.39; t = –2.45, p = .02). The parent reported Conflict Tactics Scale scores were higher in the SRB group (M = 103.91, SD = 76.68 vs. M = 51.86, SD = 41.52; t = –2.51, p = .02).

Table 3.

Follow-Up Characteristics Comparing Black Youth With and Without SRB at the Time Point Prior to SRB or Last Follow-Up.

Follow-Up Characteristics	No SRB during F/U (n = 127)	SRB during F/U (n = 12)	Test statistic	p-value
Demographic Characteristics
Age Mean, SD (range)	16.55, 2.12 16.17, 16.92	12.75, 2.70 11.03, 14.47	t = 5.79, df = 137	<.001
Pitt Site—N, %	99, 78%	10, 83%	χ² = .188	0.67
Female Sex—N, %	51, 40%	7, 58%	χ² = 1.49	.22
Hispanic Ethnicity—N, %	13, 10.24%	0, 0%	FET	.64
Household SES Mean, SD (range)	3.52, 2.16 (3.12, 3.92)	3.42, 1.88 (2.22, 4.61)	t = 0.16, df = 125	0.91
Youth Clinical and Diagnostic Characteristics
Impulsivity Mean, SD (range)	−.002, .83 (–.21, .20)	.439, .96 (–.21, 1.08)	t = –1.60, df = 76	.11
Aggression Mean, SD (range)	.12, .99 (–.06, .30)	.998, 1.24 (.21, 1.78)	t = –2.85, df = 125	.005
Impulsive Aggression Mean, SD (range)	.29, .93 (.05, .53)	1.33, 1.17 (.54, 2.12)	t = –3.29, df = 70	.002
SR Irritability Mean, SD (range)	.44, 1.41 (.14, .74)	1.42, 1.23 (.64, 2.20)	t = –2.27, df = 97	.03
SR Depression Mean, SD (range)	.23, 1.14 (–.02, .47)	1.39, 1.54 (.41, 2.37)	t = –3.16, df = 96	.002
SR Hopelessness Mean, SD (range)	−.002, .95 (–.20, .19)	.83, 1.51 (–.13, 1.79)	t = –2.65, df = 103	.009
SR Sui Ideation Mean, SD (range)	.17, 1.21 (–.08, .43)	1.89, 1.54 (.91, 2.86)	t = –4.46, df = 98	<.001
Childhood abuse (physical or sexual) N, %	34, 30%	8, 80%	FET	.008
Depressive Disorder Dx—N, %	28, 23%	9, 75%	FET	<.001
Bipolar Disorder Dx—N, %	2, 1.61%	1, 8.33%	FET	.24
Psychotic Disorder Dx—N, %	0, 0%	0, 0%	−	−
Anxiety Disorder Dx—N, %	24, 20%	3, 25%	FET	.71
Post-Traumatic Stress Disorder Dx—N, %	10, 9%	2, 17%	FET	.31
Attention Deficit Hyperactivity Disorder Dx—N, %	18, 15%	4, 33%	FET	.12
Oppositional Defiant Disorder Dx—N, %	14, 27%	4, 33%	FET	.73
Conduct Disorder Dx—N, %	7, 13%	0, 0%	FET	.33
Number of dx M, SD (range)	.98, 1.39 (0–5)	2.0, 1.35 (0–5)	t = –2.45, df = 137	.02
Parental characteristics
Parental attempt—N, %	84, 66%	9, 75%	FET	.75
Parental History of Childhood Abuse—N, %	82, 81%	8, 72%	FET	.45
Parent Dx of Bipolar Disorder—N, %	40, 32%	5, 42%	FET	.53
Parental Dx of Alcohol or Substance Use Disorder—N, %	81, 64%	11, 92%	FET	.06
Parent Dx of Post-Traumatic Stress Disorder—N, %	73, 58%	9, 75%	FET	.36
Family Characteristics
Family Adaptability and Cohesion Scale (FACES)
Child Report Mean, SD (range)	46.69, 8.67 (43.88, 49.50)	46, 4.54 (42.95, 49.05)	t = 0.25, df = 48	.80
Parent Report Mean, SD (range)	50.71, 7.76 (47.86, 53.55)	50.64, 7.03 (45.91, 55.36)	t = 0.03, df = 40	.98
Conflict Tactics Scale (CTS)
Child Report Mean, SD (range)	76.91, 61.12 (49.81, 104.01)	81.82, 59.56 (41.81, 121.83)	t = –0.22, df = 31	.83
Parent Report Mean, SD (range)	51.86, 41.52 (32.96, 70.76)	103.91, 76.68 (52.39, 155.43)	t = –2.51, df = 30	.02
Treatment Received
Outpatient Treatment—N, %	16, 32%	6, 60%	FET	.15
Inpatient Treatment—N, %	8, 7%	1, 10%	FET	.52

SRB: suicide-related behavior. FET: Fisher’s Exact Test.

