Factors mediating pre-existing autism diagnosis and later suicidal thoughts and behaviors: A follow-up cohort study

Abstract

Individuals diagnosed with autism spectrum disorder (ASD) are more likely to have suicidal thoughts and behaviors. However, little research delineates the temporal and mechanistic associations between potential risk factors and suicidality in autistic individuals. We assessed 129 autistic and 121 age-matched and sex-assigned-at-birth-matched typically developing (TD) children and youth at baseline and follow-up. We evaluated internalizing/externalizing problems, bullying experiences, and three executive function tasks at a mean baseline age of 13.4 years (standard deviation: 4.2 years). After an average of 5.8 years (standard deviation: 1.7 years), we evaluated their suicidal thoughts and behaviors. The experiences of being bullied partially mediated the temporal relationship between pre-existing ASD and later-reported suicidal thoughts. Several cognitive flexibility and sustained attention indices mediated the temporal relationship between pre-existing ASD and later-reported suicidal thoughts. A multiple mediation model confirmed that bullying victimization and cognitive flexibility jointly mediated the temporal relationship between ASD and suicidal thoughts. Better cognitive flexibility and experiences of bullying victimization in autistic children/youth were both associated with higher risks of later-reported suicidal thoughts in adolescence/adulthood. School bullying-prevention programs and tailored clinical support for autistic youth with better cognitive flexibility are warranted for personalized suicide prevention for the autistic population.

Lay Abstract

Autistic people are more likely to experience suicidal thoughts and behaviors. The underlying relationships between potential risk factors and suicidal thoughts and behaviors in autistic individuals remain unclear. To understand this, we investigated whether specific factors in childhood/youth explain the effects of pre-existing autism spectrum disorder (ASD) diagnoses on later suicidal thoughts and behaviors in adolescence/adulthood. We assessed internalizing and externalizing problems, bullying experiences, and executive functions (including cognitive flexibility, sustained attention, and spatial working memory) at an average baseline age of 13.4 years and suicidal thoughts and behaviors at an average follow-up age of 19.2 years among 129 autistic and 121 typically developing (TD) individuals. During the follow-up period in adolescence/adulthood, autistic individuals were more likely to report suicidal thoughts than TD individuals. Being bullied partially accounted for the relationship between a pre-existing ASD diagnosis and later-reported higher suicidal thoughts. Contrary to our hypothesis, higher (instead of lower) cognitive flexibility in some autistic young people appeared to partially explain their higher rates of suicidal thoughts compared with typically developing young people. The findings imply that school bullying prevention and tailored intervention programs for autistic people, especially those with higher cognitive flexibility, are warranted to reduce their risks of experiencing suicidal thoughts.

Keywords

autism bullying cognitive flexibility executive functions suicide

Introduction

Suicidal thoughts and behaviors are important but under-investigated clinical challenges faced by autistic individuals or individuals diagnosed with autism spectrum disorder (ASD). Emerging population-based data show that suicide is one of the major causes of premature mortality in autistic individuals (Hirvikoski et al., 2016; Tsai et al., 2023). One of the earliest clinic-based studies reported that in diagnosis-seeking adults who received an autism diagnosis, 66% contemplated suicide in their lifetime, and 35% had planned or attempted suicide (Cassidy et al., 2014). The latest meta-analyses of 31 studies showed that an ASD diagnosis was associated with increased risks of self-injurious behaviors and suicidal thoughts and behaviors (i.e. suicidal ideation, suicide attempt, or suicide death) in children and adults, with a pooled odds ratio (OR) of over 3 (Blanchard et al., 2021). Another meta-analysis focusing on youth summarized 29 studies and found that 25.2% of autistic youth (⩽25 years of age) experienced suicidal ideation, and 8.3% had attempted suicide (O’Halloran et al., 2022). Despite these heightened suicidal risks in autistic individuals, there is relatively little research into the risk and protective mechanisms of suicidal thoughts and behaviors in autistic individuals to inform tailored prevention and intervention strategies (Cassidy et al., 2020; Cassidy & Rodgers, 2017).

The complexity of suicidality emerges from its associations with a wide range of contributing factors, involving a diverse interplay of multi-level factors (Chesney et al., 2014; Turecki et al., 2019). Theories that seek to explain suicidal behaviors encompass various elements, including biological, psychological, social, and environmental influences (O’Connor & Nock, 2014; Turecki et al., 2019). One of the theoretical approaches that applies to understanding suicidality in autistic individuals is the stress-diathesis model. This model suggests that suicidal behaviors are influenced by the interactions between acute stressors (or proximal factors) and underlying diathesis (or distal factors) (Mann & Rizk, 2020). The stressors can be internal, such as psychiatric symptoms or disorders, or external, such as relationship problems. The diathesis represents a collection of traits that shapes the likelihood of engaging in suicidal behaviors when confronted with stressors. Our study was informed by the stress-diathesis model, taking into account internal stressors (e.g., internalizing and externalizing symptoms), external stressors (e.g., bullying experiences), and diathesis (e.g., executive functions (EFs)) to investigate their potential mediating roles for the autism-suicidality association. We focused on these factors owing to their substantial clinical relevance and high prevalence, that mental health issues are common in autistic people (Lai et al., 2019), that autistic people frequently experience bullying victimization (Park et al., 2020), and that many autistic people have variable EF issues (Demetriou et al., 2018; Velikonja et al., 2019).

Internalizing/externalizing problems and suicidal risk

In typically developing (TD) populations, suicidal thoughts and behaviors were associated with internalizing symptoms (Nock et al., 2010). Externalizing problems such as aggression were also associated with suicidal risk in children and adolescents (Brent et al., 1996). One meta-analysis of 24 longitudinal studies showed that youth and young adults with internalizing or externalizing symptoms were more likely to attempt suicide in the future (Soto-Sanz et al., 2019). In a population-based study in the Netherlands, TD children with parent-reported suicidal ideation were more likely to have co-occurring internalizing and externalizing problems than children without suicidal ideation (Herba et al., 2007).

Population-based studies showed that autistic individuals had a higher incidence of suicidal thoughts and behaviors (Widnall et al., 2022), and this association could be partly explained by the presence of psychiatric conditions (Hand et al., 2020; Hirvikoski et al., 2020; Jokiranta-Olkoniemi et al., 2021; Kõlves et al., 2021). In the autistic population, co-occurring internalizing/externalizing problems, including anxiety, mood, and behavior disorders, were cross-sectionally associated with suicidal ideation (Dell’Osso et al., 2019; Hedley & Uljarević, 2018; Zahid & Upthegrove, 2017). Anxiety and depressive symptoms were associated with increased risks of suicidal ideation, plans, and attempts in autistic youth (Hedley & Uljarević, 2018; Horowitz et al., 2018; Licence et al., 2020; South et al., 2020). Past studies also found that externalizing and internalizing problems were associated with suicidal thoughts and behaviors not only in autistic individuals (Hedley & Uljarević, 2018) but also in TD population (Commisso et al., 2023). However, little is known about the temporal relationships between internalizing/externalizing problems and later suicidal thoughts and behaviors in autistic people.

Bullying experiences and suicidal risk

Being bullied (i.e. peer victimization) is a crucial risk factor for suicide in young people (Cha et al., 2018; Cuesta et al., 2021). Longitudinal studies in the general population demonstrate the impact of being bullied during childhood and early adolescence on later suicidal thoughts and behaviors (Geoffroy et al., 2016; Winsper et al., 2012). Compared with the TD population, autistic individuals were more vulnerable to bullying victimization; more than half of autistic individuals reported such experiences (Fink et al., 2018; Maiano et al., 2016; Park et al., 2020). Autistic children and adolescents who were victims of school bullying were three times more likely to have suicidal ideation or attempts than those who were not (Mayes et al., 2013). Again, most of these studies are cross-sectional, and the temporal relationships between bullying victimization and suicidal thoughts and behaviors are unclear in the autistic population.

Executive functions and suicidal risk

Cognitive factors additionally play significant roles (i.e. as components of the diathesis) in the emergence and processing of suicidal thoughts and behaviors (Allen et al., 2019; Cha et al., 2018; Fazel & Runeson, 2020; Lengvenyte et al., 2021; Saffer & Klonsky, 2018). General population-based studies showed that low IQ might be a risk factor for suicidal thoughts and behaviors (Batty et al., 2010; Sörberg Wallin et al., 2018). However, such an association has not been consistently observed in cross-sectional studies in the autistic population. For example, some found higher IQ to be correlated with increased suicidal risks (Hedley & Uljarević, 2018; Hirvikoski et al., 2020; Hirvikoski et al., 2016). In contrast, others found no association between intellectual disability (ID) and suicidal ideation (Horowitz et al., 2018) or attempts (M.-H. Chen et al., 2017). In the general population, challenges in EF, including memory, planning, cognitive ﬂexibility, and attention control, were associated with suicidal risk (Bredemeier & Miller, 2015; Fernández-Sevillano et al., 2021; Hedley et al., 2021; Keilp et al., 2013). However, the directions of association were mixed, which may be due to the heterogeneous study populations (Fazel & Runeson, 2020). Autistic individuals tend to experience substantial EF challenges (Demetriou et al., 2018; Velikonja et al., 2019), which often persist into adulthood (Xie et al., 2020). Specifically, some autistic individuals may experience rigid thinking and inflexibility, which may increase the risk of suicidal thoughts and behaviors (Clemence & Lewis, 2018; MacPherson, Kim, et al., 2022; Novak et al., 2022). However, which specific cognitive component(s) are potential risk or protective factors of suicidal thoughts and behaviors in autistic people remains unknown. Identifying the cognitive diathesis contributing to suicidal thoughts and behaviors that autistic people uniquely experience will inform tailored prevention and intervention strategies.