Bolded values indicate significance at p<0.5.

We then examined each of the above variables as proximal risk indicators of SRB on follow-up using univariate logistic regression, controlling for age, sex, site, and parental suicide attempt, while clustering for relationship status (Table 4). Significant results included self-reported aggression (OR = 1.77, 95% CI = 1.09, 2.89, p = .02), irritability (OR = 1.65, 95% CI = 1.12, 2.43, p = .01), suicidal ideation (OR = 1.63, 95% CI = 1.003, p = .05), and a depression diagnosis at the time point prior to SRB manifestation (OR = 15.44, CI = 3.07, 77.73, p = .001). While not statistically significant, medium to large effect sizes were found for several variables, including self-reported impulsive aggression (OR = 2.44, 95% CI = .71, 8.39, p = .16), parental substance use disorder (OR = 5.67, 95% CI = .61, 52.39, p = .13), and physical or sexual abuse (OR = 8.63, 95% CI = .92, 80.60, p = .06).

Table 4.

Predictors of SRB on Follow—Up—Logistic Regression Controlling for Age, Sex, Site, and Parental Suicide Attempt Clustering by Family Membership.

Risk Factors for SRB	Odds Ratio	95% Confidence Interval	z	p
Age*	.54	.39, .74	−3.97	<.001
Sex	1.39	.35, 5.52	0.47	.64
Site	.29	.04, 2.12	−1.22	.22
Parental Suicide Attempt	1.77	.39, 7.99	0.74	.46
Age*	.54	.39, .75	−3.66	<.001
Sex	1.04	.25, 4.37	0.05	.96
Site	.33	.04, 2.41	−1.10	.28
Parental Suicide Attempt	1.71	.36, 8.01	0.68	.50
Parental Alcohol Substance Use Disorder	5.67	.61, 52.39	1.53	.13
Age*	.53	.37, .75	−3.58	<.001
Sex	1.10	.25, 4.77	0.13	.90
Site	.48	.06, 3.77	−0.70	.48
Parental Suicide Attempt	1.36	.28, 6.57	0.38	.70
Aggression*	1.77	1.09, 2.89	2.29	.02
Age*	.56	.39, .82	−2.96	.003
Sex	1.46	.31, 6.97	0.47	.64
Site	.19	.02, 1.44	−1.61	.11
Parental Suicide Attempt	2.10	.38, 11.65	.85	.40
Impulsive Aggression	2.44	.71, 8.39	1.41	.16
Age*	.51	.36, .73	−3.74	<.001
Sex	.83	.17, 4.00	−0.23	.82
Site	.36	.04, 3.21	−0.91	.36
Parental Suicide Attempt	1.65	.27, 10.11	0.54	.59
SR Irritability*	1.65	1.12, 2.43	2.53	.01
Age*	.54	.37, .78	−3.24	.001
Sex	1.00	.23, 4.27	0.00	1.00
Site	.37	.04, 2.28	−0.90	.37
Parental Suicide Attempt	1.04	.19, 5.77	0.05	.96
SR Depression	1.40	.67, 2.92	0.91	.36
Age*	.51	.38, .67	−4.81	<.001
Sex	.71	.15, 3.43	−0.42	.67
Site	.37	.03, 3.98	−0.82	.41
Parental Suicide Attempt	2.05	.41, 10.31	0.87	.39
SR Hopelessness	1.72	.81, 3.67	1.40	.16
Age*	.53	.36, .80	−3.06	.002
Sex	.67	.13, 3.42	−0.48	.63
Site	.25	.03, 2.32	−1.22	.22
Parental Suicide Attempt	.95	.19, 4.80	−0.06	.95
SR Suicidal Ideation	1.63	1.003, 2.65	1.97	.05
Age*	.53	.36, .78	−3.28	.001
Sex	.60	.05, 6.85	−0.41	.68
Site	.39	.04, 3.96	−0.80	.42
Parental Suicide Attempt	1.39	.24, 7.93	0.38	.71
Childhood Abuse	8.63	0.92, 80.60	1.89	.06
Age*	.49	.34, .70	−3.85	<.001
Sex	.78	.15, 4.16	−0.29	.77
Site	.60	.06, 5.63	−0.45	.65
Parental Suicide Attempt	1.13	.20, 6.51	0.14	.89
Depression Dx*	15.44	3.07, 77.73	3.32	.001
Age	.89	.55, 1.45	−0.47	.64
Sex	1.92	.31, 11.76	0.70	.48
Site	.34	.02, 5.89	−0.74	.46
Parental Suicide Attempt	1.83	.22, 15.12	0.56	.58
CTS-Total Score—Parent Report	1.02	1.00, 1.03	1.93	.06

SRB: suicide-related behavior.

and bolded values indicate significance at p<0.5.

Finally, using Firth logistic regression, we examined proximal risk indicators present at the time point prior to SRB. Using a backward stepwise approach, we included significant variables from the univariate regression models described above, which controlled for age, sex, site, and parental suicide attempt. In this sample, younger age (β = –.60, 95% CI = –.96, –.23, p = .001) and a diagnosis of depression (β = 2.32, 95% CI = .66, 3.98, p = .006) at the time point prior to SRB were significant proximal risk indicators of SRB (Table 5). Post hoc sensitivity analyses were conducted and showed excellent ability of the model to discriminate between groups (Area Under the Curve = 0.92, SE = 0.03, 95% CI: 0.87, 0.98).

Table 5.

Final Firth Regression Model—Predictors of SRB on Follow-Up Controlling for Age, Sex, Site, Parental Suicide Attempt.

Predictors	β Coefficient	95% Confidence Interval	z	p
Site	−.48	−2.44, 1.48	−.48	.63
Sex	−.19	−1.64, 1.26	−.26	.80
Age Prior*	−.60	−.96, –.23	−3.22	.001
Parental Suicide Attempt	.07	−1.45, 1.59	.09	.93
Depression Dx Time point Prior*	2.32	.66, 3.98	2.74	.006

and bolded values indicate significance at p<0.5.

Discussion

This is one of the few prospective studies to examine risk factors for SRB in Black youth at elevated risk for mood disorders. At the univariate level on follow-up, the clinical proximal risk indicators of SRB were similar to those found in our prior studies of the entire sample of youth (Brent et al., 2002, 2015). Self-reported aggression, irritability, and suicidal ideation, and a clinician-assessed diagnosis of depression were all significant at the univariate level on follow-up. While not statistically significant, medium to large effect sizes were also found for self-reported impulsive aggression, parental substance use disorder, and childhood abuse. These findings were consistent with our prior work in the full sample assessing for SRB risk in youth (Brent et al., 2002, 2015). However, multivariate regression found that only younger age and a diagnosis of depressive disorder were statistically significant proximal risk indicators of SRB when all significant univariate variables were included in one regression model.

The strongest and most powerful indicator of proximal risk for SRB on follow-up in this sample was the presence of a diagnosis of a depressive disorder at the time point prior, which is consistent with both our earlier cross-sectional and prospective studies of the entire sample of youth (Brent et al., 2002, 2004; Melhem et al., 2007, 2019; J. Zelazny et al., 2021), as well as community and epidemiological studies (Bridge et al., 2006). Self-reported suicidal ideation was more severe in the SRB group at the time point prior. While prior studies have reported that Black youth are less likely to disclose suicidal thoughts prior to SRB (Lee & Wong, 2020; Romanelli et al., 2022), recent Youth Risk Behavior Survey data show that Black youth are reporting increasing rates of suicidal ideation and SRB (Bommersbach et al., 2025; CDC, 2024). In our study, Black youth in the SRB group reported more severe suicidal ideation at the time point prior to engaging in SRB. This reinforces the importance of assessing Black youth for suicidal ideation.