Study aims

Previous studies indicate that internalizing/externalizing problems, bullying victimization, and specific EF components were associated with suicidal thoughts and behaviors in autistic people (Dell’Osso et al., 2019; Hedley & Uljarević, 2018; Jokiranta-Olkoniemi et al., 2021; Novak et al., 2022). For example, emotional problems, such as depression, might mediate the relationship between autism and suicidality (Y. Y. Chen et al., 2020; Hedley et al., 2018). There is still a dearth of research investigating the specific mediating roles of risk factors in the association between autism and suicidality based on the stress-diathesis model. To bridge this knowledge gap, we leveraged the data collected in a clinical cohort study in Taiwan to investigate whether childhood/youth internalizing/externalizing problems, experiences of bullying victimization, and specific EF components mediated the association between pre-existing ASD diagnoses and later-reported suicidal thoughts and behaviors in adolescence/adulthood. We hypothesized that childhood/youth ASD diagnoses would be associated with developing suicidal thoughts and behaviors later in adolescence/adulthood. Such an association could be mediated by increased childhood/youth internalizing/externalizing problems, bullying victimization experiences, and EF difficulties, especially reduced cognitive inflexibility.

Methods

Participants

Autistic individuals were clinically diagnosed with ASD-equivalent diagnoses by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR); the diagnoses were further asserted by interviewing their parents using the Chinese version of the Autism Diagnostic Interview-Revised (ADI-R) (Gau et al., 2011) to confirm the ASD diagnosis and the Chinese version of the Kiddie-Schedule for Affective Disorders and Schizophrenia for School-Age Children–Epidemiological Version (K-SADS-E) interview for other psychiatric diagnoses (M.-H. Chen et al., 2017; Gau et al., 2005) at baseline and follow-up; and the participants received the Autism Diagnostic Observation Schedule assessment at the follow-up, by experienced interviewers who had received 1-year full-time training. Participants were excluded from entering the study cohort if they had categorical diagnoses of schizophrenia, mood disorders, current anxiety disorders, or substance-use disorders at baseline assessment. Descriptions of the diagnostic instruments are provided in Supplemental Material, available online. The autistic participants were enrolled in the outpatient clinics of the National Taiwan University Hospital, a tertiary medical center in Northern Taiwan, during 2009–2015 for baseline assessments and received follow-up assessments during 2015–2020.

We recruited TD individuals from the same neighborhoods where the autistic individuals resided. The TD individuals had no lifetime or current DSM-IV-TR psychiatric disorder at baseline or follow-up. The data were extracted from a larger prospective clinical cohort study (Chang et al., 2022; Hsiao et al., 2022), which was not explicitly designed to study suicidal thoughts and behaviors in autistic individuals.

Procedures

All participants were assessed by the Wechsler Intelligence Scale for Children, Third Edition (age < 16 years) or Wechsler Adult Intelligence Scale, Third Edition (age ⩾ 16 years) at baseline, and those with a full-scale intelligence quotient (FSIQ) < 70 were excluded. Written informed consent was obtained from all participants and their parents after a detailed explanation of the study’s purpose and procedures. We included 129 autistic children and youth and 121 age- and sex-matched TD individuals (mean age across all individuals, 13.4 years) at baseline. All autistic and TD participants were re-assessed at a follow-up timepoint, with a mean follow-up interval of 5.8 years (ranging from 1.5 to 12 years), at the mean age of 19.2 years. The study was approved by the Research Ethics Committee of the National Taiwan University Hospital (approval number: 201403109RINC).

Baseline assessments

Autistic symptoms

The Social Communication Questionnaire (SCQ) is a parent-report questionnaire with 40 yes/no items to examine reciprocal social interaction, communication, and restricted, repetitive, and stereotyped interests or behaviors associated with autism. Western Psychological Services approved the Chinese SCQ in 2007 (Gau et al., 2011) with culture-relevant expressions. It has good reliability and validity (Gau et al., 2011) and is widely used in Taiwan for autism research (P. I. Lin et al., 2021; Lo et al., 2017). We used the SCQ total score as an overall index of autistic symptoms for both groups.

Internalizing/externalizing problems

The Child Behavior Checklist (CBCL) is a 118-item parent-reported questionnaire assessing behavioral and emotional problems of children and adolescents aged 6–18 years (Achenbach, 2001). Each item is scored from 0 to 2 (0 if not true, 1 if somewhat or sometimes true, and 2 if very or often true). The CBCL gives two broad-band subscales representing internalizing problems (i.e. anxious/depressed, withdrawn, and somatic complaints) and externalizing problems (i.e. rule-breaking and aggressive behaviors). The Chinese CBCL has good psychometric properties (Shang et al., 2006) and has been widely used to evaluate the emotional and behavioral problems of Taiwanese children and adolescents (Chang et al., 2020; Chiang & Gau, 2016). We used internalizing and externalizing problems T-scores to measure the psychiatric symptoms at baseline.

Bullying experiences

Two items from the parent-reported problems with peers domain of the Social Adjustment Inventory for Children and Adolescents (SAICA) (John et al., 1987) were used to present the experiences of bullying victimization (“Being teased/bullied by other kids”) or perpetration (“Bullying other kids”) (Hsiao et al., 2022). The presence of victimization or perpetration, separately, was rated as 1 (not a problem), 2 (mild problem), 3 (moderate problem), or 4 (severe problem). The Chinese SAICA has satisfactory reliability and validity (Gau et al., 2006) and is used extensively in Taiwan (e.g., Chiang et al., 2018; Y. J. Lin et al., 2021; Shang et al., 2020).

Executive functions

We used three tasks from the Cambridge Neuropsychological Test Automated Battery (CANTAB).

Cognitive flexibility

The intra-dimensional/extradimensional shifts (IED) task measures rule acquisition and shifting (i.e., cognitive flexibility) in terms of one’s ability to categorize stimuli into sets (intra-dimensional shift) and then shift attention to changes in stimuli (extradimensional shift, EDS) (Luciana & Nelson, 1998). Participants get through nine stages by scoring six consecutive correct responses at each stage. We used two indices to measure cognitive flexibility: (1) errors made before the EDS stage (pre-EDS errors) and (2) errors made in the EDS stage (EDS errors). Here, a higher score indicates less cognitive flexibility.

Sustained attention

The Rapid Visual Information Processing (RVIP) task assesses sustained attention, modified from a 4-minute visual continuous performance test and simplified from Wesnes & Warburton’s task. Digits, ranging from 2 to 9, appear one at a time in random order at 100 digits/min. When the number appears in a target sequence (e.g., 3-5-7, 2-4-6), the participants must respond as quickly as possible. We used two indices to measure visual sustained attention: (1) probability of hits, i.e., the total number of hits (the number of occasions upon which the target sequence is correctly responded to) divided by the sum of total hits and total misses; and (2) A′, a signal detection measure of sensitivity to the target, regardless of response tendency, ranging from 0 to 1, where a higher score indicates higher sensitivity (Sahgal, 1987).

Spatial working memory

The spatial working memory (SWM) task is constructed based on a self-ordered search test (Petrides & Milner, 1982) to evaluate one’s ability to plan, retain spatial information, and manipulate the information in working memory. It is adapted from Olton’s radial arm maze task (Olton, 1987). This task requires participants to search through boxes on the screen to find the hidden blue tokens. Only a single token is hidden inside one of the boxes in each trial. The participants are asked to memorize the boxes opened in each trial and the boxes in which the tokens were found in the previous trial. There are three difficulty levels (i.e., four-, six-, and eight-box levels). We used two indices to measure SWM: (1) strategy utilization, i.e., the number of search sequences starting with a novel box in the difficult problems, and (2) total errors, i.e., the sum of errors in searches across all three difficulty levels. Here, a higher score indicates a poorer SWM.