Irritability was also more severe in the SRB group. Considering the younger age of those with SRB in this sample, this is consistent with the current literature suggesting irritability as a risk factor for SRB in younger youth (Sheftall et al., 2016). Indeed, irritability has been associated with impairment and suicide risk independent of co-occurring diagnoses (Leibenluft et al., 2024). Irritability has also been associated with early bipolar disorder (Hafeman et al., 2023), which was more prevalent in the parents of the youth in this sample. Irritability is known to be a primary symptom of depression in younger children. Of note, the diagnostic criteria for ODD are also heavily weighted on irritability and related constructs. The historical overdiagnosis of ODD in Black youth is well documented (Legha, 2025) and may result from provider assumptions of “defiance” rather than irritability as a symptom of depression or trauma. Despite higher rates of diagnosis of ODD in this cohort of Black youth, a diagnosis of depression continued to be the strongest risk indicator of SRB after controlling for other variables, highlighting the importance of thoroughly screening for depression, trauma, and other potential sources of irritability. The SRB group also had higher numbers of co-occurring disorders. These findings highlight the importance of decreasing health disparities in access and utilization of mental health services, as the gap in outpatient mental health treatment significantly increased between Black and White youth from 2019 to 2021 (Olfson et al., 2024).

Prior studies have found a diagnosis of Attention Deficit Hyperactivity Disorder (ADHD) to be a predictor of both SRB and death by suicide, particularly in younger samples, including samples of Black youth (Melhem et al., 2019; Sheftall et al., 2016; J. Zelazny et al., 2021), but ADHD was not a significant risk indicator of SRB in this sample of Black youth. In fact, there were no group differences in ADHD rates between those with and without SRB on follow-up.

While not statistically significant, a large effect size was found for parental substance use disorder, which is considered by some to be a “disease of despair.” Deaths of despair (death from drug overdose, alcoholic liver disease, or suicide) have tripled from 2013 to 2022 in Black Americans (Friedman & Hansen, 2024), underscoring the urgency of improving access to mental health care and substance use treatment in communities of color. There is emerging data suggesting that parental diseases and deaths of despair may be contributing to rising adolescent suicide rates, with suicide attempt rates nearly twice as high in youth with a parent misusing opioids as those without (Brent et al., 2019).

We address the strengths and limitations of the study. First, among the strengths of this study, the high-risk prospective design provides an opportunity to detect precursors of rare events such as SRB. This is also one of the few prospective studies of Black youth at risk. The principal limitation of this study is the small sample size and the rare outcome of interest. Firth logistic regression is appropriate for rare events, but stepwise regression is interpreted cautiously. The study was conducted between 1997 and 2012, when contextual factors such as the effects of racial discrimination and exposure to community violence were not routinely measured. This is especially relevant considering the stress diathesis model, which has been widely used in the study of mood disorders, schizophrenia, and suicide (Caspi et al., 2003; Corcoran et al., 2003; Grunebaum et al., 2006; Mann & Rizk, 2020). According to the stress diathesis model, the convergence of genetic and environmental factors determines an individual’s underlying vulnerability to suicide attempt. The underlying vulnerability to suicide attempt is then moderated by an individual’s exposure to stressors. Future studies must include assessment of these domains. Youth included in this sample were specifically recruited into this study due to a clinically referred parental mood disorder, which elevated their risk for mood disorder, and in the case of offspring of a parent who also made a suicide attempt, elevated their risk for SRB. Hence, these results may not be generalizable to the general population. In addition, due to the low base rate of actual suicide attempts and the high risk for future attempts in those who have made aborted or interrupted attempts (Daray et al., 2025), we used the first episode of any SRB rather than actual suicide attempt as the primary outcome. Therefore, our results are broadly relevant to SRBs and not specific to actual suicide attempts.

Conclusion

In this sample of Black youth, the strongest proximal risk indicator of SRB was a diagnosis of depression, highlighting the importance of early identification of depression in Black youth, particularly in those with a family history of mood disorder. Overall, the clinical risk indicators of SRB in this sample were similar to those noted in the literature in general, which further emphasizes the importance of decreasing disparities in access and utilization of mental health care services. For psychiatric mental health nurses at all levels of practice working with youth, it is vital to assess for suicidal thoughts and behaviors, not only in youth with classical symptoms of mood disorders, but in those presenting with prominent symptoms of irritability. Clinicians should pay close attention to the family history of mood disorders as well as other parental diseases of despair. Psychiatric mental health nurses who work with adults with mood disorders should be aware that their children may be at higher risk of depression and SRBs at younger ages and recommend early screening if concerning symptoms are raised.

Much remains to be learned about risk factors for SRB specific to Black youth in the United States. Larger-scale longitudinal studies are needed. Additional research is needed on the effects of parental diseases of despair on offspring. Future studies must also include assessment of socioecological risk and protective factors, such as discrimination, poverty, and exposure to community violence, as well as family and social support systems.

Footnotes

Acknowledgements

The authors thank the children and families who participated in the study.

Author Roles

JZ contributed to acquisition, analysis or interpretation, drafted the manuscript, gave final approval, and agrees to be accountable for all aspects of the work, ensuring integrity and accuracy. GP contributed to analysis and interpretation of the data, served as a statistical expert, critically revised the manuscript, gave final approval and agrees to be accountable for all aspects of the work. JJM, JK, and MO contributed to interpretation of the data, critically revised the manuscript, gave final approval, and agreed to be accountable for all aspects of the work. AB contributed to the acquisition and interpretation of the data, critically revised the manuscript, gave final approval, and agrees to be accountable for all aspects of the work. DB contributed to analysis and interpretation of the data, t, critically revised the manuscript, gave final approval and agrees to be accountable for all aspects of the work. NM contributed to analysis and interpretation of the data, served as a statistical expert, critically revised the manuscript, gave final approval and agrees to be accountable for all aspects of the work.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Brent has received royalties from Guilford Press, from the electronic self-rated version of the C-SSRS from eRT, Inc., and from performing duties as an UptoDate Psychiatry Section Editor, and has received consulting fees from Healthwise. Drs. Burke, Mann, and Oquendo receive royalties from the Research Foundation for Mental Hygiene for the commercial use of the Columbia Suicide Severity Rating Scale. Dr. Oquendo serves as an advisor to Alkermes, Mind Medicine (pro bono), and Fundacion Jimenez Diaz. Her family-owned stock in Bristol Myers Squibb (sold 2024). Dr. Keilp holds stock in Pfizer, Zoetis, and Viatris, but there is no conflict with the current study. Drs. Melhem and Zelazny and Ms. Porta have reported no biomedical financial interests or potential conflicts of interest.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the National Institute of Mental Health [MH056612 (David Brent, MD) and MH056390 (J. John Mann, MD)].

Ethics Approval

The study was approved by the institutional review boards at the University of Pittsburgh and New York State Psychiatric Institute.

Informed Consent Statement

Written informed consent was obtained from all participants aged 18 and older. For youth under 18, written parental permission and youth assent were obtained prior to beginning any research procedures.

ORCID iD

Jamie Zelazny

Data Availability Statement

For data sharing requests, please contact the corresponding author for access.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). American Psychiatric Publishing.

Beck

A. T.

Kovacs

Weissman

(1979). Assessment of suicidal intention: The Scale for Suicide Ideation. Journal of Consulting and Clinical Psychology, 47(2), 343–352. https://doi.org/10.1037//0022-006x.47.2.343

Beck

A. T.

Ward

C. H.

Mendelson

Mock

Erbaugh

(1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. https://doi.org/10.1001/archpsyc.1961.01710120031004

Beck

A. T.

Weissman

Lester

Trexler

(1974). The measurement of pessimism: The hopelessness scale. Journal of Consulting and Clinical Psychology, 42(6), 861–865. https://doi.org/10.1037/h0037562

Bernstein

D. P.

Fink

Handelsman

Foote

Lovejoy

Wenzel

Sapareto

Ruggiero

(1994). Initial reliability and validity of a new retrospective measure of child abuse and neglect. American Journal of Psychiatry, 151(8), 1132–1136. https://doi.org/10.1176/ajp.151.8.1132

Bommersbach

T. J.

Olfson

Rhee

T. G.

(2025). National trends in suicidal thoughts and suicide attempts among high school students in the United States. American Journal of Psychiatry, 182(7), 639–659. https://doi.org/10.1176/appi.ajp.20240706

Brent

D. A.

Hur

Gibbons

R. D.

(2019). Association between parental medical claims for opioid prescriptions and risk of suicide attempt by their children. JAMA Psychiatry, 76(9), 941–947. https://doi.org/10.1001/jamapsychiatry.2019.0940

Brent

D. A.

Melhem

N. M.

Oquendo

Burke

Birmaher

Stanley

Biernesser

Keilp

Kolko

Ellis

Porta

Zelazny

Iyengar

Mann

J. J.

(2015). Familial pathways to early-onset suicide attempt: A 5.6-year prospective study. JAMA Psychiatry, 72(2), 160–168. https://doi.org/10.1001/jamapsychiatry.2014.2141

Brent

D. A.

Oquendo

Birmaher

Greenhill

Kolko

Stanley

Zelazny

Brodsky

Bridge

Ellis

Salazar

J. O.

Mann

J. J.

(2002). Familial pathways to early-onset suicide attempt: Risk for suicidal behavior in offspring of mood-disordered suicide attempters. Archives of General Psychiatry, 59(9), 801–807. https://doi.org/10.1001/archpsyc.59.9.801

10.