Follow-up assessments

Suicidal thoughts and behaviors

All participants received K-SADS-E interviews again at follow-up. Suicidal thoughts and behaviors (i.e., suicidal thoughts, plans, and attempts) were assessed using the K-SADS-E interview. The presence of suicidal thoughts was defined by answering “yes” to the question, “Have you ever thought of ending your own life in the past 6 months?” If the participant responded positively, the interviewer would then further ask about suicidal plans. The presence of suicidal plans was defined by answering “yes” to the follow-up question, “Have you ever planned suicide in the past 6 months?” If the participant responded positively, the interviewer would further gather the content of the suicide plans and ask, “Have you ever attempted suicide in the past 6 months?” A “yes” response was coded as positive for suicide attempts. Based on the K-SADS-E interview schedule, individuals who reported no suicidal thoughts would be coded to have no suicidal plans or attempts. In other words, those who reported suicidal plans or attempts were a subgroup of those who reported suicidal thoughts. Regarding the safety protocol, when positive suicidal risk indicators were identified, the study team led by the corresponding author (S.S.-F.G.) would immediately assess whether the participant was currently experiencing them and take appropriate actions to ensure their safety. Such efforts include informing their caregivers, creating a proper safety plan, and providing the participants with information on accessible clinical and emergency services, if necessary at National Taiwan University Hospital, Taipei, Taiwan.

Statistical analysis

We used independent sample t-tests to examine the differences between the autistic and TD groups in age, IQ, internalizing/externalizing problems, bullying victimization, and EF (i.e., pre-EDS errors and EDS errors in the IED task; probability of hit and A′ in the RVIP task; strategy utilization; and total errors in the SWM task) at baseline. Group differences in suicidal thoughts and behaviors at follow-up were examined using chi-square tests. For mediation analyses, we calculated mediation effects, or indirect effects, according to the suggestions by Preacher and Hayes (2008). Figure 1 presents the framework of the simple mediation models, where paths a and b represent the indirect effects of the baseline internalizing/externalizing problems, bullying victimization, and EF (i.e., pre-EDS errors and EDS errors in the IED task; probability of hit and A’ in the RVIP task; strategy utilization and total errors in the SWM task) on the link between pre-existing ASD diagnosis and follow-up suicidal thoughts and behaviors. We tested the indirect effects (i.e., path ab, Figure 1) and assessed their uncertainties by generating the 95% confidence intervals (CI) using 5000 bootstrap samples; we also estimated the single-step indirect effects (i.e., path a and path b). We followed the suggestions of MacKinnon et al. (2007) that we only interpreted the findings where paths a and b were both significant because this, when coupled with significant ab (i.e., the 95% CI) of the indirect effect, did not include zero), suggested stronger evidence for mediation effects than significant ab alone. To avoid biased parameter estimates and to consider the effects of multiple mediators simultaneously, we further ran a multiple mediation model (Figure 2). A bootstrapping procedure was also used to test the significance of the indirect effects with 5000 bootstrap samples (Preacher & Hayes, 2008). Participants’ baseline age, follow-up interval, and sex assigned at birth were covariates included in the mediation analyses. Finally, to explore the potential influence of intellectual capability on the mediation effects of EF, we conducted sensitivity analyses by including FSIQ as a moderator in the mediation models (Hayes, 2017) (Supplemental Figure S1). All analyses were conducted using SPSS 25.0 (IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0; IBM Corp., Armonk, NY).

Figure 1.

Conceptual representation of the mediation models in a regression framework examih significant ab (i.e., the 95% C the mediating roles of internalizing/externalizing problems, bullying victimization, and specific executive function components between pre-existing autism spectrum disorder (ASD) diagnostic status and later-reported suicidal thoughts and behaviors. ASD was diagnosed before baseline assessment and coded as 0 (typically developing) or 1 (ASD). Suicidal thoughts and behaviors were coded as 0 (absent) or 1 (present). Control variables: age at baseline, follow-up interval, and sex assigned at birth. a = correlation between ASD diagnostic status and internalizing/externalizing problems, bullying victimization, or individual executive functions; b = correlation between internalizing/externalizing problems, bullying victimization, or individual executive functions and suicidal thoughts and behaviors, independent of ASD diagnostic status; c = total (direct and indirect) effect from ASD diagnostic status to suicidal thoughts and behaviors; c’ = direct effect from ASD diagnostic status to suicidal thoughts and behaviors.

Figure 2.

The multiple mediation models.

Community involvement

Community members were not involved in the design and implementation of the study.

Results

Sample characteristics

There were no differences in the baseline and follow-up age, follow-up interval, and sex assigned at birth distribution between the ASD (n = 129) and TD (n = 121) groups (Table 1) except for lower IQ scores in ASD than TD at baseline. More autistic individuals (n = 37, 28.7%) reported suicidal thoughts and behaviors than did TD individuals (n = 8, 6.6%) at follow-up, with an OR of 5.68 (95% CI (2.52, 12.80), p < 0.001). At baseline, autistic individuals had more internalizing and externalizing problems, were more frequently bullied, and performed worse on IED and SWM metrics than TD individuals.

Table 1.

Demographics, IQ profile, internalizing/externalizing problems, bullying victimization, specific executive function components, and suicidal thoughts and behaviors of autistic and typically developing individuals.

	ASD (n = 129)	TD (n = 121)	Effect size (Cohen’s d or odds ratio)	p Value	FDR-adjusted p value
Age (years) at baseline (childhood/youth) (range)	13.71 ± 4.04 (8–27)	13.07 ± 4.26 (7–29)	0.16	0.219	–
Age (years) at follow-up (adolescence/adulthood) (range)	19.60 ± 4.18 (12.6–34.0)	18.73 ± 4.28 (12.5–34.9)	0.21	0.103	–
Follow-up interval (years)	5.89 ± 2.04 (1.5–12.0)	5.66 ± 1.37 (3.3–9.7)	0.13	0.295	–
Sex assigned at birth (male/female) (n/%)	120/9 (93.0/7.0)	113/8 (93.4/6.6)		0.909	–
SCQ total score at baseline ^a	17.92 (7.06)	4.17 (2.70)	2.57	<0.001	–
Assessment at baseline
Intelligence quotient (IQ)
Verbal IQ	102.4 (18.7)	111.7 (10.4)	0.61	<0.001	<0.001^h
Performance IQ	101.5 (19.6)	108.9 (13.1)	0.44	<0.001	<0.001^h
Full-scale IQ	101.6 (19.1)	111.2 (10.7)	0.62	<0.001	<0.001^h
CBCL^b
Internalizing problems	19.05 ± 12.82	5.89 ± 6.15	1.31	<0.001	<0.001ⁱ
Externalizing problems	15.54 ± 11.09	4.82 ± 5.43	1.23	<0.001	<0.001ⁱ
SAICA—bullying experiences^c
Being teased/bullied by other kids	1.96 ± 0.89	1.11 ± 0.32	1.27	<0.001	<0.001^j
Bullying other kids	1.16 ± 0.45	1.03 ± 0.20	0.37	0.008	0.008^j
Intradimensional/extradimensional shifts^d
Pre-extradimensional shift errors	8.63 ± 5.80	7.09 ± 3.31	0.33	0.011	0.022^k
Extradimensional shift errors	9.82 ± 9.58	7.18 ± 7.90	0.30	0.019	0.029^k
Rapid Visual Information Processing^d
Probability of hit	0.55 ± 0.19	0.58 ± 0.21	0.15	0.219	0.219^k
A’	0.87 ± 0.06	0.89 ± 0.06	0.33	0.133	0.160^k
Spatial working memory^e
Strategy utilization	34.74 ± 6.60	31.56 ± 5.13	0.54	<0.001	<0.001^k
Total errors	36.02 ± 21.15	19.38 ± 17.14	0.86	<0.001	<0.001^k
Assessment at follow-up: suicidal thoughts and behaviors^f			OR (95% CI)
Thoughts	37	8	5.68 (2.52–12.80)	<0.001^g	–
Plans	7	5	1.33 (0.41–4.31)	1.000^g	–
Attempts	6	2	2.90 (0.57–14.67)	0.283^g	–

Sample size: ASD/TD = 127/78.

Sample size: ASD/TD = 127/121.

Sample size: ASD/TD = 89/114.

Sample size: ASD/TD = 126/120.

Sample size: ASD/TD = 129/120.

Data from Kiddie-Schedule for Affective Disorders and Schizophrenia Epidemiological version. Individuals endorsing suicidal plans or attempts were subsets of those reporting suicidal thoughts.

Fisher’s exact test.

The FDR adjustment was applied for multiple comparisons across three intelligence quotient measures.

The FDR adjustment was applied for multiple comparisons across two CBCL measures.

The FDR adjustment was applied for multiple comparisons across two SAICA-bullying experiences measures.

The FDR adjustment was applied for multiple comparisons across six executive function measures.

ASD: autism spectrum disorder; TD: typically developing; CBCL: Child Behavior Checklist; SAICA: Social Adjustment Inventory for Children and Adolescents; SCQ: Social Communication Questionnaire; OR: odds ratio; CI: confidence interval; FDR: false discovery rate.

Mediators between ASD diagnostic status and suicidal thoughts

We ran mediation analyses for suicidal thoughts but not plans/attempts as there was an ASD-TD group difference on suicidal thoughts but not on plans/attempts. Table 2 presents the indirect effects of pre-existing ASD diagnosis on suicidal thoughts reported at follow-up through internalizing/externalizing problems, bullying victimization, and specific EF components in simple mediation models. Internalizing/externalizing problems and bullying perpetration were not significant mediators. However, bullying victimization partially mediated the effects of pre-existing ASD on suicidal thoughts, with significant indirect (i.e., path ab, Figure 1) and direct effects (i.e., path c, Figure 1) in the same direction. Children/youth who experienced more bullying victimization reported more suicidal thoughts later in their adolescence/adulthood, significantly explaining the associations between pre-existing ASD diagnoses and later-reported suicidal thoughts.

Table 2.

Indirect effects of pre-existing ASD diagnostic status on suicidal thoughts at follow-up through internalizing/externalizing problems, bullying victimization, and specific executive function components at baseline.

	Unstandardized estimates (SE)				Indirect effects
Mediators	a	b	c	c’	ab	Bootstrap ab	SE	95% CI
CBCL
Internalizing problems	12.53 (1.43)***	0.01 (0.02)	1.69 (0.48)***	1.52 (0.52)**	0.163	0.163	0.199	[–0.219, 0.572]
Externalizing problems	10.92 (1.26)***	0.01 (0.02)	1.69 (0.48)***	1.59 (0.52)**	0.098	0.098	0.204	[–0.307, 0.509]
SAICA—bullying experiences
Being teased/bullied by other kids	0.82 (0.09)***	0.70 (0.28)*	1.73 (0.44)***	1.12 (0.51)*	0.574	0.574	0.253	[0.116, 1.113]
Bullying other kids	0.14 (0.05)**	−0.50 (0.65)	1.73 (0.44)***	1.79 (0.45)***	−0.073	−0.073	0.350	[–1.313, 0.077]
Executive function
Intradimensional/extradimensional shifts
Pre-extradimensional shift errors	1.68 (0.61)**	−0.16 (0.07)*	1.68 (0.42)***	1.89 (0.44)***	–0.269	–0.269	0.157	[–0.704, –0.063]
Extradimensional shift errors	2.84 (1.11)*	−0.03 (0.02)	1.68 (0.42)***	1.77 (0.43)***	−0.087	−0.087	0.076	[–0.302, 0.014]
Rapid Visual Information Processing
Probability of hit	−0.05 (0.02)*	2.60 (1.16)*	1.68 (0.42)***	1.88 (0.44)***	–0.136	–0.136	0.097	[–0.403, –0.009]
A’ (signal detection sensitivity)	−0.02 (0.01)**	9.61 (3.94)*	1.68 (0.42)***	1.92 (0.44)***	–0.170	–0.170	0.111	[–0.468, –0.023]
Spatial working memory
Strategy utilization	3.35 (0.75) ***	−0.01 (0.03)	1.65 (0.42)***	1.70 (0.44)***	−0.046	−0.046	0.107	[–0.335, 0.107]
Total errors	17.70 (2.38)***	−0.01 (0.01)	1.65 (0.42)***	1.92 (0.46)***	−0.260	−0.260	0.187	[–0.676, 0.067]

Covaried with baseline age, follow-up interval, and sex assigned at birth, with 5000 bootstrap samples. SE: standard error; CI: confidence interval; ASD, autism spectrum disorder; CBCL, Child Behavior Checklist; SAICA, Social Adjustment Inventory for Children and Adolescents. Bolded text indicates significant indirect effects of pre-existing ASD diagnostic status on suicidal thoughts at follow-up.

p < 0.05; **p < 0.01; ***p < 0.001.

We found that pre-EDS errors in IED and the probability of hit and A’ in RVIP mediated the effects of pre-existing ASD diagnosis on suicidal thoughts, while controlling for baseline age, follow-up interval, and sex assigned at birth. Surprisingly, indirect path ab and direct path c were opposite signs, indicating competitive mediations (i.e. indirect and direct effects both exist and point in opposite directions) (Zhao et al., 2010). Children/youth with better cognitive flexibility and attentional control experienced more suicidal thoughts later in their adolescence/adulthood, significantly explaining the associations between pre-existing ASD diagnoses and later-reported suicidal thoughts.

In the multiple mediation models, we found that IED pre-EDS errors and bullying victimization mediated the effects of pre-existing ASD diagnoses on later-reported suicidal thoughts, controlling for baseline age, follow-up interval, and sex assigned at birth (Table 3 and Figure 2). That is, when considering significant mediators shown in the simple mediation models altogether, the fact that children/youth who experienced more bullying victimization and had better cognitive flexibility reported more suicidal thoughts later in their adolescence/adulthood significantly explained the associations between pre-existing ASD diagnoses and later-reported suicidal thoughts. We added FSIQ as a moderator in the simple mediation models for EFs (Supplemental Figure S1), and FSIQ did not moderate the mediation effects (Supplemental Table S1).

Table 3.

Indirect effects of pre-existing ASD diagnostic status on suicidal thoughts at follow-up through a multiple mediation model.

	Unstandardized estimates (SE)				Indirect effects
Mediators	a	b	c	c’	ab	Bootstrap ab	SE	95% CI
Being teased/bullied by other kids	0.85 (0.09)***	0.66 (0.29)*	1.78^a (0.44)***	1.57^a (0.55)**	0.559	0.559	0.284	[0.036, 1.156]
Pre-extradimensional shift errors	2.27 (0.71)**	−0.13 (0.07)*			−0.295	−0.295	0.223	[–0.866, –0.031]
RVIP—Probability of hit	−0.04 (0.02)	−0.18 (5.18)			0.007	0.007	0.415	[–0.499, 0.950]
RVIP—A’ (signal-detection sensitivity)	−0.02 (0.01)*	10.65 (17.75)			−0.161	−0.161	0.585	[–1.843, 0.230]

Covaried with baseline age, follow-up interval, and sex assigned at birth, with 5000 bootstrap samples. SE: standard error; CI: confidence interval; ASD, autism spectrum disorder; RVIP: Rapid Visual Information Processing. Bolded text indicates significant indirect effects of pre-existing ASD diagnostic status on suicidal thoughts at follow-up.

The values of the c and c’ represent the total (direct and indirect) and direct effects in the overall multiple mediation model.

p < 0.05; **p < 0.01; ***p < 0.001.

Discussion

In a Taiwanese clinical cohort first assessed at baseline in childhood/youth and followed up on average 5.8 years later in adolescence/adulthood, we found more frequent suicidal thoughts (but not plans or attempts) reported in adolescence/adulthood in autistic individuals than in TD individuals. We did not find that baseline internalizing/externalizing problems mediated the association between pre-existing ASD diagnoses and later-reported suicidal thoughts. In contrast, the experiences of being bullied partially mediated the ASD-suicidal thoughts temporal association. Contrary to our hypotheses, better rather than poorer cognitive flexibility was associated with a higher risk of suicidal thoughts, and such a relationship mediated the ASD-suicidal thoughts association. The mediation effects of bullying victimization and cognitive flexibility remained significant when considering multiple mediators altogether, and FSIQ did not moderate these mediation effects.

Our finding of more autistic adolescents/adults reporting suicidal thoughts than TD adolescents/adults echoes previous cross-sectional findings across the globe (Blanchard et al., 2021) and a national registry study in Taiwan showing increased suicidal attempts in autistic compared with non-autistic youth (M.-H. Chen et al., 2017). The absence of a statistically significant difference between autistic and TD individuals regarding suicidal plans or attempts in our study could be partly attributable to potential under-reporting in self-reports (Howe et al., 2020) and/or the sample size and statistical power limitation. Contrary to our hypotheses, internalizing and externalizing problems did not mediate the relations between pre-existing ASD and later-reported suicidal thoughts. Several cross-sectional studies reported that externalizing problems in autistic people were associated with suicidal thoughts and behaviors (Hedley & Uljarević, 2018; Licence et al., 2020). A Canadian longitudinal study of autistic children also showed that externalizing problems at age 7 years were associated with suicidal ideation or self-injurious behavior at any time point. Still, internalizing problems were not associated with suicidal ideation or attempts in autistic children (Hunsche et al., 2020). A general population longitudinal study in Quebec, Canada, showed that childhood externalizing and internalizing symptoms were related to later suicide attempts (Commisso et al., 2023). The inconsistency of our findings from previous studies regarding externalizing problems may be due to different study designs, statistical power, or other socio-cultural factors. However, the consistent findings imply that while rates of internalizing symptoms and suicidal thoughts and behaviors are reportedly elevated in autistic individuals, they may not always co-occur (Hunsche et al., 2020).