Brent

D. A.

Oquendo

Birmaher

Greenhill

Kolko

Stanley

Zelazny

Brodsky

Firinciogullari

Ellis

S. P.

Mann

J. J.

(2003). Peripubertal suicide attempts in offspring of suicide attempters with siblings concordant for suicidal behavior. American Journal of Psychiatry, 160(8), 1486–1493. https://doi.org/10.1176/appi.ajp.160.8.1486

11.

Brent

D. A.

Oquendo

Birmaher

Greenhill

Kolko

Stanley

Zelazny

Brodsky

Melhem

Ellis

S. P.

Mann

J. J.

(2004). Familial transmission of mood disorders: Convergence and divergence with transmission of suicidal behavior. Journal of the American Academy of Child & Adolescent Psychiatry, 43(10), 1259–1266. https://doi.org/10.1097/01.chi.0000135619.38392.78

12.

Bridge

J. A.

Goldstein

T. R.

Brent

D. A.

(2006). Adolescent suicide and suicidal behavior. Journal of Child Psychology and Psychiatry, 47(3–4), 372–394. https://doi.org/10.1111/j.1469-7610.2006.01615.x

13.

Bridge

J. A.

Horowitz

L. M.

Fontanella

C. A.

Sheftall

A. H.

Greenhouse

Kelleher

K. J.

Campo

J. V.

(2018). Age-related racial disparity in suicide rates among US youths from 2001 through 2015. JAMA Pediatrics, 172(7), 697–699. https://doi.org/10.1001/jamapediatrics.2018.0399

14.

Brown

G. L.

Goodwin

F. K.

(1986). Human aggression and suicide. Suicide and Life-Threatening Behavior, 16(2), 223–243. https://doi.org/10.1111/j.1943-278x.1986.tb00353.x

15.

Buss

A. H.

Durkee

(1957). An inventory for assessing different kinds of hostility. Journal of Consulting Psychology, 21(4), 343–349. https://doi.org/10.1037/h0046900

16.

Caspi

Sugden

Moffitt

T. E.

Taylor

Craig

I. W.

Harrington

McClay

Mill

Martin

Braithwaite

Poulton

(2003). Influence of life stress on depression: Moderation by a polymorphism in the 5-HTT gene. Science, 301(5631), 386–389. https://doi.org/10.1126/science.1083968

17.

Centers for Disease Control and Prevention. (2023). Disparities in suicide. https://www.cdc.gov/suicide/facts/disparities-in-suicide.html

18.

Centers for Disease Control and Prevention. (2024). Youth risk behavior survey data summary & trends report: 2011-2021. https://www.cdc.gov/healthyyouth/data/yrbs/yrbs_data_summary_and_trends.htm

19.

Centers for Disease Control and Prevention. (2025). National vital statistics system, 2024 provisional mortality on CDC WONDER Online Database. CDC WONDER Database.

20.

Chaffin

Wherry

J. N.

Newlin

Crutchfield

Dykman

(1997). The Abuse Dimensions Inventory: Initial data on a research measure of abuse severity. Journal of Interpersonal Violence, 12(4), 569–589. https://doi.org/10.1177/088626097012004006

21.

Corcoran

Walker

Huot

Mittal

Tessner

Kestler

Malaspina

(2003). The stress cascade and schizophrenia: Etiology and onset. Schizophrenia Bulletin, 29(4), 671–692. https://doi.org/10.1093/oxfordjournals.schbul.a007038

22.

Daray

F. M.

Grendas

L. N.

Neupane

S. P.

Chiapella

L. C.

Fisher

P. W.

Gould

M. S.

Posner

Galfalvy

Jaffe

Mann

J. J.

(2025). Prediction of fatal and non-fatal suicide attempts by the Columbia Suicide Severity Rating Scale (C-SSRS): Systematic review and meta-analysis. British Journal of Psychiatry. Advance online publication. https://doi.org/10.1192/bjp.2025.10316

23.

Friedman

Hansen

(2024). Trends in deaths of despair by race and ethnicity from 1999 to 2022. JAMA Psychiatry. Advance online publication. https://doi.org/10.1001/jamapsychiatry.2024.0303

24.