In the present clinical cohort, experiences of being bullied in childhood/youth significantly mediated the effect of pre-existing ASD on later-reported suicidal thoughts. Similarly, a retrospective cohort study on the longitudinal association between experiences of being bullied and suicidal thoughts and behaviors also showed that autistic adolescents who experienced bullying victimization at first clinical contact showed a nearly doubled risk of suicidal thoughts and behaviors over the follow-up period (Holden et al., 2020). A recent systemic review also concludes that adverse childhood experiences, including bullying experiences, are important risk factors for suicidal thoughts and behaviors in autistic youth (O’Halloran et al., 2022). There is an urgent need to prevent bullying victimization and support victims of autistic people; these prevention and intervention efforts should be the responsibility of all school community members, including students, teachers, and school personnel (Park et al., 2020). Our findings implicate the temporal or mechanistic associations between bullying victimization and later suicidal thoughts, potentially explaining why autistic people suffer from increased suicidal risk. This finding adds further impetus to improve recognition and timely intervention for autistic young people who experience bullying victimization.

To our knowledge, this is the first study investigating whether specific EF components mediate the association between ASD diagnoses and suicidal thoughts and behaviors. Cognitive flexibility and attention control significantly mediated this association, and cognitive flexibility remained significant in the multiple mediation analysis. Contrary to our hypotheses, EF showed a competitive mediation effect: that is, what potentially explained the higher suicidal thoughts in autistic than non-autistic young people seemed to be the higher (instead of lower) cognitive flexibility in some of them. This unique finding partially echoes retrospective clinical research showing significantly lower reported suicidal thoughts and behaviors in autistic young people with IDs than in autistic individuals without ID (Holden et al., 2020). Population-based data also indicate that autistic individuals with ID are less likely to have suicide attempts than autistic individuals without ID (Hirvikoski et al., 2020; Jokiranta-Olkoniemi et al., 2021; Kõlves et al., 2021). Moreover, autistic individuals without ID are more likely to die by suicide than autistic individuals with ID (Hirvikoski et al., 2016; Tsai et al., 2023). Similarly, cognitive ability was not found to be protective against the risk of suicide attempts in the autistic population (Mayes et al., 2013). These findings mirror the observation that autistic adults with better cognitive flexibility and social cognition have more negative self-concepts and lower self-esteem (Zimmerman et al., 2017). Autistic individuals with higher cognitive abilities may be more prone to attribute social challenges to their own abilities and efforts rather than external causes, which may fuel more critical self-appraisal and poorer mental health (Barnhill, 2001; Sterling et al., 2008). Uniquely, these findings are in contrast with those in TD people, that large population-based studies show a lower level of intelligence/cognitive ability being associated with increased suicidal risk (Andersson et al., 2008; Batty et al., 2010; Hedley et al., 2021; Sörberg Wallin et al., 2018). In hospitalized individuals, adolescents with suicidal attempts show lower set-shifting/cognitive flexibility than adolescents without suicidal thoughts and behaviors; however, there are no differences in cognitive flexibility between adolescents with suicidal ideation and adolescents without suicidal thoughts and behaviors (MacPherson, Kim, et al., 2022). Studies on individuals with bipolar disorder (MacPherson, Kudinova, et al., 2022) and psychiatric inpatients (Novak et al., 2022) similarly found that higher cognitive flexibility was associated with lower suicidal thoughts and behaviors. Our findings run against a prevailing view that developmental challenges in childhood act cumulatively to shape one’s diathesis to predict further adversity (Kraemer et al., 2005), suggesting that the mechanistic relations between EF and suicide risk may vary across diagnostic categories and operate in non-linear manners.

This study is unique as it is the first to use a follow-up design to examine the temporal relationships between internalizing/externalizing problems, bullying victimization, specific EF components, and suicidal thoughts and behaviors in autistic individuals. Understanding the inter-relations between childhood/youth risk factors and suicidal thoughts and behaviors in adolescence/adulthood may offer mechanistic insights to develop tailored supports for autistic young people to navigate the challenges of living in neurotypical contexts, especially in their transition from childhood to adulthood.

Nevertheless, several limitations should be noted. First, suicidal thoughts and behaviors were only ascertained at follow-up. Suicidal thoughts and behaviors usually start around the age of 10–15 years (Borges et al., 2008; Lee et al., 2007). In contrast, the onset of autism is much earlier (Tan et al., 2021), circumventing reverse causality. Second, our assessment for suicidal thoughts and behaviors was not specifically designed for autistic individuals. So far, no suicide assessment tool has been adequately validated in autistic young people, and there is a crucial need to develop validated tools for measuring suicidality in this group (Howe et al., 2020). Autistic individuals may under-report their suicidal ideations or behaviors when using measures designed for the general population, particularly self-reported measures, due to challenges related to social communication, EF, self-awareness of mental states, and emotional regulation (Howe et al., 2020; Roos et al., 2013). Nevertheless, K-SADS-E is widely used to evaluate suicidal risk across the general population (Gili et al., 2019). Third, clinical ascertainment bias may contribute to the small number of females in the cohort, which limited sex-stratified/focused investigations due to insufficient statistical power. Female sex assigned at birth was found to be a risk factor for increased psychiatric problems (Martini et al., 2022) and suicidal thoughts and behaviors in autistic individuals (Holden et al., 2020; Kõlves et al., 2021), although this was not found consistently in other studies (Horowitz et al., 2018; Storch et al., 2013). The moderating roles of sex assigned at birth and gender identity in suicide risks in autistic people await better-powered investigations in the future. Sub-optimally, we treated sex assigned at birth as a covariate in the mediation models. Fourth, the autistic group had an on-average lower IQ at baseline than the TD group. We performed sensitivity analyses, including FSIQ as a moderator in the mediation models, which showed no impact on the significant findings compared with the main results. Fifth, it is important to acknowledge that other risk factors not included in this study, such as loneliness and social support (Hedley et al., 2018), may also contribute to suicidality in autistic individuals (Mournet et al., 2023). However, measures of these factors were not available in the study dataset. Further longitudinal research is needed to investigate their potential roles as mediators or moderators of suicidality. Finally, because of the strict inclusion/exclusion criteria regarding psychiatric diagnoses and the exclusion of people with ID in the samples when the cohort was set up >15 years ago, we must exercise caution in generalizing the findings to autistic people at large, including those diagnosed with anxiety, mood, psychotic, or substance-use disorders, given the high prevalence of psychiatric diagnoses (Lai et al., 2019), and especially to those with ID as different mechanisms for suicidal risk and protection may be at play.

Conclusion

In a clinical follow-up cohort, bullying victimization and cognitive flexibility in childhood/youth mediated the effect of pre-existing ASD diagnoses on increased suicidal thoughts reported later in adolescence/adulthood. A better mechanistic understanding of the roles of internalizing/externalizing problems, bullying victimization, and specific EF components on the associations between ASD diagnoses and suicidal thoughts and behaviors will identify key avenues for tailored prevention and personalized intervention for the autistic population. School bullying-prevention programs and personalized clinical support for autistic individuals with higher EF are warranted.

Supplemental Material

sj-docx-1-aut-10.1177_13623613231223626 – Supplemental material for Factors mediating pre-existing autism diagnosis and later suicidal thoughts and behaviors: A follow-up cohort study

Supplemental material, sj-docx-1-aut-10.1177_13623613231223626 for Factors mediating pre-existing autism diagnosis and later suicidal thoughts and behaviors: A follow-up cohort study by Jung-Chi Chang, Meng-Chuan Lai, Shu-Sen Chang and Susan Shur-Fen Gau in Autism

Footnotes

Acknowledgements

The authors are grateful to all the participants and their families who participated in this study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by National Health Research Institute, Taiwan (NHRI-EX104-10404PI, NHRI-EX105-10404PI, NHRI-EX106-10404PI, NHRI-EX107-10404PI, NHRI-EX108-10404PI) to S.S.-F. Gau. M.-C. Lai is supported by the Academic Scholars Award from the Department of Psychiatry, University of Toronto, the Canadian Institutes of Health Research (CIHR Sex and Gender Science Chair, GSB 171373, and project grant, PJT 173351), the Innovation Fund of the Alternative Funding Plan for the Academic Health Sciences Centers of Ontario (CAM-20-004, CAM-20-005), and the CAMH Foundation.

ORCID iD

Susan Shur-Fen Gau

Supplemental material

Supplemental material for this article is available online.

References

Achenbach

T. M.

(2001). Manual for ASEBA school-age forms & profiles. University of Vermont, Research Center for Children, Youth & Families.

Allen

K. J. D.

Bozzay

M. L.

Edenbaum

E. R.

(2019). Neurocognition and suicide risk in adults. Current Behavioral Neuroscience Reports, 6(4), 151–165. https://doi.org/10.1007/s40473-019-00189-y

Andersson

Allebeck

Gustafsson

J. E.

Gunnell

(2008). Association of IQ scores and school achievement with suicide in a 40-year follow-up of a Swedish cohort. Acta Psychiatrica Scandinavica, 118(2), 99–105.

Barnhill

G. P.

(2001). Social attributions and depression in adolescents with Asperger syndrome. Focus on Autism and Other Developmental Disabilities, 16(1), 46–53. https://doi.org/10.1177/108835760101600112

Batty

G. D.

Whitley

Deary

I. J.

Gale

C. R.

Tynelius

Rasmussen

(2010). Psychosis alters association between IQ and future risk of attempted suicide: Cohort study of 1,109,475 Swedish men. BMJ, 340, Article c2506.

Blanchard

Chihuri

DiGuiseppi

C. G.

(2021). Risk of self-harm in children and adults with autism spectrum disorder: A systematic review and meta-analysis. JAMA Network Open, 4(10), Article e2130272. https://doi.org/10.1001/jamanetworkopen.2021.30272

Borges

Benjet

Medina-Mora

M. E.

Orozco

Nock

(2008). Suicide ideation, plan, and attempt in the Mexican adolescent mental health survey. Journal of the American Academy of Child and Adolescent Psychiatry, 47(1), 41–52. https://doi.org/10.1097/chi.0b013e31815896ad

Bredemeier

Miller

I. W.

(2015). Executive function and suicidality: A systematic qualitative review. Clinical Psychology Review, 40, 170–183. https://doi.org/10.1016/j.cpr.2015.06.005

Brent

D. A.

Bridge

Johnson

B. A.

Connolly

(1996). Suicidal behavior runs in families: A controlled family study of adolescent suicide victims. Archives of General Psychiatry, 53(12), 1145–1152.

10.

Cassidy

Bradley

Robinson

Allison

McHugh

Baron-Cohen

(2014). Suicidal ideation and suicide plans or attempts in adults with Asperger’s syndrome attending a specialist diagnostic clinic: A clinical cohort study. The Lancet Psychiatry, 1(2), 142–147. http://linkinghub.elsevier.com/retrieve/pii/S2215036614702482

11.

Cassidy

Rodgers

(2017). Understanding and prevention of suicide in autism. The Lancet Psychiatry, 4(6), Article e11. https://doi.org/10.1016/S2215-0366(17)30162-1

12.

Cassidy

S. A.

Gould

Townsend

Pelton

Robertson

A. E.

Rodgers

(2020). Is camouflaging autistic traits associated with suicidal thoughts and behaviours? Expanding the interpersonal psychological theory of suicide in an undergraduate student sample. Journal of Autism and Developmental Disorders, 50(10), 3638–3648. https://doi.org/10.1007/s10803-019-04323-3

13.

Cha

C. B.

Franz

P. J. E. M. G.

Glenn

C. R.

Kleiman

E. M.

Nock

M. K.

(2018). Annual Research Review: Suicide among youth—Epidemiology, (potential) etiology, and treatment. Journal of Child Psychology and Psychiatry and Allied Disciplines, 59(4), 460–482. https://doi.org/10.1111/jcpp.12831

14.

Chang

J.-C.

Huang

W.-L.

Chen

Y.-L.

Gau

S.-F.

(2020). The mental health of immigrant mother’s offspring in Taiwan: A national epidemiological study. Journal of the Formosan Medical Association, 119(2), 601–609. https://doi.org/10.1016/j.jfma.2019.08.033

15.

Chang

J.-C.

Lai

M.-C.

Tai

Y.-M.

Gau

S. S.-F.

(2022). Mental health correlates and potential childhood predictors for the wish to be of the opposite sex in young autistic adults. Autism, 26, 146–159.

16.

Chen

M.-H.

Pan

T. L.

Lan

W. H.

Hsu

J. W.

Huang

K. L.

T. P.

. . .Bai

Y. M.

(2017). Risk of suicide attempts among adolescents and young adults with autism spectrum disorder: A nationwide longitudinal follow-up study. Journal of Clinical Psychiatry, 78(9), e1174–e1179. https://doi.org/10.4088/JCP.16m11100

17.

Chen

Y. Y.

Chen

Y. L.

Gau

S. S.

(2020). Suicidality in children with elevated autistic traits. Autism Research, 13(10), 1811–1821. https://doi.org/10.1002/aur.2333

18.

Chesney

Goodwin

G. M.

Fazel

(2014). Risks of all-cause and suicide mortality in mental disorders: A meta-review. World Psychiatry, 13(2), 153–160.

19.

Chiang

H. L.

Gau

S. S.

(2016). Comorbid psychiatric conditions as mediators to predict later social adjustment in youths with autism spectrum disorder. Journal of Child Psychology and Psychiatry, 57(1), 103–111. https://doi.org/10.1111/jcpp.12450

20.

Chiang

H. L.

Kao

W. C.

Chou

M. C.

Chou

W. J.

Chiu

Y. N.

Y. Y.

Gau

S. S.

(2018). School dysfunction in youth with autistic spectrum disorder in Taiwan: The effect of subtype and ADHD. Autism Research, 11(6), 857–869. https://doi.org/10.1002/aur.1923

21.

Clemence

A. J.

Lewis

(2018). Flexibility and rigidity in object relational functioning: Assessing change in suicidal ideation and global psychiatric functioning using the SCORS-G. Journal of Personality Assessment, 100(2), 135–144. https://doi.org/10.1080/00223891.2017.1418747

22.

Commisso

Temcheff

Orri

Poirier

Lau

Côté

. . .Geoffroy

M.-C.

(2023). Childhood externalizing, internalizing and comorbid problems: Distinguishing young adults who think about suicide from those who attempt suicide. Psychological Medicine, 53, 1030–1037. https://doi.org/10.1017/S0033291721002464

23.

Cuesta

Montesó-Curto

Metzler Sawin

Jiménez-Herrera

Puig-Llobet

Seabra

Toussaint

(2021). Risk factors for teen suicide and bullying: An international integrative review. International Journal of Nursing Practice, 27(3), Article e12930. https://doi.org/10.1111/ijn.12930

24.

Dell’Osso

Carpita

Muti

Morelli

Salarpi

Salerni

. . .Maj

(2019). Mood symptoms and suicidality across the autism spectrum. Comprehensive Psychiatry, 91, 34–38. https://doi.org/10.1016/j.comppsych.2019.03.004

25.

Demetriou

E. A.

Lampit

Quintana

D. S.

Naismith

S. L.

Song

Y. J. C.

Pye

J. E.

. . .Guastella

A. J.

(2018). Autism spectrum disorders: A meta-analysis of executive function. Molecular Psychiatry, 23(5), 1198–1204. https://doi.org/10.1038/mp.2017.75

26.

Fazel

Runeson

(2020). Suicide. New England Journal of Medicine, 382(3), 266–274. https://doi.org/10.1056/NEJMra1902944

27.

Fernández-Sevillano

González-Pinto

Rodríguez-Revuelta

Alberich

Gónzalez-Blanco

Zorrilla

. . .Sáiz

P. A.

(2021). Suicidal behaviour and cognition: A systematic review with special focus on prefrontal deficits. Journal of Affective Disorders, 278, 488–496. https://doi.org/10.1016/j.jad.2020.09.044

28.

Fink

Olthof

Goossens

van der Meijden

Begeer

(2018). Bullying-related behaviour in adolescents with autism: Links with autism severity and emotional and behavioural problems. Autism, 22(6), 684–692.

29.

Gau

S. S.

Chong

M. Y.

Chen

T. H.

Cheng

A. T.

(2005). A 3-year panel study of mental disorders among adolescents in Taiwan. American Journal of Psychiatry, 162(7), 1344–1350. https://doi.org/10.1176/appi.ajp.162.7.1344

30.

Gau

S. S.

Lee

C. M.

Lai

M. C.

Chiu

Y. N.

Huang

Y. F.

Kao

J. D.

Y. Y.

(2011). Psychometric properties of the Chinese version of the social communication questionnaire. Research in Autism Spectrum Disorders, 5(2), 809–818. https://doi.org/10.1016/j.rasd.2010.09.010

31.

Gau

S. S.

Shen

H. Y.

Soong

W. T.

Gau

C. S.

(2006). An open-label, randomized, active-controlled equivalent trial of osmotic release oral system methylphenidate in children with attention-deficit/hyperactivity disorder in Taiwan. Journal of Child and Adolescent Psychopharmacology, 16(4), 441–455. https://doi.org/10.1089/cap.2006.16.441

32.

Geoffroy

M. C.

Boivin

Arseneault

Turecki

Vitaro

Brendgen

. . .Côté

S. M.

(2016). Associations between peer victimization and suicidal ideation and suicide attempt during adolescence: Results from a Prospective Population-Based Birth Cohort. Journal of the American Academy of Child and Adolescent Psychiatry, 55(2), 99–105. https://doi.org/10.1016/j.jaac.2015.11.010

33.