Grunebaum

M. F.

Ramsay

S. R.

Galfalvy

H. C.

Ellis

S. P.

Burke

A. K.

Sher

Printz

D. J.

Kahn

D. A.

Mann

J. J.

Oquendo

M. A.

(2006). Correlates of suicide attempt history in bipolar disorder: A stress-diathesis perspective. Bipolar Disorder, 8(5 Pt. 2), 551–557. https://doi.org/10.1111/j.1399-5618.2006.00304.x

25.

Hafeman

D. M.

Merranko

Joseph

H. M.

Goldstein

T. R.

Goldstein

B. I.

Levenson

Axelson

Monk

Sakolsky

Iyengar

Birmaher

(2023). Early indicators of bipolar risk in preschool offspring of parents with bipolar disorder. Journal of Child Psychology and Psychiatry, 64(10), 1492–1500. https://doi.org/10.1111/jcpp.13739

26.

Kaufman

Birmaher

Brent

Rao

Flynn

Moreci

Williamson

Ryan

(1997). Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry, 36(7), 980–988. https://doi.org/10.1097/00004583-199707000-00021

27.

Kazdin

A. E.

French

N. H.

Unis

A. S.

Esveldt-Dawson

Sherick

R. B.

(1983). Hopelessness, depression, and suicidal intent among psychiatrically disturbed inpatient children. Journal of Consulting and Clinical Psychology, 51(4), 504–510. https://doi.org/10.1037//0022-006x.51.4.504

28.

Kazdin

A. E.

Rodgers

Colbus

(1986). The hopelessness scale for children: Psychometric characteristics and concurrent validity. Journal of Consulting and Clinical Psychology, 54(2), 241–245. https://doi.org/10.1037//0022-006x.54.2.241

29.

Kazdin

A. E.

Rodgers

Colbus

Siegel

(1987). Children’s Hostility Inventory: Measurement of aggression and hostility in psychiatric inpatient children. Journal of Clinical Child Psychology, 16(4), 320–328. https://doi.org/10.1207/s15374424jccp1604_5

30.

Kovacs

(1985). The Children’s Depression, Inventory (CDI). Psychopharmacology Bulletin, 21(4), 995–998.

31.

Lee

C. S.

Wong

Y. J.

(2020). Racial/ethnic and gender differences in the antecedents of youth suicide. Cultural Diversity & Ethnic Minority Psychology, 26(4), 532–543. https://doi.org/10.1037/cdp0000326

32.

Legha

R. K.

(2025). There are no bad kids: An antiracist approach to oppositional defiant disorder. Pediatrics, 155(2). https://doi.org/10.1542/peds.2024-068415

33.

Leibenluft

Allen

L. E.

Althoff

R. R.

Brotman

M. A.

Burke

J. D.

Carlson

G. A.

Dickstein

D. P.

Dougherty

L. R.

Evans

S. C.

Kircanski

Klein

D. N.

Malone

E. P.

Mazefsky

C. A.

Nigg

Perlman

S. B.

Pine

D. S.

Roy

A. K.

Salum

G. A.

Shakeshaft

Stringaris

(2024). Irritability in youths: A critical integrative review. American Journal of Psychiatry, 181(4), 275–290. https://doi.org/10.1176/appi.ajp.20230256

34.

Mann

J. J.

McBride

P. A.

Brown

R. P.

Linnoila

Leon

A. C.

DeMeo

Mieczkowski

Myers

J. E.

Stanley

(1992). Relationship between central and peripheral serotonin indexes in depressed and suicidal psychiatric inpatients. Archives of General Psychiatry, 49(6), 442–446. https://doi.org/10.1001/archpsyc.1992.01820060022003

35.

Mann

J. J.

Rizk

M. M.

(2020). A brain-centric model of suicidal behavior. American Journal of Psychiatry, 177(10), 902–916. https://doi.org/10.1176/appi.ajp.2020.20081224

36.

Melhem

N. M.

Brent

D. A.

Ziegler

Iyengar

Kolko

Oquendo

Birmaher

Burke

Zelazny

Stanley

Mann

J. J.

(2007). Familial pathways to early-onset suicidal behavior: Familial and individual antecedents of suicidal behavior. American Journal of Psychiatry, 164(9), 1364–1370. https://doi.org/10.1176/appi.ajp.2007.06091522

37.

Melhem

N. M.

Porta

Oquendo

M. A.

Zelazny

Keilp

J. G.

Iyengar

Burke

Birmaher

Stanley

Mann

J. J.

Brent

D. A.

(2019). Severity and variability of depression symptoms predicting suicide attempt in high-risk individuals. JAMA Psychiatry, 76(6), 603–613. https://doi.org/10.1001/jamapsychiatry.2018.4513

38.

Olfson

McClellan

Zuvekas

S. H.

Wall

Blanco

(2024). Mental health impairment and outpatient mental health care of US children and adolescents. JAMA Psychiatry. Advance online publication. https://doi.org/10.1001/jamapsychiatry.2024.0088

39.

Olsen

D. H.

Portner

Bell

R. Q.

(1982). FACES II: Family Adaptability and Cohesion Evaluation Scales. Family Social Science, University of Minnesota.

40.

Plomin

(1983). Childhood temperament ( Lahey

Kazdin

A. E.

, Eds. Vol. 76). Springer Science & Business Media.

41.

Posner

Oquendo

M. A.

Gould

Stanley

Davies

(2007). Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of suicidal events in the FDA’s pediatric suicidal risk analysis of antidepressants. American Journal of Psychiatry, 164(7), 1035–1043. https://doi.org/10.1176/ajp.2007.164.7.1035

42.

Puhr

Heinze

Nold

Lusa

Geroldinger

(2017). Firth’s logistic regression with rare events: Accurate effect estimates and predictions? Statistics in Medicine, 36(14), 2302–2317. https://doi.org/10.1002/sim.7273

43.

Ramchand

Gordon

J. A.

Pearson

J. L.

(2021). Trends in suicide rates by race and ethnicity in the United States. JAMA Network Open, 4(5), Article e2111563. https://doi.org/10.1001/jamanetworkopen.2021.11563

44.

Romanelli

Sheftall

A. H.

Irsheid

S. B.

Lindsey

M. A.

Grogan

T. M.

(2022). Factors associated with distinct patterns of suicidal thoughts, suicide plans, and suicide attempts among US adolescents. Prevention Science, 23(1), 73–84. https://doi.org/10.1007/s11121-021-01295-8

45.

Ruch

D. A.

Horowitz

L. M.

Hughes

J. L.

Sarkisian

Luby

J. L.

Fontanella

C. A.

Bridge

J. A.

(2024). Suicide in US preteens aged 8 to 12 years, 2001 to 2022. JAMA Network Open, 7(7), Article e2424664. https://doi.org/10.1001/jamanetworkopen.2024.24664

46.

Sheftall

A. H.

Asti

Horowitz

L. M.

Felts

Fontanella

C. A.

Campo

J. V.

Bridge

J. A.

(2016). Suicide in elementary school-aged children and early adolescents. Pediatrics, 138(4), Article e20160436. https://doi.org/10.1542/peds.2016-0436

47.

Sheftall

A. H.

Vakil

Ruch

D. A.

Boyd

R. C.

Lindsey

M. A.

Bridge

J. A.

(2022). Black youth suicide: Investigation of current trends and precipitating circumstances. Journal of the American Academy of Child & Adolescent Psychiatry, 61(5), 662–675. https://doi.org/10.1016/j.jaac.2021.08.021

48.

Spitzer

R. L.

Williams

J. B.

Gibbon

First

M. B.

(1992). The Structured Clinical Interview for DSM-III-R (SCID). I: History, rationale, and description. Archives of General Psychiatry, 49(8), 624–629. https://doi.org/10.1001/archpsyc.1992.01820080032005

49.

StataCorp. (2023). Stata Statistical Software: Release 18.

50.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and Family, 41(1), 75–88. https://doi.org/10.2307/351733

51.

Zelazny

Melhem

Porta

Biernesser

Keilp

J. G.

Mann

J. J.

Oquendo

M. A.

Stanley

Brent

D. A.

(2019). Childhood maltreatment, neuropsychological function and suicidal behavior. Journal of Child Psychology and Psychiatry, 60(10), 1085–1093. https://doi.org/10.1111/jcpp.13096

52.

Zelazny

Stanley

Porta

Mann

J. J.

Oquendo

Birmaher

Melhem

Brent

D. A.

(2021). Risk factors for pre-adolescent onset suicidal behavior in a high-risk youth sample. Journal of Affective Disorders, 290, 292–299.