Gili

Castellví

Vives

de la Torre-Luque

Almenara

Blasco

M. J.

. . .Roca

(2019). Mental disorders as risk factors for suicidal behavior in young people: A meta-analysis and systematic review of longitudinal studies. Journal of Affective Disorders, 245, 152–162. https://doi.org/10.1016/j.jad.2018.10.115

34.

Hand

B. N.

Benevides

T. W.

Carretta

H. J.

(2020). Suicidal ideation and self-inflicted injury in Medicare enrolled autistic adults with and without co-occurring intellectual disability. Journal of Autism and Developmental Disorders, 50(10), 3489–3495. https://doi.org/10.1007/s10803-019-04345-x

35.

Hayes

A. F.

(2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford.

36.

Hedley

Uljarević

(2018). Systematic review of suicide in autism spectrum disorder: Current trends and implications. Current Developmental Disorders Reports, 5(1), 65–76. https://doi.org/10.1007/s40474-018-0133-6

37.

Hedley

Uljarevic

Cai

R. Y.

Bury

S. M.

Stokes

M. A.

Evans

D. W.

(2021). Domains of the autism phenotype, cognitive control, and rumination as transdiagnostic predictors of DSM-5 suicide risk. PLOS ONE, 16(1), Article e0245562. https://doi.org/10.1371/journal.pone.0245562

38.

Hedley

Uljarevic

Foley

K. R.

Richdale

Trollor

(2018). Risk and protective factors underlying depression and suicidal ideation in autism spectrum disorder. Depress Anxiety, 35(7), 648–657. https://doi.org/10.1002/da.22759

39.

Herba

C. M.

Ferdinand

R. F.

Verhulst

F. C.

(2007). Long-term associations of childhood suicide ideation. Journal of the American Academy of Child and Adolescent Psychiatry, 46(11), 1473–1481.

40.

Hirvikoski

Boman

Chen

D’Onofrio

B. M.

Mittendorfer-Rutz

Lichtenstein

. . .Larsson

(2020). Individual risk and familial liability for suicide attempt and suicide in autism: A population-based study. Psychological Medicine, 50(9), 1463–1474. https://doi.org/10.1017/S0033291719001405

41.

Hirvikoski

Mittendorfer-Rutz

Boman

Larsson

Lichtenstein

Bölte

(2016). Premature mortality in autism spectrum disorder. British Journal of Psychiatry, 208(3), 232–238. https://doi.org/10.1192/bjp.bp.114.160192

42.

Holden

Mueller

McGowan

Sanyal

Kikoler

Simonoff

. . .Downs

(2020). Investigating bullying as a predictor of suicidality in a clinical sample of adolescents with autism spectrum disorder. Autism Research, 13(6), 988–997. https://doi.org/10.1002/aur.2292

43.

Horowitz

L. M.

Thurm

Farmer

Mazefsky

Lanzillo

Bridge

J. A.

, . . . Autism and Developmental Disorders Inpatient Research Collaborative. (2018). Talking about death or suicide: prevalence and clinical correlates in youth with autism spectrum disorder in the psychiatric inpatient setting. Journal of Autism and Developmental Disorders, 48(11), 3702–3710. https://doi.org/10.1007/s10803-017-3180-7

44.

Howe

S. J.

Hewitt

Baraskewich

Cassidy

McMorris

C. A.

(2020). Suicidality among children and youth with and without autism spectrum disorder: A systematic review of existing risk assessment tools. Journal of Autism and Developmental Disorders, 50(10), 3462–3476. https://doi.org/10.1007/s10803-020-04394-7

45.

Hsiao

M.-N.

Tai

Y.-M.

Y.-Y.

Tsai

W.-C.

Chiu

Y.-N.

Gau

S. S.-F.

(2022). Psychopathologies mediate the link between autism spectrum disorder and bullying involvement: A follow-up study. Journal of the Formosan Medical Association, 121, 1739–1747.

46.

Hunsche

M. C.

Saqui

Mirenda

Zaidman-Zait

Bennett

Duku

. . .Kerns

C. M.

(2020). Parent-reported rates and clinical correlates of suicidality in children with autism spectrum disorder: A longitudinal study. Journal of Autism and Developmental Disorders, 50(10), 3496–3509. https://doi.org/10.1007/s10803-020-04373-y

47.

John

Gammon

G. D.

Prusoff

B. A.

Warner

(1987). The Social Adjustment Inventory for Children and Adolescents (SAICA): Testing of a new semistructured interview. Journal of the American Academy of Child and Adolescent Psychiatry, 26(6), 898–911. https://doi.org/10.1097/00004583-198726060-00015

48.

Jokiranta-Olkoniemi

Gyllenberg

Sucksdorff

Suominen

Kronstrom

Chudal

Sourander

(2021). Risk for premature mortality and intentional self-harm in autism spectrum disorders. Journal of Autism and Developmental Disorders, 51, 3098–3108. https://doi.org/10.1007/s10803-020-04768-x

49.

Keilp

Gorlyn

Russell

Oquendo

Burke

Harkavy-Friedman

Mann

(2013). Neuropsychological function and suicidal behavior: Attention control, memory and executive dysfunction in suicide attempt. Psychological Medicine, 43(3), 539–551.

50.

Kõlves

Fitzgerald

Nordentoft

Wood

S. J.

Erlangsen

(2021). Assessment of suicidal behaviors among individuals with autism spectrum disorder in Denmark. JAMA Network Open, 4(1), Article e2033565. https://doi.org/10.1001/jamanetworkopen.2020.33565

51.

Kraemer

H. C.

Lowe

K. K.

Kupfer

D. J.

(2005). To your health: How to understand what research tells us about risk. Oxford University Press.

52.

Lai

M. C.

Kassee

Besney

Bonato

Hull

Mandy

. . .Ameis

S. H.

(2019). Prevalence of co-occurring mental health diagnoses in the autism population: A systematic review and meta-analysis. The Lancet Psychiatry, 6(10), 819–829. https://doi.org/10.1016/s2215-0366(19)30289-5

53.

Lee

Fung

S. C.

Tsang

Liu

Z. R.

Huang

Y. Q.

Y. L.

. . .Kessler

R. C.

(2007). Lifetime prevalence of suicide ideation, plan, and attempt in metropolitan China. Acta Psychiatrica Scandinavica, 116(6), 429–437. https://doi.org/10.1111/j.1600-0447.2007.01064.x

54.

Lengvenyte

Conejero

Courtet

Olie

. (2021). Biological bases of suicidal behaviours: A narrative review. European Journal of Neuroscience, 53(1), 330–351. https://doi.org/10.1111/ejn.14635

55.

Licence

Oliver

Moss

Richards

(2020). Prevalence and risk-markers of self-harm in autistic children and adults. Journal of Autism and Developmental Disorders, 50(10), 3561–3574. https://doi.org/10.1007/s10803-019-04260-1

56.

Lin

P. I.

Moni

M. A.

Gau

S. S.

Eapen

(2021). Identifying subgroups of patients with autism by gene expression profiles using machine learning algorithms. Frontiers in Psychiatry, 12, Article 637022. https://doi.org/10.3389/fpsyt.2021.637022

57.

Lin

Y. J.

Tseng

W. L.

Gau

S. S.

(2021). Psychiatric comorbidity and social adjustment difficulties in children with disruptive mood dysregulation disorder: A national epidemiological study. Journal of Affective Disorders, 281, 485–492. https://doi.org/10.1016/j.jad.2020.12.039

58.

Y. C.

Chen

Y. J.

Hsu

Y. C.

Tseng

W. I.

Gau

S. S.

(2017). Reduced tract integrity of the model for social communication is a neural substrate of social communication deficits in autism spectrum disorder. Journal of Child Psychology and Psychiatry, 58(5), 576–585. https://doi.org/10.1111/jcpp.12641

59.

Luciana

Nelson

C. A.

(1998). The functional emergence of prefrontally-guided working memory systems in four- to eight-year-old children. Neuropsychologia, 36(3), 273–293.

60.

MacKinnon

D. P.

Fairchild

A. J.

Fritz

M. S.

(2007). Mediation analysis. Annual Review of Psychology, 58, 593–614.

61.

MacPherson

H. A.

Kim

K. L.

Seymour

K. E.

Wolff

Esposito-Smythers

Spirito

Dickstein

D. P.

(2022). Cognitive flexibility and impulsivity deficits in suicidal adolescents. Research on Child and Adolescent Psychopathology, 50, 1643–1656. https://doi.org/10.1007/s10802-022-00952-y

62.

MacPherson

H. A.

Kudinova

A. Y.

Schettini

Jenkins

G. A.

Gilbert

A. C.

Thomas

S. A.

. . .Dickstein

D. P.

(2022). Relationship between cognitive flexibility and subsequent course of mood symptoms and suicidal ideation in young adults with childhood-onset bipolar disorder. European Child & Adolescent Psychiatry, 31(2), 299–312. Advance online publication. https://doi.org/10.1007/s00787-020-01688-0

63.

Maiano

Normand

C. L.

Salvas

M. C.

Moullec

Aimé

(2016). Prevalence of school bullying among youth with autism spectrum disorders: A systematic review and meta-analysis. Autism Research, 9(6), 601–615.

64.

Mann

J. J.

Rizk

M. M.

(2020). A brain-centric model of suicidal behavior. American Journal of Psychiatry, 177(10), 902–916.

65.

Martini

M. I.

Kuja-Halkola

Butwicka

Du Rietz

D’Onofrio

B. M.

Happé

. . .Taylor

M. J.

(2022). Sex differences in mental health problems and psychiatric hospitalization in autistic young adults. JAMA Psychiatry, 79(12), 1188–1198. https://doi.org/10.1001/jamapsychiatry.2022.3475

66.

Mayes

S. D.

Gorman

A. A.

Hillwig-Garcia

Syed

(2013). Suicide ideation and attempts in children with autism. Research in Autism Spectrum Disorders, 7(1), 109–119. https://doi.org/10.1016/j.rasd.2012.07.009

67.

Mournet

A. M.

Wilkinson

Bal

V. H.

Kleiman

E. M.

(2023). A systematic review of predictors of suicidal thoughts and behaviors among autistic adults: Making the case for the role of social connection as a protective factor. Clinical Psychology Review, 99, Article 102235. https://doi.org/10.1016/j.cpr.2022.102235

68.

Nock

M. K.

Hwang

Sampson

N. A.

Kessler

R. C.

(2010). Mental disorders, comorbidity and suicidal behavior: Results from the National Comorbidity Survey Replication. Molecular Psychiatry, 15(8), 868–876.

69.

Novak

L. A.

Carter

S. P.

LaCroix

J. M.

Perera

K. U.

Neely

L. L.

Soumoff

Ghahramanlou-Holloway

(2022). Cognitive flexibility and suicide risk indicators among psychiatric inpatients. Psychiatry Research, 313, Article 114594. https://doi.org/10.1016/j.psychres.2022.114594

70.

O’Connor

R. C.

Nock

M. K.

(2014). The psychology of suicidal behaviour. The Lancet Psychiatry, 1(1), 73–85. https://doi.org/10.1016/s2215-0366(14)70222-6

71.

O’Halloran

Coey

Wilson

(2022). Suicidality in autistic youth: A systematic review and meta-analysis. Clinical Psychology Review, 93, Article 102144. https://doi.org/10.1016/j.cpr.2022.102144

72.

Olton

D. S.

(1987). The radial arm maze as a tool in behavioral pharmacology. Physiology & Behavior, 40(6), 793–797.

73.

Park

Gong

Lyons

G. L.

Hirota

Takahashi

Kim

. . .Leventhal

B. L.

(2020). Prevalence of and factors associated with school bullying in students with autism spectrum disorder: A cross-cultural meta-analysis. Yonsei Medical Journal, 61(11), 909–922. https://doi.org/10.3349/ymj.2020.61.11.909

74.

Petrides

Milner

(1982). Deficits on subject-ordered tasks after frontal-and temporal-lobe lesions in man. Neuropsychologia, 20(3), 249–262.

75.

Preacher

K. J.

Hayes

A. F.

(2008). Assessing mediation in communication research. In Hayes

A. F.

Slater

M. D.

Snyder

L. B.

(Eds.), The Sage sourcebook of advanced data analysis methods for communication research. Sage.

76.

Roos

Sareen

Bolton

J. M.

(2013). Suicide risk assessment tools, predictive validity findings and utility today: Time for a revamp. Neuropsychiatry, 3(5), 483–495.

77.

Saffer

B. Y.

Klonsky

E. D.

(2018). Do neurocognitive abilities distinguish suicide attempters from suicide ideators? A systematic review of an emerging research area. Clinical Psychology: Science and Practice, 25(1), Article e12227. https://doi.org/10.1111/cpsp.12227

78.

Sahgal

(1987). Some limitations of indices derived from signal detection theory: Evaluation of an alternative index for measuring bias in memory tasks. Psychopharmacology, 91, 517–520.

79.

Shang

C. Y.

Gau

S. S.

Soong

W. T.

(2006). Association between childhood sleep problems and perinatal factors, parental mental distress and behavioral problems. Journal of Sleep Research, 15(1), 63–73. https://doi.org/10.1111/j.1365-2869.2006.00492.x

80.

Shang

C. Y.

Shih

H. H.

Pan

Y. L.

Lin

H. Y.

Gau

S. S.

(2020). Comparative efficacy of methylphenidate and atomoxetine on social adjustment in youths with attention-deficit/hyperactivity disorder. Journal of Child and Adolescent Psychopharmacology, 30(3), 148–158. https://doi.org/10.1089/cap.2019.0139

81.

Sörberg Wallin

Zeebari

Lager

Gunnell

Allebeck

Falkstedt

. (2018). Suicide attempt predicted by academic performance and childhood IQ: A cohort study of 26,000 children. Acta Psychiatrica Scandinavica, 137(4), 277–286.

82.

Soto-Sanz

Castellví

Piqueras

J. A.

Rodríguez-Marín

Rodríguez-Jiménez

Miranda-Mendizábal

. . .Alonso

(2019). Internalizing and externalizing symptoms and suicidal behaviour in young people: A systematic review and meta-analysis of longitudinal studies. Acta Psychiatrica Scandinavica, 140(1), 5–19. https://doi.org/10.1111/acps.13036

83.

South

Beck

J. S.

Lundwall

Christensen

Cutrer

E. A.

Gabrielsen

T. P.

. . .Lundwall

R. A.

(2020). Unrelenting depression and suicidality in women with autistic traits. Journal of Autism and Developmental Disorders, 50(10), 3606–3619. https://doi.org/10.1007/s10803-019-04324-2

84.

Sterling

Dawson

Estes

Greenson

(2008). Characteristics associated with presence of depressive symptoms in adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 38(6), 1011–1018. https://doi.org/10.1007/s10803-007-0477-y

85.

Storch

E. A.

Sulkowski

M. L.

Nadeau

Lewin

A. B.

Arnold

E. B.

Mutch

P. J.

. . .Murphy

T. K.

(2013). The phenomenology and clinical correlates of suicidal thoughts and behaviors in youth with autism spectrum disorders. Journal of Autism and Developmental Disorders, 43(10), 2450–2459. https://doi.org/10.1007/s10803-013-1795-x

86.

Tan

Frewer

Cox

Williams

Ure

(2021). Prevalence and age of onset of regression in children with autism spectrum disorder: A systematic review and meta-analytical update. Autism Research, 14(3), 582–598. https://doi.org/10.1002/aur.2463

87.

Tsai

S. J.

Chang

W. H.

Cheng

C. M.

Liang

C. S.

Bai

Y. M.

Hsu

J. W.

. . .Chen

M. H.

(2023). All-cause mortality and suicide mortality in autistic individuals: An entire population longitudinal study in Taiwan. Autism, 27, 2496–2506. https://doi.org/10.1177/13623613231167287

88.

Turecki

Brent

D. A.

Gunnell

O’Connor

R. C.

Oquendo

M. A.

Pirkis

Stanley

B. H.

(2019). Suicide and suicide risk. Nature Reviews Disease Primers, 5(1), Article 74. https://doi.org/10.1038/s41572-019-0121-0

89.

Velikonja

Fett

A.-K.

Velthorst

(2019). Patterns of nonsocial and social cognitive functioning in adults with autism spectrum disorder: A systematic review and meta-analysis. JAMA Psychiatry, 76(2), 135–151.

90.

Widnall

Epstein

Polling

Velupillai

Jewell

Dutta

. . .Downs

(2022). Autism spectrum disorders as a risk factor for adolescent self-harm: A retrospective cohort study of 113,286 young people in the UK. BMC Medicine, 20(1), Article 137. https://doi.org/10.1186/s12916-022-02329-w

91.

Winsper

Lereya

Zanarini

Wolke

(2012). Involvement in bullying and suicide-related behavior at 11 years: A prospective birth cohort study. Journal of the American Academy of Child and Adolescent Psychiatry, 51(3), 271–282, e273. https://doi.org/10.1016/j.jaac.2012.01.001

92.

Xie

Sun

Yang

Guo

(2020). Characteristic executive dysfunction for high-functioning autism sustained to adulthood. Autism Research, 13(12), 2102–2121. https://doi.org/10.1002/aur.2304

93.

Zahid

Upthegrove

(2017). Suicidality in autistic spectrum disorders. Crisis, 38(4), 237–246. https://doi.org/10.1027/0227-5910/a000458

94.

Zhao

Lynch

J. G.

Jr. Chen

(2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206.

95.

Zimmerman

Ownsworth

O’Donovan

Roberts

Gullo

M. J.

(2017). Associations between executive functions and mental health outcomes for adults with autism spectrum disorder. Psychiatry Research, 253, 360–363. https://doi.org/10.1016/j.psychres.2017.04.023

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB